# Research Papers — 2026-08 · bundle page 234 of 240 > Full-text markdown of papers #11651–11700 of 11995 published in 2026-08, newest first. AI-readable mirrors of the full 11995-paper 2026-08 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-08/page/234 [← Previous bundle](https://aipod.dev/papers/archive/2026-08/page/233.md) [Next bundle →](https://aipod.dev/papers/archive/2026-08/page/235.md) ### Inter-Residue Geometry Attention for Antibody-Specific Epitope Prediction Authors: Chuanliu Fan, Nan Yu, Junjie Wu, Guohong Fu Published: 2026-08-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.01092v1 Antibody-specific epitope prediction aims to identify which antigen residues are recognized by a given antibody, a task that depends on the three-dimensional complementarity between antibody CDRs and the antigen surface. Existing methods usually leverage PLM embeddings and inject structure through additional graph, surface, or point-cloud encoders, where the positional mechanism inside attention remains largely tied to one-dimensional sequence order. For proteins, the analogue of a token offset is not only sequence separation, but also the three-dimensional displacement between residues after folding. This raises a question, can folded residue geometry serve as the positional mechanism of attention itself? We propose Local-Frame 3D Rotary Position Encoding (LF3DRoPE), which expresses inter-residue displacements in backbone-defined local frames and injects them directly into rotary attention. This design preserves continuous directional geometry while ensuring invariance to global $\mathrm{SE}(3)$ transformations. On the AsEP benchmark, LF3DRoPE achieves state-of-the-art $\mathrm{MCC}$ on both ratio and epitope-group splits. Ablations and rigid transformation tests show that local three-dimensional geometry provides information beyond sequence-order attention while preserving invariance to arbitrary global coordinate systems. Mutation ranking results further indicate that LF3DRoPE captures antigen-specific structural compatibility. --- ### Factorized AdaBoost.MH Achieves the Same Convergence Rate as AdaBoost.MH Authors: Xin Zou, Jingyuan Xu Published: 2026-08-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.01091v2 {AdaBoost.MH} reduces multi-class classification to a collection of binary subproblems and enjoys the classical boosting-type convergence guarantee under a weak learning condition. A more structured variant, Factorized {AdaBoost.MH}, uses base classifiers of the form $\mathbf{h}(x)=α\mathbf{v} \bm{\varphi}(x)$, where a single binary classifier $\bm{\varphi}$ is shared across all classes and the label dependence is carried by a vote vector $\mathbf{v} \in\{\pm1\}^K$. This factorization is algorithmically attractive and achieves better performance in practice, but its convergence depends on whether one can always choose a vote vector with sufficiently large induced binary weight mass. Previous work resolved this question with a lower bound $\max\{1/n,1/\sqrt{2K}\}$, which still leaves a dimension-dependent slowdown relative to the original {AdaBoost.MH} analysis. In this paper, we sharpen this combinatorial step. For the minimax quantity $\mathfrak{W}_{n,K}$ governing the factorized edge, we prove $\mathfrak{W}_{n,K} = C_{\min\{n+1,K\}}$, where $C_q=1$ for $q=1$, $C_q=q/(3q-4)$ for even $q\ge2$, and $C_q=(q+1)/(3q-1)$ for odd $q\ge2$. Since $C_q\downarrow 1/3$, our bounds show that $\mathfrak{W}_{n,K}=Θ(1)$ uniformly over $n$ and $K$. Consequently, Factorized {AdaBoost.MH} achieves the same boosting-type convergence rate as {AdaBoost.MH} up to a universal constant factor, removing the previously suggested additional dependence on $n$ or $K$ in the number of boosting rounds. --- ### Factorized AdaBoost.MH Achieves the Same Convergence Rate as AdaBoost.MH Authors: Xin Zou, Jingyuan Xu Published: 2026-08-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.01091v1 AdaBoost.MH reduces multi-class classification to a collection of binary subproblems and enjoys the classical boosting-type convergence guarantee under a weak learning condition. A more structured variant, Factorized AdaBoost.MH, uses base classifiers of the form $\mathbf{h}(x)=α\mathbf{v} \bm{\varphi}(x)$, where a single binary classifier $\bm{\varphi}$ is shared across all classes and the label dependence is carried by a vote vector $\mathbf{v} \in\{\pm1\}^K$. This factorization is algorithmically attractive and achieves better performance in practice, but its convergence depends on whether one can always choose a vote vector with sufficiently large induced binary weight mass. Previous work resolved this question with a lower bound $\max\{1/n,1/\sqrt{2K}\}$, which still leaves a dimension-dependent slowdown relative to the original AdaBoost.MH analysis. In this paper, we sharpen this combinatorial step. For the minimax quantity $\mathfrak{W}_{n,K}$ governing the factorized edge, we prove $\max\{1/n,C_K\}\le\mathfrak{W}_{n,K}\le C_{\min\{n,K\}}$, where $C_q=1$ for $q=1$, $C_q=q/(3q-4)$ for even $q\ge2$, and $C_q=(q+1)/(3q-1)$ for odd $q\ge2$. Since $C_q\downarrow 1/3$, our bounds show that $\mathfrak{W}_{n,K}=Θ(1)$ uniformly over $n$ and $K$. Consequently, Factorized AdaBoost.MH achieves the same boosting-type convergence rate as AdaBoost.MH up to a universal constant factor, removing the previously suggested additional dependence on $n$ or $K$ in the number of boosting rounds. --- ### Breaking Diversity Collapse in Spiking Pseudo-Ensembles for Efficient OOD Detection in Remote Sensing Authors: Srinivas Anumasa, Rushi Shah, Qiran Zou, Dianbo Liu Published: 2026-08-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.01090v1 Spiking Neural Networks (SNNs) are attractive for resource-constrained remote-sensing systems, but reliable out-of-distribution (OOD) detection remains challenging. Deep ensembles provide strong predictive uncertainty, yet require multiple complete models and backbone evaluations. We propose an efficient spiking pseudo-ensemble that attaches multiple lightweight classification heads to a frozen SNN backbone. Naively training these heads with cross-entropy can lead to diversity collapse, where independently parameterized heads may produce correlated predictions. To address this, we introduce an agree--disagree objective that preserves correct predictions on clean in-distribution samples while encouraging diversity on structured, uncertainty-inducing transformations of the same inputs. This provides a diversity-promoting training signal without requiring external OOD data. Experiments with Spikformer and ResNet19-SNN on EuroSAT demonstrate consistent improvements over conventionally trained pseudo-ensembles. Using three backbones with five heads each matches or improves upon a five-model deep ensemble on UCM and AID, while requiring approximately 38% fewer parameters and 40% fewer backbone evaluations. These results show that explicit diversity promotion can recover useful ensemble-style uncertainty at substantially lower deployment cost. --- ### When Collaboration Becomes a Trigger: Collective Evidence-Threshold Backdoors in Multi-Agent Systems Authors: Jia-Hao Xiao, Lei Feng, Min-Ling Zhang Published: 2026-08-02 Categories: cs.MA, cs.LG arXiv: https://arxiv.org/abs/2608.01085v1 LLM-based multi-agent systems (MAS) extend LLM capabilities through iterative communication and shared contexts. However, this collaboration introduces a vulnerability: backdoor behavior can be activated when peer evidence reaches a hidden threshold, rather than being determined by any single message. We introduce a collective evidence-threshold backdoor paradigm for MAS and Boundary-Conditioned Backdoor Injection (BCBI), which constructs counterfactual boundary pairs to separate benign behavior before the threshold from the adversarial objective after it, and learns latent progression aligned with evidence. To mitigate this threat, we propose LAtent Transition Test-time Evaluation (LATTE), a clean-only latent-transition defense that learns benign communication dynamics and quarantines anomalous agent updates before their responses propagate. Across several benchmarks, BCBI yields selective activation with little premature activation; without knowing the attack target or trigger, LATTE limits propagation with minimal disruption. --- ### Attend to Your Own Thoughts: Breaking the Barrier for Post-Training Quantization of Reasoning LLMs through the Lens of 1.58-Bit Quantization Authors: Shigeng Wang, Chao Li, Yangyuxuan Kang, Jiawei Fan, Anbang Yao Published: 2026-08-02 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.01078v1 We propose ScaleQ-1.58, a scalable ternary post-training quantization (PTQ) framework for reasoning LLMs. Its core insight stems from an empirical finding: although modern LLMs are typically trained to exhibit chain-of-thought reasoning capabilities, in the PTQ regime, even the latest CAT-Q method based on learning-based differentiable ternarization still leads to performance collapse on challenging mathematics and coding tasks when using conventional calibration schemes that ignore the model's reasoning process. Driven by this finding, we introduce a simple calibration approach, Attend to Your Own Thoughts (AYOT), where reasoning traces and final answers generated by the pre-trained high-precision target LLM on a proper set of calibration samples are used as the context input during the ternarization process, along with the corresponding questions. ScaleQ-1.58 is formed by simply integrating AYOT with CAT-Q, which demonstrates several scaling properties: (1) with only 4M calibration tokens, Qwen3-1.7B ternarized by ScaleQ-1.58 reaches over 90.52% of the performance of the prior best BitNet b1.58 2B4T averaged over 4 mathematics and coding tasks, and our ternary Qwen3-4B shows an absolute gain of 8.97%, while requiring 1,000,000x fewer calibration tokens for quantization; (2) ScaleQ-1.58 generalizes well to both dense and MoE architectures, with performance improving as model scale increases (up to 235B parameters); (3) ScaleQ-1.58 demonstrates strong generalization across tasks of varying difficulty levels, including mathematics, coding and scientific logic reasoning, as well as commonsense reasoning and basic language generation; (4) its performance continues to improve as the number of calibration tokens increases. Notably, AYOT also exhibits strong generalization ability across other quantization bit-widths. Code will be available at https://github.com/IntelChina-AI/BitTern. --- ### Role-Decoupled Attention Residuals: Separating Matching and Content Retrieval Across Depth Authors: Kehan Wang Published: 2026-08-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.01075v1 Depth-routing residual architectures allow Transformer layers to retrieve earlier representations instead of inheriting only the immediately preceding state. Existing Block Attention Residuals, however, use a single content-dependent depth mixture to construct the inputs to queries, keys, and values. This design couples two functionally different decisions: queries and keys determine where attention matches, whereas values determine what content is retrieved. We therefore ask whether matching and content retrieval should be forced to read from the same depth. We introduce Role-Decoupled Attention Residuals (RD-AttnRes), a minimal extension that shares one depth route between queries and keys while learning an independent value route over the same residual sources. Tying the two routing queries exactly recovers the parent architecture, while decoupling them adds only one model-width vector per layer and introduces no additional token-to-token attention operation. We evaluate RD-AttnRes using a frozen, paired pretraining protocol on FineWeb-Edu with five matched seeds for both 120M- and 343M-parameter models and a 2.0B-token training budget. RD-AttnRes improves validation negative log-likelihood in all 10 matched comparisons. The mean reductions are 0.0301 and 0.0247, corresponding to perplexity reductions of 2.97 percent and 2.43 percent at 120M and 343M parameters, respectively. Early-budget controls indicate that neither the additional parameter count, duplicated routing execution, nor a fixed value route reproduces the improvement. Routing diagnostics further reveal persistent divergence between the query-key and value depth distributions. These results suggest that, within the evaluated training regime, attention matching and content retrieval benefit from distinct reads over the residual hierarchy. --- ### Logit-Origin Centering for Singleton Test-Time Adaptation Authors: Mayank Sharma, Rohit Kumar Mourya, Pratik Mazumder Published: 2026-08-02 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2608.01074v1 Tabular data is used extensively in many real-world use cases. Deep learning models have been developed to deal with tabular data, but generally perform poorly when the test data distribution differs from that of the training data. Researchers have proposed test-time adaptation approaches to deal with this problem. The fully test-time adaptation (FTTA) setting involves adapting deployed classifiers to shifted target distributions using only unlabeled test data. Leading FTTA methods inherit a batch-dependent approach from computer vision literature. This paper demonstrates for the first time that such approaches degrade sharply in strict streaming regimes where examples arrive and must be classified one at a time. This occurs because at a batch size of one, batch-level statistics become unavailable or poorly estimated. We argue that singleton tabular FTTA is not merely a small-batch variant of ordinary FTTA, but a distinct identifiability problem where only the location of the model's score stream remains directly observable. To address this, we propose Prequential Logit-Origin Centering (PLOC), a lightweight approach that keeps the source model frozen and shifts the logit space at each step. PLOC stores only a single running number (the mean of past logits), requires no labels, estimates no priors, and bypasses weight updates entirely. A deferred variant applies a static shift that preserves the source ranking, and thus the AUROC, exactly. Evaluated across five tabular benchmarks, three architectures (MLP, FT-Transformer, and TabTransformer), and five independent source checkpoints, PLOC significantly outperforms strong tabular and entropy-based baselines. --- ### PlantRig - From Bones to Branches: Adaptation of Autoregressive Rigging Models for Plant Skeletal Reconstruction Authors: Nathan Hu, Yang Yang, Fumio Okura Published: 2026-08-02 Categories: cs.CV, cs.GR arXiv: https://arxiv.org/abs/2608.01072v1 Autoregressive rigging models such as UniRig and SkinTokens perform well on articulated characters, but their ability to generalize to plant structures remains largely unexplored, since plant topologies exhibit highly variable, non-canonical branching patterns that challenge learned skeletal priors. We evaluate these models for plant skeletal reconstruction using synthetic L-system-generated trees and real scanned data spanning monopodial, sympodial, whorled, and vine-like archetypes. Preliminary testing showed UniRig collapsing complex branching into near-linear chains, while SkinTokens preserved topology better but over-segmented branches and produced an unstable output space, so we focused on UniRig for its greater stability. Diagnosis traced the collapse to sampling-level suppression of branch tokens, and further analysis showed the frozen mesh encoder had limited sensitivity to structural variation, pointing to a geometric bottleneck in the tokenization pipeline rather than a purely learned bias. Building on these findings, we applied multi-round fine-tuning over multiple procedurally generated synthetic datasets. Across rounds, the model progressively recovered accurate branching topology and generalized beyond branch-only structures to plants with foliage, a harder case given the zero-thickness, mesh-normal-dependent geometry of leaves. The resulting model generalized well across diverse plant forms without leaf-specific architectural changes, indicating that targeted fine-tuning can substantially close the domain gap between character-rigging priors and plant skeletal structure. As such, our work points toward a viable path for automated plant rigging across both branch topology and foliage type, even those not considered in our findings. --- ### Sharp Characterization of Bias in Post-Bandit Inference Authors: Lisu Wang, Yilun Chen, Jiaqi Lu Published: 2026-08-02 Categories: cs.LG, math.PR, stat.ML arXiv: https://arxiv.org/abs/2608.01069v2 Bandit algorithms generate data for downstream inference, but adaptive sampling biases post-bandit sample means. We analyze this bias for stable index algorithms, including UCB1 and its generalizations, and derive sharp leading-order expressions for the sample-mean bias and expected $Z$-statistic, in bandit experiments of fixed horizon $T$. Our characterization reveals the algorithmic origin of bias through a key index-function-dependent quantity, which we term effective exploration rate. For example, under UCB1, the effective exploration rate is of order $\sqrt{\log T}$, and the standardized bias of any arm (that is not uniquely optimal) decays at the extremely slow rate $1/\sqrt{\log T}$. We also show how the choice of the index function affects both regret and bias, which reveals a regret-bias trade-off: more exploratory algorithm reduces bias but increases regret. We further show how bias most severely distorts confidence intervals and hypothesis tests when the tested arm is one of the tied-optimal arms. Our sharp characterization for bias uses a novel empirical fluid approximation of the algorithm's sampling dynamics, which may be of independent interest. --- ### Characterizing Bias in Post-Bandit Inference under Index Algorithms Authors: Lisu Wang, Yilun Chen, Jiaqi Lu Published: 2026-08-02 Categories: cs.LG, math.PR, stat.ML arXiv: https://arxiv.org/abs/2608.01069v1 Bandit algorithms generate data for downstream inference, but adaptive sampling biases post-bandit sample means. We analyze this bias for stable index algorithms, including UCB1 and its generalizations, and derive sharp leading-order expressions for the sample-mean bias and expected $Z$-statistic. Our characterization reveals the algorithmic origin of bias through a key index-function-dependent quantity, which we term effective exploration rate. For example, under UCB1, the effective exploration rate is of order $\sqrt{\log T}$, and the standardized bias of any arm (that is not uniquely optimal) decays at the extremely slow rate $1/\sqrt{\log T}$. We also show how the choice of the index function affects both regret and bias, which reveals a regret-bias trade-off: more exploratory algorithm reduces bias but increases regret. Our sharp characterization for bias uses a novel empirical fluid approximation of the algorithm's sampling dynamics, which may be of independent interest. --- ### ReACT-CLIP: Response-Aware Test-Time Defense for Vision--Language Models Authors: Hashmat Shadab Malik, Toluwani Aremu, Samuele Poppi, Muzammal Naseer, Salman Khan Published: 2026-08-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.01067v2 Training-free test-time defenses offer a practical way to improve the adversarial robustness of CLIP-style vision--language models without modifying the pretrained model. However, their correction strength is typically fixed for a narrow range of attack budgets, even though the attack budget is unknown at inference and the required correction varies across samples. We show that this mismatch causes existing defenses to degrade sharply as attacks strengthen. We introduce ReACT-CLIP, a response-conditioned test-time defense that separately determines how strongly each input should be corrected and whether defensive intervention is necessary. Our key observation is that the relative increase in CLIP visual-feature drift between low- and high-noise probes provides a graded, sample-specific proxy for correction demand. ReACT-CLIP maps this relative cross-noise drift to the Gaussian noise scale used to construct a stable, noise-averaged feature anchor, enabling the corrective reach to adapt to each input. To determine whether intervention is necessary, we further observe that clean inputs retain stable class-probability distributions under weak spatial augmentations, whereas adversarial inputs exhibit greater variation. ReACT-CLIP quantifies this variation using a prediction-instability score computed by Jensen--Shannon divergence and combines it with relative cross-noise drift to form the defensive intervention score. ReACT-CLIP requires no model or prompt training, and its correction-strength mapping is calibrated once and fixed across datasets and attack budgets. Across 12 downstream datasets, as well as ImageNet and its distribution-shifted variants, ReACT-CLIP delivers substantial robustness gains across diverse attack types and strengths while largely preserving clean accuracy. --- ### ReACT-CLIP: Response-Aware Test-Time Defense for Vision--Language Models Authors: Hashmat Shadab Malik, Toluwani Aremu, Samuele Poppi, Muzammal Naseer, Salman Khan Published: 2026-08-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.01067v1 Training-free test-time defenses offer a practical way to improve the adversarial robustness of CLIP-style vision--language models without modifying the pretrained model. However, their correction strength is typically fixed for a narrow range of attack budgets, even though the attack budget is unknown at inference and the required correction varies across samples. We show that this mismatch causes existing defenses to degrade sharply as attacks strengthen. We introduce ReACT-CLIP, a response-conditioned test-time defense that separately determines how strongly each input should be corrected and whether defensive intervention is necessary. Our key observation is that the relative increase in CLIP visual-feature drift between low- and high-noise probes provides a graded, sample-specific proxy for correction demand. ReACT-CLIP maps this relative cross-noise drift to the Gaussian noise scale used to construct a stable, noise-averaged feature anchor, enabling the corrective reach to adapt to each input. To determine whether intervention is necessary, we further observe that clean inputs retain stable class-probability distributions under weak spatial augmentations, whereas adversarial inputs exhibit greater variation. ReACT-CLIP quantifies this variation using a prediction-instability score computed by Jensen--Shannon divergence and combines it with relative cross-noise drift to form the defensive intervention score. ReACT-CLIP requires no model or prompt training, and its correction-strength mapping is calibrated once and fixed across datasets and attack budgets. Across 12 downstream datasets, as well as ImageNet and its distribution-shifted variants, ReACT-CLIP delivers substantial robustness gains across diverse attack types and strengths while largely preserving clean accuracy. --- ### One Query, Many Scales: Sparse Mixture-of-Experts for Efficient Hierarchical Cross-View Geo-Localization Authors: Ruijie Fan, Junyan Ye, Qi Zhu, Weijia Li Published: 2026-08-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.01060v2 Cross-view geo-localization (CVGL) retrieves geo-tagged satellite imagery for a ground-view query. Most systems exhaustively search a flat, fixed-resolution gallery, incurring high cost over large areas and adapting poorly to satellite resolution changes. Autoregressive coarse-to-fine alternatives reduce comparisons but bind later predictions to earlier decisions and a predefined hierarchy. We introduce GeoMoE, a sparse mixture-of-experts dual encoder that decouples global multi-scale representation learning from local hierarchical search. Global multi-scale supervision and content-adaptive routing map ground and satellite images across resolutions into a globally comparable embedding space. At inference, each image is encoded once, and probabilistic beam search follows parent--child links to score a small candidate subset. Later levels reuse these descriptors rather than features generated by preceding levels, limiting feature-level error propagation and hierarchy coupling. We further introduce VIGOR-M, a four-city benchmark with an explicit parent--child satellite hierarchy and held-out half-step galleries for single-resolution, cross-resolution, and hierarchical evaluation. GeoMoE achieves 95.78% R@40m on Just Zoom In, 2.77 percentage points above the previous best, and 62.39% R@1 on VIGOR-M. The latter requires 0.885 MMAC/query for descriptor matching, 5.27% of an exhaustive L3 scan, while exceeding the strongest exhaustive baseline by 3.12 percentage points in R@1. One model trained on L1, L2, and L3 also outperforms a matched dense control across all six galleries and transfers to three withheld resolutions. By decoupling globally trained embeddings from local hierarchical search, GeoMoE jointly improves localization accuracy, search efficiency, and cross-resolution transfer. --- ### One Query, Many Scales: Sparse Mixture-of-Experts for Efficient Hierarchical Cross-View Geo-Localization Authors: Ruijie Fan, Junyan Ye, Qi Zhu, Weijia Li Published: 2026-08-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.01060v1 Cross-view geo-localization (CVGL) retrieves geo-tagged satellite imagery for a ground-view query. Most systems exhaustively search a flat, fixed-resolution gallery, incurring high cost over large areas and adapting poorly to satellite resolution changes. Autoregressive coarse-to-fine alternatives reduce comparisons but bind later predictions to earlier decisions and a predefined hierarchy. We introduce GeoMoE, a sparse mixture-of-experts dual encoder that decouples global multi-scale representation learning from local hierarchical search. Global multi-scale supervision and content-adaptive routing map ground and satellite images across resolutions into a globally comparable embedding space. At inference, each image is encoded once, and probabilistic beam search follows parent--child links to score a small candidate subset. Later levels reuse these descriptors rather than features generated by preceding levels, limiting feature-level error propagation and hierarchy coupling. We further introduce VIGOR-M, a four-city benchmark with an explicit parent--child satellite hierarchy and held-out half-step galleries for single-resolution, cross-resolution, and hierarchical evaluation. GeoMoE achieves 95.78% R@40m on Just Zoom In, 2.77 percentage points above the previous best, and 62.39% R@1 on VIGOR-M. The latter requires 0.885 MMAC/query for descriptor matching, 5.27% of an exhaustive L3 scan, while exceeding the strongest exhaustive baseline by 3.12 percentage points in R@1. One model trained on L1, L2, and L3 also outperforms a matched dense control across all six galleries and transfers to three withheld resolutions. By decoupling globally trained embeddings from local hierarchical search, GeoMoE jointly improves localization accuracy, search efficiency, and cross-resolution transfer. --- ### Extended KAFR: A kinematic-adaptive paradigm for the efficient analysis of surgical video Authors: Huu Phong Nguyen, Shekhar Madhav Khairnar, Ganesh Sankaranarayanan Published: 2026-08-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.01058v1 Artificial Intelligence is increasingly applied to surgical video analysis for phase segmentation, skill assessment, and workflow optimization. A key challenge is the length of surgical recordings, often one to several hours, creating substantial computational burden. We previously developed Kinematics-Adaptive Frame Recognition (KAFR) for robotic surgery, showing that tracking tool motion effectively identifies informative frames while filtering redundant content. However, laparoscopic surgery introduces additional challenges: manual camera control causes frequent motion artifacts, and image quality is generally lower than robotic systems. This study evaluates whether KAFR generalizes to laparoscopic surgery using the Cholec80 benchmark, comprising 80 laparoscopic cholecystectomy procedures annotated for seven surgical phases. KAFR operates in three stages: a fine-tuned YOLO model detects and segments surgical tools; frames are adaptively selected based on tool displacement or velocity variation; and an X3D model classifies selected frames into surgical phases. KAFR achieved a 91.0\% F1 score using only 0.58\% of frames for phase classification, representing an approximately seven-fold reduction compared to typical 4\% frame sampling, while maintaining performance comparable to LoViT (90.2\%) and Trans-SVNet (89.7\%). These results demonstrate that kinematics-based frame selection transfers effectively to the challenging laparoscopic environment. --- ### Control Under Compression: Reliability Frontiers for Tool-Using Agents Authors: Yinghan Hou, Zongyou Yang Published: 2026-08-02 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.01056v1 Tool-using language-model agents are governed not only by task prompts but also by persistent system-side instructions that specify tools, arguments, policies, execution protocols, and recovery. Compressing these agent control contexts (ACCs) can reduce input cost and context use, yet existing prompt-compression evaluations do not reveal whether the resulting control remains operationally reliable. We introduce CompressAgent, an environment-verified benchmark for ACC compression across nine independently constructed ACCs, three task families, three fixed Qwen API model identifiers, six retained-context budgets, and 15,525 runs. We uncover a nonlinear, method-dependent reliability frontier. At 75% retained context, generic rewriting and section-based compression achieve 92.7% and 92.4% success, close to the 93.8% full-context baseline. Between 50% and 35%, methods diverge sharply; at 35%, section-based, obligation-aware, and generic rewriting achieve 47.0%, 39.0%, and 19.9%. At retained-context budgets from 25% to 10%, executable protocols become fragile. Reliability also varies substantially across ACCs, making universal compressor rankings inappropriate and motivating per-context qualification. Failure analysis shows that compression primarily surfaces as tool-execution and action-parsing errors. These findings recast ACC compression from token reduction into a runtime-reliability problem that must be evaluated through executable outcomes. --- ### Credit the Right Box: Marginal Contribution Assignment for Structured Visual Perception Authors: Xinheng Han, Jianfei Wang, Yu Chen, Xiang Wang, Shuai Li, Weixing Li, Feng Pan Published: 2026-08-02 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.01055v1 Multimodal Large Language Models (MLLMs) are increasingly expected to solve structured perception tasks that require visual recognition, language-to-object binding, object cardinality preservation, and precisely localized grounding and segmentation outputs. However, existing group-relative reinforcement learning methods provide only response-level supervision, creating a granularity mismatch for structured multi-object prediction: a single advantage is broadcast to all tokens in a response, without distinguishing individual box contributions. To address this mismatch, we propose MCR-GRPO, a marginal contribution assignment framework that derives box-level credit directly from each sampled response. Specifically, Marginal Contribution Reward (MCR) estimates each predicted box's contribution through a leave-one-out comparison, measuring how the matched set value changes when the box is removed from the response. After within-response normalization, records that improve the set value receive positive credit, while redundant or harmful ones are suppressed. To make marginal attribution stable and informative, we further introduce a Continuous Matched Set Value Evaluator that integrates permutation-invariant matching, count-aware normalization, and graded localization. MCR-GRPO maps normalized box-level marginal advantages to the token spans that generated each box, preserving GRPO's response-level comparison while enabling box-aware optimization of structured multi-object grounding. Experiments across REC, DOD, segmentation, and counting benchmarks show state-of-the-art performance over prior GRPO-based baselines. --- ### Struct-GStream: Towards Efficient Free-Viewpoint Video Streaming at Low-Bitrates with Structured 3D Gaussians Authors: Han Jiao, Jiakai Sun, Lei Zhao, Wei Xing, Huaizhong Lin, Zhanjie Zhang, Ao Ma Published: 2026-08-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.01053v1 Constructing photorealistic Free-Viewpoint Videos (FVVs) of dynamic scenes from a set of posed 2D images has been an intriguing yet challenging task in computer vision. Methods based on neural rendering achieve high-fidelity image quality in FVV construction. However, most of these methods are unable to achieve real-time rendering and often require complete video sequences to train. Despite the existence of some online training methods capable of rendering FVVs in real time, they struggle to meet the requirements for storage and training time for downstream applications. To overcome this problem, we propose Struct-GStream, which can achieve efficient FVV streaming using structured 3D Gaussians (3DGs). Specifically, we introduce dynamic anchor points to generate structured 3DGs to construct basic scenes and model approximate scene movements based on the assumption of local rigidity in object motion. Besides, we introduce a global free 3DGs patching strategy involving free 3DGs' generation, pruning, and optimization to patch and model deficient areas and emerging objects. Our method achieves fast training at low bitrates while maintaining high rendering quality. Extensive experiments demonstrate that Struct-GStream significantly outperforms existing online training methods for FVV construction in terms of training time, storage, and rendering quality while maintaining competitive rendering speed. --- ### Don't Offer What Can't Be Done: Deterministic Executability Gating for LLM Skill Selection at Scale Authors: Ortal Ashkenazi, Vitalii Kloz, Mykhailo Ulianchenko Published: 2026-08-02 Categories: cs.AI, cs.CL, cs.SE arXiv: https://arxiv.org/abs/2608.01050v1 Production LLM agents that select from large skill libraries face a limitation that semantic relevance alone cannot resolve: a skill may match a user's topic yet be impossible to execute in the current account state. We present a deployed three-stage selection pipeline for Helpmate, Wix's customer-care assistant. First, a recall-oriented semantic matcher identifies messages related to a ten-skill domain family without consulting account state. Second, a deterministic executability gate removes candidates whose internal hard-stop conditions hold. Because the gate and the skill evaluate the same exit predicates, every blocked candidate would be unable to complete under the same account state, provided predicate parity is preserved and both checks observe fresh authoritative state. Finally, the LLM decides whether to invoke one of the remaining candidates. In a post-launch production analysis of 756.6K user messages across 267.6K conversations, semantic matching retained 174,927 messages (23.1%). Within this matched stream, the gate removed 1,039,462 of 1,749,270 skill-message pairs (59.4%), saving 228.8 million skill-description tokens -- 59.1% of the post-semantic skill-description footprint. Together, semantic matching and executability gating reduced skill-description context by 90.5% relative to exposing all ten skills to every message. To test whether this pruning affects model behavior rather than context size alone, we replayed a risk-enriched cohort of 1,000 conversations with all ten skills exposed. The model selected a production-blocked skill in 78 conversations (7.8%). This counterfactual result shows that deterministic gating prevents non-executable candidates from influencing model selection, while not claiming downstream tool execution or customer-outcome effects. --- ### FactorJEPA: Factorizing Monolithic Futures into Layout-Agent-Interaction Channels for Crowded and Chaotic Global South Urban Worlds Authors: Kapil Wanaskar, Gaytri Jena, Aman Chadha, Vinija Jain, Vasu Sharma, Amitava Das Published: 2026-08-02 Categories: cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2608.01049v1 World models have attracted significant attention for their ability to capture and predict the structure and dynamics of the physical world. In this emerging landscape, Joint Embedding Predictive Architectures (JEPA) offer a particularly compelling direction. We study a largely unexplored regime: populous, crowded, and chaotic Global South urban environments, which we call DENSEWORLD. Unlike the lower-density, lane-structured settings that dominate existing evaluations, these scenes exhibit soft spatial boundaries, extreme agent heterogeneity, persistent occlusion, and rapid social negotiation under mixed traffic. We introduce the first large-scale dataset for this regime: 1,000 hours of drive-through, walk-through, and aerial video across 22 cities. Existing JEPA formulations struggle to preserve dense interaction dynamics under heterogeneity and partial observability. We introduce FactorJEPA, which makes world structure a first-class predictive primitive. Rather than encoding the future in a monolithic latent, it composes layout, entities, and interactions, using a visibility gate and separated subspaces to preserve partially observed agents and discourage cross-factor shortcuts. FactorJEPA improves (i) future-latent accuracy (Future-frame L1), (ii) intervention-sensitive prediction (Causal L1), and (iii) robustness to reduced visual evidence (Mask-ratio slope), while exposing (iv) a reproducible motion-information trade-off (Motion cosine). Method rankings replicate across 2B and 1B V-JEPA 2.1 backbones, with rho = 0.895 to 0.978. We publicly release the DENSEWORLD-115k dataset (https://huggingface.co/datasets/anonymousML123/denseworld-115k) and the surgery-trained FactorJEPA checkpoints (https://huggingface.co/datasets/anonymousML123/factorjepa-outputs/tree/main/outputs/full/vjepa_2_1_vitg_1B/train/m09c_surgery_3stage_DI_diheavy_encoder). --- ### DeBERTa-Sentinel: Toward Transparent and Trustworthy Detection of AI-Generated Text Authors: Muhammad Yousaf Rehman, Muhammad Islam Published: 2026-08-02 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.01046v1 The rapid spread of large language models (LLMs) across the web raises concerns about misinformation, academic integrity, automated content manipulation, and risks to vulnerable online communities. Existing transformer-based detectors, such as GPT-Sentinel, show promise but struggle to generalize to diverse model outputs and paraphrasing attacks, limiting their role in building trustworthy web ecosystems. This work introduces DeBERTa-Sentinel, a responsible AI-generated text detection framework leveraging DeBERTa-v3's disentangled attention to capture subtle structural irregularities in synthetic content. A central design principle is transparency: unlike black-box commercial detectors, DeBERTa-Sentinel exposes token-level explanations of its decisions, enabling affected stakeholders journalists, educators, and platform trust and safety teams to audit, challenge, and contextualize detection outcomes. Using the GLC-AIText dataset of 28,057 human and LLM-generated samples (GPT, LLaMA, and Claude) with a 60-20-20 split, DeBERTa-Sentinel achieves 98.21\% validation accuracy and surpasses the RoBERTa-Sentinel baseline from NeurIPS 2025, achieving 97.53\% test accuracy, 95.89\% precision, 99.33\% recall, and 99.53\% ROC-AUC, and maintaining a 0.665\% false negative rate. The model's interpretability reveals linguistic markers such as academic phrasing and formal transitions associated with synthetic text, directly supporting stakeholder needs for verifiable, auditable content-authenticity decisions. By advancing responsible detection methods that reduce bias and enhance explainability, DeBERTa-Sentinel promotes trustworthy, ethical, and human-centric AI systems. Code and data are available at https://github.com/Galileo-Galili/HUMAN-VS-AI-TEXT-DETECTION. --- ### Decoy Images Amplify Caption-Mediated Defenses Against Encoded Jailbreaks Authors: Haoyu Zhang, Xiangchen Guan, Shibo Zheng, Mohammad Zandsalimy, Shanu Sushmita Published: 2026-08-02 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2608.01043v2 We report a counter-intuitive interaction between image inputs and existing black-box defenses on Vision--Language Models (VLMs): pairing an encoded jailbreak prompt with an unrelated decoy image can sharply lower attack success rate (ASR). The operative change is in the defense pipeline, not in the image. Across five frontier VLMs, two encoded-attack families, and three black-box defenses, a caption-mediated defense (ECSO) that leaves ASR essentially unchanged on text-only encoded input drops it by up to $73$pp once a content-free decoy is attached; every non-saturated contrast is significant under exact McNemar tests. We advance two hypotheses for this pattern, supported by indirect evidence rather than pipeline introspection, since a black-box threat model precludes inspecting vendor internals: caption-mediated defenses branch on image presence, and intrinsic image-side safety engages on image-resident content. Three controls constrain the explanation. Blank-canvas and natural-photograph decoys reproduce the effect on every model, implicating image presence rather than content; the effect replicates on three open-weight VLMs served with no moderation layer, so it is not a vendor-filtering artifact; and a non-symbolic, meaning-based encoder reproduces it, so it is not specific to symbolic obfuscation. Attaching a decoy unconditionally is not deployable --- it raises benign refusal to $20$--$79\%$, an inflation of $+10$ to $+67$pp --- but gating attachment on a lightweight encoded-input detector returns benign refusal to the text baseline while preserving the safety gain wherever the detector fires, making detector recall the binding constraint. Under adaptive attacks that target the caption-mediated re-check, the effect degrades but holds. We frame this as an observation about pipeline interaction, not as a robust defense. --- ### Decoy Images Amplify Caption-Mediated Defenses Against Encoded Jailbreaks Authors: Haoyu Zhang, Xiangchen Guan, Shibo Zheng, Mohammad Zandsalimy, Shanu Sushmita Published: 2026-08-02 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2608.01043v1 We report a counter-intuitive interaction between image inputs and existing black-box defenses on Vision--Language Models (VLMs): pairing an encoded jailbreak prompt with an unrelated decoy image can sharply lower attack success rate (ASR). The operative change is in the defense pipeline, not in the image. Across five frontier VLMs, two encoded-attack families, and three black-box defenses, a caption-mediated defense (ECSO) that leaves ASR essentially unchanged on text-only encoded input drops it by up to $73$pp once a content-free decoy is attached; every non-saturated contrast is significant under exact McNemar tests. We advance two hypotheses for this pattern, supported by indirect evidence rather than pipeline introspection, since a black-box threat model precludes inspecting vendor internals: caption-mediated defenses branch on image presence, and intrinsic image-side safety engages on image-resident content. Three controls constrain the explanation. Blank-canvas and natural-photograph decoys reproduce the effect on every model, implicating image presence rather than content; the effect replicates on three open-weight VLMs served with no moderation layer, so it is not a vendor-filtering artifact; and a non-symbolic, meaning-based encoder reproduces it, so it is not specific to symbolic obfuscation. Attaching a decoy unconditionally is not deployable --- it raises benign refusal to $20$--$79\%$, an inflation of $+10$ to $+67$pp --- but gating attachment on a lightweight encoded-input detector returns benign refusal to the text baseline while preserving the safety gain wherever the detector fires, making detector recall the binding constraint. Under adaptive attacks that target the caption-mediated re-check, the effect degrades but holds. We frame this as an observation about pipeline interaction, not as a robust defense. --- ### What Could the Agent See at 19:05? Generating Temporal Enterprise Scenarios from Real Research and Replaying Them to Evaluate Agents Authors: Tezan Sahu, Himani Arora Published: 2026-08-02 Categories: cs.SE, cs.AI, cs.HC, cs.LG arXiv: https://arxiv.org/abs/2608.01042v1 Enterprise AI agents act across many apps whose data changes continuously, so an answer is correct only relative to what data existed and who could see it at the moment it was asked. Offline evaluation today grades against a single static snapshot, effectively the end of the episode. So, it can only evaluate one situation, the final one, even though every earlier moment of the episode is a different situation that invites its own realistic questions with its own correct answers. Recreating each of those moments as a separate snapshot would mean re-provisioning a whole tenant per instant, which is prohibitively costly; and even a single snapshot leaks future state hidden inside records and cannot represent the multi-app, time-ordered way real work happens. Our system closes two gaps at once: it generates a realistic, persona-driven, temporally-evolving enterprise world from real research, and replays that world at any chosen moment to evaluate any pluggable agent. A schema-inferred temporal description drives a deterministic-plus-LLM rebuild of each record's past state; because the queryable moments are finite, all rebuilds are precomputed into a compact difference cache, making evaluation a fast, reproducible lookup with no model in the path. We describe the design, an architecture spanning both flows, and early experience evaluating enterprise agents. --- ### On the Limits of Machine-Learned Ranking for Modern Microarchitectural Policies Authors: Yanxin Zhang, Shayne Wadle, Yuxuan Xiong, Zheyu Fu, Trivikram Krishnamurthy, Karthikeyan Sankaralingam Published: 2026-08-02 Categories: cs.AR, cs.LG arXiv: https://arxiv.org/abs/2608.01041v2 Machine-learning predictors estimate processor performance far faster than cycle-level simulation. For design-space exploration, however, the valuable test is not merely reproducing the usual hardware ordering, but identifying how different hardware configurations rank on individual program phases. We evaluate four ML-predictors in two design regimes: \emph{Structural Parameters} (SP), varying hardware resources such as issue width, ROB size, and cache capacity; and \emph{Behavioral Policies} (BP), varying prefetching and replacement algorithms. In the SP regime, aggregate ranking is strong, yet counter-intuitive windows(CIW)---where the configuration expected to be slower is faster---constitute $22.4\%$ of non-tied windows across five pairs with a clear architectural prior. CIW match across these pairs is only $23.3$--$39.9\%$; every point estimate is below the $50\%$ random strict-ordering reference. The BP regime presents a different failure: ground-truth ties cover $37.8\%$ of pair-windows, most strict pairs have margins of only a few cycles, and no model family reliably beats a feature-free majority baseline. NeuroScalar and SimNet fall below that baseline, Concorde is statistically tied with it, and the best selected OneDSE head improves by only $2.1$ percentage points. Accuracy rises mainly at large margins. We further show that this failure is not a matter of model capacity: an information-theoretic analysis reveals that when ranking outcomes depend on hidden microarchitectural state absent from the instruction stream, no trace-based predictor can exceed the Bayes accuracy determined by observable inputs alone. Thus high cycle or aggregate ranking accuracy can reflect mastery of easy, high-margin cases while missing the local reversals that carry the most architectural insight and for which cycle-level simulation remains indispensable. --- ### On the Limits of Machine-Learned Ranking for Modern Microarchitectural Policies Authors: Yanxin Zhang, Shayne Wadle, Yuxuan Xiong, Zheyu Fu, Trivikram Krishnamurthy, Karu Sankaralingam Published: 2026-08-02 Categories: cs.AR, cs.LG arXiv: https://arxiv.org/abs/2608.01041v1 Machine-learning predictors estimate processor performance far faster than cycle-level simulation. For design-space exploration, however, the valuable test is not merely reproducing the usual hardware ordering, but identifying how different hardware configurations rank on individual program phases. We evaluate four ML-predictors in two design regimes: \emph{Structural Parameters} (SP), varying hardware resources such as issue width, ROB size, and cache capacity; and \emph{Behavioral Policies} (BP), varying prefetching and replacement algorithms. In the SP regime, aggregate ranking is strong, yet counter-intuitive windows(CIW)---where the configuration expected to be slower is faster---constitute $22.4\%$ of non-tied windows across five pairs with a clear architectural prior. CIW match across these pairs is only $23.3$--$39.9\%$; every point estimate is below the $50\%$ random strict-ordering reference. The BP regime presents a different failure: ground-truth ties cover $37.8\%$ of pair-windows, most strict pairs have margins of only a few cycles, and no model family reliably beats a feature-free majority baseline. NeuroScalar and SimNet fall below that baseline, Concorde is statistically tied with it, and the best selected OneDSE head improves by only $2.1$ percentage points. Accuracy rises mainly at large margins. We further show that this failure is not a matter of model capacity: an information-theoretic analysis reveals that when ranking outcomes depend on hidden microarchitectural state absent from the instruction stream, no trace-based predictor can exceed the Bayes accuracy determined by observable inputs alone. Thus high cycle or aggregate ranking accuracy can reflect mastery of easy, high-margin cases while missing the local reversals that carry the most architectural insight and for which cycle-level simulation remains indispensable. --- ### Using Lower-Bound Representations for Trajectory Similarity Learning Authors: Liwei Deng, Haotian Meng, Yupu Zhang, Yan Zhao, Torben Bach Pedersen, Kai Zheng, Christian S. Jensen Published: 2026-08-02 Categories: cs.DB, cs.LG arXiv: https://arxiv.org/abs/2608.01039v1 Trajectory similarity learning is fundamental to efficient trajectory retrieval under complex distance measures. Existing learning-based methods typically rely on embeddings trained to approximate trajectory distances or rankings, but they often lack guarantees with respect to the original distances, exhibit unstable performance across distance measures, and incur substantial training costs. We revisit trajectory similarity learning from a lower-bound representation perspective and propose LB-TrajRep, a unified lower-bound representation framework independent of deep neural embeddings. This framework constructs single-vector representations from a set of lower-bound components, enabling admissible and interpretable lower bounds for multiple classical trajectory distances, including Dynamic Time Warping (DTW), Hausdorff distance, and Discrete Fréchet Distance (DFD). Within this framework, we instantiate point-pivot components, which naturally support both metric and non-metric distances and remain compatible with standard vector-based retrieval pipelines. To improve ranking quality, we develop two data-driven pivot selection strategies that explicitly optimize lower-bound tightness and prioritize hard near-neighbor trajectory pairs, respectively. Extensive experiments on real-world trajectory datasets show that the proposed lower-bound representations are able to consistently outperform state-of-the-art neural trajectory embeddings across diverse distance measures, improving top-$k$ ranking accuracy by up to 20\%--60\% on the Hausdorff distance and DFD and by 15\%--40\% on DTW. --- ### WAM-Diff2: Hierarchical AR-to-Diffusion Distillation for Highly Efficient Autonomous Driving VLA Authors: Zhihao Zhu, Hanlin Shang, Mingwang Xu, Feipeng Cai, Zhuolin He, Yaoyi Li, Jianhua Han, Hang Xu, Siyu Zhu Published: 2026-08-02 Categories: cs.RO, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2608.01035v4 Vision-Language-Action (VLA) models have emerged as a prominent paradigm for end-to-end autonomous driving; however, their efficient deployment is severely constrained by high computational latency and exposure bias arising from sequential autoregressive decoding. Conversely, while specialized diffusion policies enable low-latency, parallel execution, training them from scratch typically yields narrow, single-task architectures that lack holistic visual-linguistic reasoning. Successfully transforming pre-trained autoregressive generalists into parallel diffusion models could combine multi-task cognitive intelligence with execution efficiency, yet this transition presents a formidable architectural challenge due to mismatched attention patterns (causal versus bidirectional) and divergent optimization objectives. To bridge this divide, we introduce WAM-Diff2, a multi-task discrete diffusion VLA framework powered by a three-stage hierarchical distillation strategy. By structuring the architectural shift through progressive block-wise adaptation, block-wise distillation, and model-wise cross-scale distillation, WAM-Diff2 preserves the underlying semantic foundations of the base model while accelerating inference. Extensive evaluations across driving understanding, perception, and planning benchmarks demonstrate that WAM-Diff2 effectively mitigates exposure bias and achieves performance parity with autoregressive baselines. Crucially, the autoregressive-to-diffusion transition yields a 2.8x decoding speedup, which scales to an ultimate 15.1x acceleration when combined with system-level optimizations including FlashInfer and CUDA Graphs. --- ### WAM-Diff2: Hierarchical AR-to-Diffusion Distillation for Highly Efficient Autonomous Driving VLA Authors: Zhihao Zhu, Hanlin Shang, Mingwang Xu, Feipeng Cai, Zhuolin He, Yaoyi Li, Jianhua Han, Hang Xu, Siyu Zhu Published: 2026-08-02 Categories: cs.RO, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2608.01035v1 Vision-Language-Action (VLA) models have emerged as a prominent paradigm for end-to-end autonomous driving; however, their efficient deployment is severely constrained by high computational latency and exposure bias arising from sequential autoregressive decoding. Conversely, while specialized diffusion policies enable low-latency, parallel execution, training them from scratch typically yields narrow, single-task architectures that lack holistic visual-linguistic reasoning. Successfully transforming pre-trained autoregressive generalists into parallel diffusion models could combine multi-task cognitive intelligence with execution efficiency, yet this transition presents a formidable architectural challenge due to mismatched attention patterns (causal versus bidirectional) and divergent optimization objectives. To bridge this divide, we introduce WAM-Diff2, a multi-task discrete diffusion VLA framework powered by a three-stage hierarchical distillation strategy. By structuring the architectural shift through progressive block-wise adaptation, block-wise distillation, and model-wise cross-scale distillation, WAM-Diff2 preserves the underlying semantic foundations of the base model while accelerating inference. Extensive evaluations across driving understanding, perception, and planning benchmarks demonstrate that WAM-Diff2 effectively mitigates exposure bias and achieves performance parity with autoregressive baselines. Crucially, the autoregressive-to-diffusion transition yields a 2.8x decoding speedup, which scales to an ultimate 15.1x acceleration when combined with system-level optimizations including FlashInfer and CUDA Graphs. --- ### Opt.Gear Technical Report Authors: Juneyoung Park, Youngwook Kwon Published: 2026-08-02 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.01034v2 We introduce OptGear, a foundation model designed for efficient on-device deployment, real-tim inference, and strong task capability. It includes a dense model (1M, 270M, and 1B) with a context length of 64K. We designed a new hybrid architecture that combines a convolutional key-value gated mixer with local-global attention to reduce the KV-cache memory that tends to increase exponentially with long context. This architecture delivers up to X4.9 faster prefill and decoding speeds on the NPUs compared to models of a similar scale models. From a 2T tokens candidate corpus, OptGear is trained on a curated 0.5T tokens subset without knowledge distillation. This is the most data-efficient of the existing foundation models. All models are released with open weights and deployment binaries for ONNX, Qualcomm NPU, and Apple ANE making OptGear a practical base for edge applications that need fast, memory-efficient inference and strong task capabilities. Furthermore, to expand the ecosystem of on-device generative language models, we are introducing the OptGear-1M that can be deployed on Micro-Controller Units (MCUs), a Tiny Language Model (TLM). OptGear-1M is the first generative language model to achieve 20 TPS with W4A32 quantization on the ARM Cortex-M7 CPU of the STM32H747I-DISCO. --- ### CallScreenBench: Benchmarking Small Language Models as Phone Secretaries Authors: Jiaqi Gan, Haoyuan Tang, Jamey Z. Liang, Siying Chen, Ankit Raj, Kidus Zewde, Yuchen Zhou, Yuxin Zhang, Simiao Ren Published: 2026-08-02 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2608.01033v2 Language models small enough to run on a handset, quantized to a few bits, are increasingly capable of acting on their user's behalf -- which makes on-device task automation newly plausible. One such task is answering the phone. A phone secretary takes an unknown inbound call on its owner's behalf, and unlike the agents most benchmarks evaluate, it has no cooperative caller-assigned task to complete: the caller holds the goal and may be an adversary, while the secretary must begin deciding how to respond without an oracle. What matters is not task success but whether the owner would endorse how their proxy handled the call. We evaluate only the text-domain conversational decision layer; speech recognition, audio interaction, end-to-end latency, and handset execution are outside scope. We present CallScreenBench, which reports five automated call-and-note measure groups motivated by owner endorsement. Each is paired, where available, with a counter-metric and an uncertainty estimate; no benchmark-wide Q1-Q5 composite or leaderboard score is defined. Three guardedness diagnostics identify candidate cases for a toolless proxy that holds no credentials and calls no tools. Across three model families represented by paired 4-bit checkpoints (0.6-4B), the primary scoring snapshot gives the larger checkpoint higher point estimates on several service, recall, and plausibility measures, while triage discrimination follows a different ordering. Bare scam-side TPR rewards universal suspicion, and pairwise separation changes when legitimate-side false positives are included and across judge snapshots. Scripted degenerate agents expose further floors, including a hangup-and-echo policy with entity recall 1.000. We report quality measures and guardedness channels separately so that a single pass/fail score does not hide their trade-offs. --- ### CallScreenBench: Benchmarking On-Device Models as Phone Secretaries Authors: Simiao Ren Published: 2026-08-02 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2608.01033v1 Language models small enough to run on a handset, quantized to a few bits, are increasingly capable of acting for their user, making on-device task automation newly plausible. One such task is answering the phone. A phone secretary takes an unknown inbound call on its owner's behalf. Unlike the agents evaluated by most benchmarks, it has no task to complete and no cooperative user: the caller holds the goal, may be an adversary, and must be judged from the opening turn with no oracle. What matters is not task success, but whether the owner would endorse how their proxy handled the call. We present CallScreenBench, which scores this setting on five quality dimensions. Each dimension is printed beside the counter-metric that bills it and is never averaged into a single number. We also report a guardedness profile for a toolless proxy that holds no credentials and calls no tools. Across six on-device models (0.6-4B parameters, 4-bit quantization), quality scales with capability, but triage does not. The appearance that it does is an artifact of measurement. Scripted degenerate agents supply the missing floors: after correcting for them, the number of model pairs whose triage performance separates falls from 11 of 15 to zero at the preregistered operating point. An agent that simply hangs up and echoes the caller also scores perfect message fidelity. We report which of our own metrics these floors defeat and declare no pass/fail threshold. --- ### The Fourth Quadrant: A Stylized View of Benign Misfitting Authors: Gireeja Ranade, Anant Sahai Published: 2026-08-02 Categories: cs.LG, cs.IT arXiv: https://arxiv.org/abs/2608.01032v1 Training error is what we can observe on a training set; test error is the quantity we actually care about. We study linear regression with squared-error in a deterministic $(d+1)$-dimensional single-spike model. Each stylized training vector has the same informative spike coordinate, of amplitude $\sqrtγ$ with $γ>1$. The remaining directions are nuisance, and the nuisance components of distinct training vectors all have equal norm and are mutually orthogonal. The training labels are all $1$. Fresh test points are drawn from $\vec{x}_{\rm test} \sim \mathcal{N}(\vec{0},\operatorname{diag}(γ,1,\ldots,1))$, with the noise-free test labels being the normalized spike coordinate $x_{\rm test}(1)/\sqrtγ$. We focus on linear predictors in the span of the training vectors, the class naturally reached by zero-initialized linear gradient methods. We exhibit a range of training-set sizes $n$ in which every span predictor that generalizes well must fit the training data \emph{worse} than the zero predictor. We call this regime \emph{benign misfitting}, or the fourth quadrant. The best span predictor begins to generalize when $n\gg d/γ^2$, while interpolation does not generalize until the later threshold $n\gg d/γ$. In the window $d/γ^2 \ll n \ll d/γ$, useful prediction within the linear span lies beyond interpolation: predictions on the training points overshoot the labels. We show that one-pass stochastic gradient descent (SGD), with a large constant learning rate, reaches small test error throughout this window---matching the best span predictor up to a logarithmic factor. We also verify directly that it indeed has \emph{large} empirical training error (despite the descent premise in its name). Finally, we show that the unavoidable nuisance component responsible for the training misfit also controls the predictor's adversarial sensitivity. --- ### VLAGuard: A Framework for Evaluating and Mitigating Physical Attention Hijacking in Vision-Language-Action Robots within Wireless Sensor Networks Authors: Dongfu Yin, Jinquan Zhang Published: 2026-08-02 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2608.01028v1 Deploying Vision-Language-Action (VLA) robots as mobile edge nodes within wireless sensor networks (WSNs) requires robust protection against physical adversarial threats. We present VLAGuard, a framework to assess and mitigate a critical vulnerability: policy-critical action-to-vision attention hijacking. We first introduce a stress-test module, Visuomotor Attention-guided Semantic Attack (VASA), using printable patches to severely distract the robot's action-conditioned cross-attention. To counter this, we propose Attention-Protective Fine-Tuning (APFT), a defense that stabilizes spatiotemporal attention and enforces geometric consistency with zero inference overhead. Evaluations across simulated and physical WSN-assisted smart environments demonstrate significant robustness gains. APFT reduces the OpenVLA failure rate from 100.0% to 25.9% in LIBERO simulations. Furthermore, across 2,000 real-world trials, APFT improves the average success rate from 23.0% to 67.4% under severe patch attacks. This highlights that protecting attention pathways is important for improving the robustness of VLA-driven edge nodes in sensor networks. --- ### Stress-Relief Annealing: Polynomial-Time Simulation-Free Layout Optimization for Automated Warehouses Authors: Xiangjie Luo, Yulun Zhang, Miyuki Koshimura, Makoto Yokoo, Jiaoyang Li Published: 2026-08-02 Categories: cs.MA, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2608.01024v1 We study the problem of optimizing physical layouts for automated warehouses, where hundreds to thousands of robots are coordinated to transport packages. Previous works have shown that optimizing the warehouse layout (e.g., the physical location of the storage shelves) significantly improves throughput. However, state-of-the-art layout optimization approaches are based on evolutionary optimization methods, which treat the entire warehouse as a black box and rely on random mutation to search for high-quality layouts. While the optimization outcomes are promising, these methods require a massive number of simulations to evaluate candidate solutions, making them sample-inefficient. In this paper, we present Stress-Relief Annealing (SRA), a polynomial-time simulation-free layout optimization algorithm. SRA turns the task demand into a per-vertex \emph{stress field} that predicts where traffic will concentrate in the warehouse; the field's peak provably caps the throughput. Our experimental results show that (1) SRA improves both the throughput and the scalability of a human-designed warehouse, roughly doubling the number of robots it can sustain, (2) it matches or exceeds the throughput of the evolutionary baselines while taking only $19$ minutes on one CPU core, against their $25{,}000$ simulations and $25$ hours on a $64$-core machine, and (3) the gain generalizes across different Multi-Agent Path Finding algorithms, non-uniform task demands, and a warehouse with doubled dimensions. --- ### Caliber: Cross-Architecture Extraction-Cost Control for Score-Returning APIs Authors: Chi Wang, Hanwen Wang, Yu Xia, Zihan Wang, Guangdong Bai Published: 2026-08-02 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.01023v1 We present Caliber, an output-perturbation defense against model extraction that formulates noise selection as a calibration problem: how much the defense degrades the supervision signal used to train a surrogate, and the provable per-input query cost of recovering the clean logits. To defend against an attacker that uses returned scores for knowledge distillation, Caliber adds independent and identically distributed Gaussian noise to the internal logits. We establish two properties of the resulting perturbed predictions. Monotone agreement degradation: When the clean logits have a unique maximizer, agreement with the clean prediction decreases strictly with the noise scale, so every target in $(1/K,1)$ corresponds to a unique positive scale; task accuracy is bounded by computable lower and upper envelopes. Per-input recovery cost: We derive a closed-form minimax lower bound on the repeated queries needed to recover the clean logits for a fixed input. Caliber normalizes noise variance by the squared median top-two logit margin and fits the resulting noise-utility relationship with a logistic curve, either per model or shared within a task. Across more than thirty model-dataset combinations, per-model calibration achieves mean absolute relative errors of 0.6-1.4%. End-to-end experiments show that surrogate performance generally tracks the configured degradation, while fixed-input averaging follows the expected variance reduction. --- ### Can Humans Dream of Electric Sheep? Human-Written Samples for Fine-Grained Vision-and-Language Hallucination Benchmarking Authors: Timothee Mickus, Claudio Savelli, Eduardo Calò, Emilio Raimond, Stella Frank, Hengyu Luo, Flavio Giobergia, Vincent Segonne, Chuyuan Li, Aman Sinha, Lorenzo Vaiani, Jörg Tiedemann et al. Published: 2026-08-02 Categories: cs.CV, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.01021v1 In an age of rapid model turnover, how do we make hallucination evaluation more perennial? We explore whether human-written hallucination samples could take the place of model-generated hallucinations, in order to make benchmarking detection independent of particular models. To this end, we construct a dataset of 1,600 human-written samples, spanning four languages (Chinese, English, French, Italian), and 18,400 samples from five vision-and-language models, all annotated for hallucinations using a fine-grained span-level labeling scheme. We find that human-written samples result in higher agreement and allow greater control of dataset contents, while remaining distributionally similar to samples derived from vision-and-language samples and providing a reasonable portrayal of detection capabilities - suggesting that human data is a viable substitute for model-based hallucination benchmarks. --- ### Why LLMs Give In: Conversational Factors and Reasoning Behind Medical Sycophancy Authors: Kaike Ping, Buse Çarık, Caleb Wohn, Xiaohan Ding, Tongshuai Wang, Eugenia Rho Published: 2026-08-02 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.01017v2 Large language models can answer a medical question correctly and still abandon that answer when a user pushes back. We study this failure as medical sycophancy and ask when models are most likely to give in. Across five open-weight models, 500 MedQuAD questions, and 1.2 million trials, we use a fully crossed design over four conversational factors: user role, user evidence, interaction structure, and grounding. Medical sycophancy is nearly three times more common when users challenge an answer the model has already given than when the false claim appears in the initial query. Models are also more susceptible to users presented as physicians or medical students. Most strikingly, fabricated evidence has opposite effects across interaction structures. It increases sycophancy in single-turn interactions but reduces it after the model has already answered. Grounding helps, but does not eliminate the behavior. Sycophancy varies more across medical questions than across models, making question selection an important part of benchmark design. Reasoning traces suggest that multi-turn failures are associated with models turning back toward their own prior answer, while fabricated evidence receives more scrutiny after an initial response. Together, the results show that medical sycophancy depends as much on how a model is challenged and evaluated as on which model is tested. --- ### Why LLMs Give In: Conversational Factors and Reasoning Behind Medical Sycophancy Authors: Kaike Ping, Buse Çarık, Caleb Wohn, Xiaohan Ding, Tongshuai Wang, Eugenia Rho Published: 2026-08-02 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.01017v1 A language model that abandons a correct medical answer under user pushback is more dangerous than one that was simply wrong, because it lends the credibility of a correct answer to the user's misinformation. Such model behavior, described as medical sycophancy, is usually reported as a single rate per model, but we find it is a property of the conversation, not the model. We study medical sycophancy in language models with a fully crossed factorial design over four conversational factors, user role, the evidence behind a false claim, whether the challenge precedes or follows the model's answer, and whether the correct answer is grounded in the prompt, across five open-weight models and 500 MedQuAD questions (1.2M trials). The factors interact sharply: fabricated sources raise sycophancy 2.0x when they accompany the question but halve it once the model has answered, so the same evidence helps or hurts depending only on timing. Sycophancy varies far more across questions than across models (67x vs. 3x), so a single rate reflects the conversation and the questions sampled as much as the model. Chain-of-thought traces explain why. Models that re-examine their own prior answer concede, while those that reason about the medical facts hold, and only a model that has already answered can spend a round auditing the fabricated source. --- ### KING: Embodiment-Aware Kinematic Graph Neural Network for Unified Motion Representation of Legged and Wheeled Robots Authors: Taku Okawara, Aoki Takanose, Kenji Koide, Shuji Oishi, Masashi Yokozuka Published: 2026-08-02 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2608.01015v1 Kinematic models provide reliable motion constraints for odometry estimation in featureless environments, where exteroceptive sensing degrades and IMU integration drifts. Learning-based kinematic models can achieve more accurate odometry estimation than model-based methods by capturing nonlinear effects; however, most existing learning-based models are trained on a single embodiment and generalize poorly to new embodiments. This generalization is difficult because the meanings and structures of proprioceptive measurements vary across embodiments, including the number of joints and ground-contact elements (e.g., wheels, feet). To address this challenge, we propose KING, a Graph Neural Network (GNN)-based kinematic model that explicitly incorporates robot embodiments by representing them as a common graph. We show that wheel and leg kinematic models can be expressed by a unified representation, enabling a single model for both wheeled and legged robots. Trained on datasets spanning diverse embodiments, KING provides a unified representation of wheeled and legged kinematics and achieves high-accuracy odometry estimation in real environments. KING estimates accurate odometry using only an embodiment description (e.g., a URDF file) and on-board proprioception (encoders and an IMU) and can be adapted to new robot embodiments through few-shot learning with only one minute of data, avoiding retraining from scratch on a new dataset for each robot. The project page is available at: https://smrg-aist.github.io/king_project_page/ --- ### Cloud-ScPO: Hidden-State Geometry for Semi-Supervised Preference Optimization in LLM Reasoning Authors: Yuzhou Liu, Xiyang Hu Published: 2026-08-02 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.01014v2 Preference optimization improves mathematical reasoning in large language models (LLMs), but reliable chosen-rejected pairs usually require verified answers, human annotations, or external reward models. We investigate whether preference supervision can instead be derived from the model's internal representation geometry in a semi-supervised setting. Our analysis shows that reasoning trajectories generated across different mathematical problems form structured global point clouds in which correct and incorrect trajectories exhibit different geometric organization. Based on this observation, we propose Cloud-ScPO, a topology-guided preference-mining framework that uses a small labeled set to construct multiple correct and incorrect reference Clouds. Each trajectory is represented by a mean-pooled hidden state and scored against connectivity-induced components using a component-level soft $k$-nearest-neighbor measure averaged across reference banks. We combine this cross-problem Cloud signal with prompt-level self-consistency: self-consistency determines the answer-level preference direction, while Cloud scoring selects concrete trajectories and filters pairs by their score margin. Experiments on GSM8K and MATH-Numeric across four model settings show that Cloud-ScPO consistently improves over ScPO, with gains of up to 4.49% on GSM8K and 4.19% on MATH-Numeric. Pair-level analyses further show that Cloud-ScPO maintains comparable correctness reliability while more effectively separating informative chosen trajectories from incomplete, repetitive, or otherwise low-quality rejected responses. --- ### Cloud-ScPO: Hidden-State Geometry for Semi-Supervised Preference Optimization in LLM Reasoning Authors: Yuzhou Liu, Xiyang Hu Published: 2026-08-02 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.01014v1 Preference optimization improves mathematical reasoning in large language models (LLMs), but reliable chosen-rejected pairs usually require verified answers, human annotations, or external reward models. We investigate whether preference supervision can instead be derived from the model's internal representation geometry in a semi-supervised setting. Our analysis shows that reasoning trajectories generated across different mathematical problems form structured global point clouds in which correct and incorrect trajectories exhibit different geometric organization. Based on this observation, we propose Cloud--ScPO, a topology-guided preference-mining framework that uses a small labeled set to construct multiple correct and incorrect reference Clouds. Each trajectory is represented by a mean-pooled hidden state and scored against connectivity-induced components using a component-level soft $k$-nearest-neighbor measure averaged across reference banks. We combine this cross-problem Cloud signal with prompt-level self-consistency: self-consistency determines the answer-level preference direction, while Cloud scoring selects concrete trajectories and filters pairs by their score margin. Experiments on GSM8K and MATH-Numeric across four model settings show that Cloud--ScPO consistently improves over ScPO, with gains of up to 4.49\% on GSM8K and 4.19\% on MATH-Numeric. Pair-level analyses further show that Cloud--ScPO maintains comparable correctness reliability while more effectively separating informative chosen trajectories from incomplete, repetitive, or otherwise low-quality rejected responses. --- ### MedUPS: Towards Diagnostic Assistance in Uncommon Medical Cases with Large Language Models Authors: Ofir Ben Shoham, Oriel Perets, Nir Grinberg, Nadav Rappoport Published: 2026-08-02 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.01012v1 Uncommon and off-guideline cases are difficult for clinical decision support, because physicians must make a series of management decisions under diagnostic uncertainty and rarely see the full case at once. Most large language model (LLM) benchmarks for medicine score only the final diagnosis, yet much of clinical care turns on the next appropriate action: the next test to order, the imaging study to obtain, the specialist to involve, or the differential to pursue. We introduce MedUPSQA, a dataset of 21,874 mid-stream clinical decision points built from 5,535 real case reports, and MedUPS, an alignment framework that supervises models on these intermediate decisions as they unfold along a patient's trajectory. We segment free-text case presentations into chronologically ordered, accumulating clinical chunks and align models to predict the next step with reinforcement learning (GRPO), using an external LLM-as-a-Judge reward. This objective mirrors how clinicians actually meet patients, reasoning forward from accumulating evidence toward the next decision, rather than committing to a final label. Across three backbones, mid-stream alignment raises next-step accuracy from 55.2 to 66.7 for Qwen3.6-27B, from 47.2 to 57.8 for Qwen3.5-9B, and from 37.8 to 44.4 for HuatuoGPT-3-8B, with 95% CI. In several model scales we test the objective improves accuracy more than scale, with smaller models surpassing larger, frontier models we evaluate. We further train supervised fine-tuning (SFT) baselines on the mid-stream task, SFT improves all backbones above base, indicating the target framwork carries signal independently of the optimizer. We release the dataset, code, and aligned checkpoints. --- ### Toward Fine-Grained Forgetting:Attribute Unlearning for Multimodal Large Language Models Authors: Junkai Lin, Junkai Chen, Siqi Hou, Yuhao He, Ruiqi Liu, Chenhan Jin, Shengze Xu, Tieyong Zeng Published: 2026-08-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.01008v1 Multimodal large language models (MLLMs) exhibit strong vision--language capabilities but may also memorize and disclose sensitive information. Machine unlearning seeks to remove designated knowledge without retraining from scratch while preserving general utility. Existing privacy-oriented benchmarks primarily adopt profile-level deletion, whereas practical requests are often finer grained: a model should forget a specified attribute while retaining non-sensitive information about the same identity. We therefore introduce attribute-level MLLM unlearning as a finer-grained task and construct a benchmark spanning long-text, numeric, and short-text targets, multiple forget ratios, and diverse question types. Our evaluation reveals that target and retained attributes share identity-specific and visual evidence, making selective forgetting susceptible to residual leakage or collateral degradation; accordingly, existing methods exhibit unstable forgetting--retention trade-offs in this setting. To address this challenge, we propose Causal Localization and Retain-Aware Projection (CLRP), a lightweight training-free framework. CLRP uses activation patching to identify the layer that causally mediates target-attribute disclosure, then applies a retain-aware projection that removes the target-attribute subspace while preserving same-identity evidence. Experiments across multiple widely used MLLMs with distinct architectures and parameter scales demonstrate the effectiveness of CLRP. --- ### Fused Bayesian Flow Networks for Dual-Target Molecular Design Authors: Jingyuan Zhou, Shikui Tu, Lei Xu Published: 2026-08-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.01007v1 Dual-target drug design aims to generate 3D molecules that can simultaneously interact with two target proteins, offering a promising route for discovering polypharmacological compounds against complex diseases. While recent generative models have shown encouraging performance in single-target drug design, existing dual-target approaches either focus on sequence generation or introduce an additional predictive drift term into the diffusion-based generative trajectory, which limits their ability to fully integrate feature information from both targets. We propose FusedBFN, a fused Bayesian flow network (BFN) for dual-target molecular design. FusedBFN formulates dual-target generation as distribution fusion in a unified continuous parameter space and employs a product-of-experts formulation to incorporate dual-target information throughout the generative process. To address the scarcity of dual-target structural data, we leverage a pretrained target-aware BFN model as the shared backbone. We further introduce a chemically aware prior-based alignment method and a prior-free pocket alignment strategy to construct aligned dual-target contexts. Extensive experiments demonstrate that FusedBFN generates molecules with strong binding affinity toward dual targets while maintaining favorable molecular properties. --- ### Hierarchical Solomonoff Induction: An Unbounded Machine Learning Model Authors: Nathan Young Published: 2026-08-02 Categories: cs.LG, cs.AI, cs.IT arXiv: https://arxiv.org/abs/2608.01005v1 Solomonoff Induction, or SolInd, provides an ideal unbounded model of a priori sequence prediction but cannot naturally describe extrapolation from a given training dataset, as performed by Large Language Models. We apply de Finetti's theorem on exchangeable distributions to SolInd to produce what we call Hierarchical Solomonoff Induction, or HSI, which maintains a hyperprior over all Solomonoff priors that can be conditioned on previously observed sequences. We extend Wood et al.'s proof that universal mixtures of semimeasures are equivalent to SolInd to show that universal mixtures of these mixtures are also equivalent, proving that HSI=SolInd. We also prove that HSI's excess error on any distribution, compared to its true generator, is bounded by that generator's complexity in the hyperprior. This result is directly comparable to SolInd's prediction error being bounded by the Kolmogorov complexity of the sequence being predicted, and forces HSI's average excess error to converge to 0 as a dataset grows, leading to optimal prediction in the limit. We claim that HSI is an ideal unbounded model of sequence prediction given a dataset in the same way that SolInd is ideal over individual sequences. --- ### Who Belongs in the Eval Set? A Capability-Taxonomy-Driven Pipeline for Curating Regression Eval Sets in Agent-Extensibility Platforms Authors: Tezan Sahu, Aritra Das, Pankaj Mittal, Sudipta Das Published: 2026-08-02 Categories: cs.LG, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2608.01004v1 Platform teams hosting agent-extensibility surfaces face a regression-economics paradox: every onboarding customer ships an evaluation set tuned to their domain, but the platform's regression set must live under a hard query-count ceiling bounded by release cadence. To our knowledge, no published industrial pipeline addresses this platform-side curation problem: existing evaluation frameworks are customer-side, and benchmark-compression work treats benchmarks as fixed pools rather than streams of incoming sets. We describe a capability-taxonomy-driven curation pipeline applied to declarative agents with custom actions in Microsoft 365 Copilot. It takes an agent specification and a customer's eval set as input, projects each query into a platform-owned capability taxonomy, and outputs per-query decisions (admit, drop, swap, or human review), under the philosophy that a healthy regression set is the minimal set of queries capturing the maximal spread of capability signatures -- distinct combinations of capabilities a query exercises together. Three components instantiate this: a classifier producing per-(query, capability) verdicts via a hybrid of deterministic specification-based extraction and large-language-model (LLM) semantic inference; an Invocation Quality (IQ) rater scoring how thoroughly a query exercises each capability, so a new query sharing a signature with an existing entry can still be recognized as a better test and displace it; and a consolidator comparing incoming queries against the regression set on coverage and quality through a rule-based decision cascade, backed by a conservative curator that only suggests evictions. The mechanism is taxonomy-agnostic and applies to any regression eval-set curation problem with a typed capability taxonomy, including taxonomies that evolve in response to the very evidence the pipeline surfaces. --- ### From AI Technical Debt to Agentic Technical Debt: A Systematic Mapping of Root Causes and Manifestations in Agentic AI Systems Authors: Muhammad Tukur, Hayatullahi B. Adeyemo, Tao Chen, Nour Ali, Anis Zarrad, Marco Agus, Rick Kazman, Rami Bahsoon Published: 2026-08-02 Categories: cs.AI, cs.CR, cs.SE arXiv: https://arxiv.org/abs/2608.01001v1 The emergence of Agentic AI systems, characterized by autonomous reasoning, multi-agent collaboration, tool orchestration, adaptive decision-making, and persistent memory, represents a fundamental shift from traditional AI pipelines to dynamic software ecosystems. While AI Technical Debt (AITD) has been widely studied in machine learning and software engineering, existing models assume static, component-level architectures and fail to capture the dynamic and emergent behaviors of agentic environments. To address this gap, this paper introduces Agentic Technical Debt (AgTD), defined as technical debt that emerges, accumulates, propagates, and amplifies due to the autonomous and collaborative nature of Agentic AI systems. Building on our prior systematic scoping review of 31 AITDs across seven root-cause categories, we employ a theory-informed transformation methodology to reinterpret these debts in Agentic AI through direct transformation, contextual transformation, and manifestation expansion. We present the first systematic mapping of established AITDs to their agentic manifestations, showing how conventional debts evolve into system-level liabilities, including memory inconsistencies, orchestration fragility, cascading failures, and unsafe autonomous decision-making. Our findings show that technical debt extends beyond software artifacts to encompass agent behaviors, coordination mechanisms, and interactions among agents, tools, and execution environments. We further examine its implications for AI Trust, Risk, and Security Management (AI TRiSM), highlighting impacts on trustworthiness, governance, security, operational resilience, and Sustainability Technical Debt. Overall, this work establishes AgTD as a foundational software engineering construct and provides a transformation framework, taxonomy, and research agenda for managing technical debt in autonomous multi-agent AI systems. --- ### Judging Is Not Enumerating: Silent Omissions in LLM-Authored Acceptable Sets Authors: Wenhui Chen, Jianlin Chen, Ziyao Lin, Peiji Long, Chi Man Vong Published: 2026-08-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.01000v1 Language models are increasingly promoted from examinees to examiners: they write the test suites, answer keys, rubrics, and reward functions that define correctness for other systems. We measure the capability that role assumes and find it lacking under the protocol the role is usually deployed with, one-shot greedy authoring with no test-time reasoning. Across four reference constructions - two with complete finite truth, one with a hardened executable reference (HumanEval+/MBPP+), one with an explicitly incomplete lexical reference (WordNet) - models judge whether a candidate belongs far better than they author the set itself. On the incompleteness-proof algorithmic construction the gap is +0.34 to +0.29 F1 over a 24x parameter range and does not close; on executable code, models judging at F1 0.74-0.90 author suites admitting only 19-42% of oracle-correct solutions. A control locates the deficit: asked to emit the predicate rather than its extension, the same models reach F1 about 0.99. The failure is not missing knowledge or an inability to specify, but an inability to materialise the region a specification induces. The dominant error is omission, which resists audit: an over-inclusion is a token a reviewer can challenge, a missing member an absence whose discovery is the authoring problem itself. Models detect planted over-inclusions 6-7x more often than planted omissions, and a production deployment of 43,227 items fails omission-first at 10:1. Wired into RLVR, an authored key costs 1.9 points of accuracy against an exact oracle and 18.5 WordNet-relative (six paired seeds, p=0.031). Gating authored verifiers on a known-correct probe cuts false rejection from 58-92% to at most 5%, but keeps only 5-39% of suites. Repairing them instead, by rewriting each wrong expected value to what a reference execution returns, raises yield 3.3-10.6x across four author families. ---