# Research Papers — 2026-08 · bundle page 173 of 240 > Full-text markdown of papers #8601–8650 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/173 [← Previous bundle](https://aipod.dev/papers/archive/2026-08/page/172.md) [Next bundle →](https://aipod.dev/papers/archive/2026-08/page/174.md) ### Dramarrator: Object-Based Audio Editing for Audio Drama Production from Books Authors: Karim Benharrak, Oriol Nieto, Bryan Wang, Zeyu Jin, Amy Pavel Published: 2026-08-08 Categories: cs.HC, cs.AI, cs.MM arXiv: https://arxiv.org/abs/2608.08349v1 Audio dramas weave dialogue, sound effects, and music into immersive stories. Creators often adapt books into audio dramas, but this process remains labor-intensive, requiring them to interpret source material, author scripts, generate audio assets, and assemble them on a timeline. Because story elements like characters and scenes manifest across many interdependent assets, a single change can ripple into manual updates across the entire project. We present Dramarrator, an audio drama authoring tool built around object-based audio editing, where these story elements are represented as editable objects. Dramarrator extracts these objects from a book, generates linked audio assets (speech, sound effects, and music), and composes a multi-track audio drama. Edits to any object (e.g., a character's voice) automatically propagate to all dependent assets. In a user study with professionals (N=8), Dramarrator significantly lowered task load when creating audio dramas. A listener study (N=300) shows that creator-refined output from Dramarrator approaches the quality of productions made with existing professional tools, and an exploratory study (N=3) suggests object-based editing lowers entry barriers and generalizes beyond audio dramas. --- ### PRISM: A Predictive Protocol for Permutation Optimization via Landscape Diagnostics Authors: Blessings Mambwe Published: 2026-08-08 Categories: cs.LG, cs.AI, math.OC arXiv: https://arxiv.org/abs/2608.08344v1 Permutation optimization arises whenever the components of a system are fixed but their ordering affects performance. We introduce PRISM, a predictive protocol for permutation optimization that measures a fitness landscape before selecting a search strategy. PRISM uses inexpensive landscape diagnostics, including one-step move autocorrelation and fitness-distance correlation, to predict useful mutation operators, identify when structured search is likely to outperform random sampling, and detect regimes in which search provides little advantage. Across synthetic permutation landscapes, neural architecture benchmarks, scientific machine learning pipelines, and large-language-model instruction ordering, the protocol makes testable predictions about search behavior before optimization begins. Exhaustive instruction-ordering experiments reveal substantial performance variation induced solely by permutation, while cross-model experiments show that useful ordering structure can transfer across model families and task difficulty. Additional experiments demonstrate that instruction ordering remains consequential after prompt wording is optimized, indicating that content optimization and ordering optimization are complementary. The results position PRISM not as a universally superior optimizer, but as a framework for determining when permutation search is useful, which representation and operator should be used, and when simpler alternatives are preferable. --- ### Circuit Fine-Tuning for Compute-Efficient Transformer Adaptation Authors: Uri Z. Kialy, Gil Ben-Artzi Published: 2026-08-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.08336v1 Parameter-Efficient Fine-Tuning (PEFT) has become the de facto standard for adapting Vision Transformers (ViTs) to downstream tasks. While parameter count has been the dominant efficiency metric in PEFT, it does not imply \textit{compute efficiency}: parameter-sparse methods can still incur full-model training cost per step, and typically need long schedules to reach peak accuracy. We introduce Circuit Fine-Tuning (CFT), a compute-efficient framework that uses circuit discovery---conventionally used to explain trained models---to select modules for fine-tuning before training. Whereas attribution is conventionally formulated against a trained task head, we formulate it against a near-zero-initialized probe head, which isolates the response of the backbone to the target distribution rather than the preferences of a particular classifier. CFT then fine-tunes only the recovered subgraph. CFT needs no learning-rate warmup and reaches peak accuracy in ${\sim}20$ epochs on average---versus $44$--$96$ for strong PEFT baselines---yielding $2.3$--$6.6\times$ fewer training FLOPs and up to $16\times$ less wall-clock time, while adding zero parameters and no inference operations. Experiments across a standard visual transfer benchmark (VTAB-1k), hierarchical backbones (Swin), domain-shifted medical imaging (CBIS-DDSM), and a vision-language model (Gemma-3 on CUB-200) demonstrate the effectiveness of CFT. Code is available at https://github.com/UriKialy/CFT --- ### StructReward: Efficient Structured Process Rewards for Self-Correcting Multimodal Reasoning Authors: Yifan Li, Ruxin Sun, Tongzhou Zhao Published: 2026-08-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.08326v2 Reinforcement learning with verifiable rewards (RLVR) has emerged as an effective approach for improving multimodal reasoning. However, most existing methods evaluate an entire response using a binary reward based only on final-answer correctness, thereby discarding the supervision available in intermediate reasoning steps. Process reward models offer finer-grained feedback, but they typically rely on separately trained verifiers, costly chain-of-thought annotations, or online judging by large language models (LLMs). In this work, we introduce StructReward, a compute-efficient framework that provides dense reinforcement signals through structured step-level reward alignment. StructReward represents each generated solution as a sequence of reasoning steps and aligns them with process-labeled reference steps using lightweight numerical, symbolic, and lexical matching rules. The aligned labels are aggregated into a dense process reward and combined with final-answer consistency and output-validity rewards through a gated Group Relative Policy Optimization (GRPO) objective. We further recycle policy rollouts into complementary supervision for response comparison and reflective self-correction, rather than discarding them after policy updates. Separately, we use a strong LLM to rewrite sampled correct trajectories into reflection-oriented training instances, further strengthening the policy's ability to evaluate and refine its reasoning. Since reward computation is performed online without an additional learned verifier or external LLM judge, StructReward substantially reduces the computational overhead of multimodal reinforcement learning. Experimental results show that structured process supervision and rollout recycling provide an efficient path toward self-improving multimodal reasoning. --- ### SDF-Aware Weighting: Adaptive Eikonal Regularisation for Three-Dimensional Level-Set Physics-Informed Neural Networks Authors: Muhammad Akbar Khan Published: 2026-08-08 Categories: physics.flu-dyn, cs.LG, physics.comp-ph arXiv: https://arxiv.org/abs/2608.08322v2 Adaptive loss-balancing schemes for physics-informed neural networks rest on a premise that every residual should be driven to zero. For level-set advection with an eikonal regulariser that premise fails: the eikonal term penalises deviation of $\lVert\nablaφ\rVert$ from unity, a property transport preserves only under rigid motion; where the exact solution departs from a signed-distance function the eikonal residual of the correct answer is nonzero, and driving it to zero moves the network away from that answer. We show that standard gradient-norm balancing fails in exactly this way, its weight remaining near its initial value throughout training on benchmarks where the property is violated, and we introduce SDF-Aware Weighting (SAW), which combines a residual-quantile gate with a gradient-norm ratio so that points exhibiting legitimate departure are excluded before the surviving term is scaled. Across four three-dimensional benchmarks SAW selects an eikonal weight within an order of magnitude of the value located by an eighteen-run manual sweep, spanning four decades from $10^{-1}$ to $10^{-5}$ with a single fixed configuration. On the slotted sphere, where the initial field is non-differentiable at reentrant edges, SAW attains a lower error than any weight in that sweep. Two smooth rigid benchmarks serve as controls: SAW is worse there, as expected when its premise does not hold. An ablation with the gate disabled shows the slot is nearly entirely filled while the relative $L_2$ error reads $1.06\%$, indistinguishable from a field that never represented the slot. We give a feature-restricted measure that separates the two cases. --- ### Spatial Heterogeneity-Aware Multi-Hazard Susceptibility and Risk Mapping at Regional Scale Authors: Aswathi Mundayatt, Siddharth Anil, Hitanshu Seth, Jaya Sreevalsan-Nair Published: 2026-08-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.08321v1 Floods and landslides often co-occur, but their relationships with environmental controls vary spatially. This study develops a spatial heterogeneity-aware framework for flood-landslide susceptibility and relative-risk mapping in Kerala, India, and Nepal. It combines 15 km x 15 km grid cells with region-specific contextual zones and compares proximity-gated cross-zone training (S1) and ecology-gated zone-constrained training (S2). S1 permits geographically nearby models to be assigned across contextual boundaries, whereas S2 restricts model development and assignment to the same zone. Random Forest models for each hazard use strategy-specific predictor sets and are evaluated on spatially held-out test samples. Susceptibility surfaces are integrated with CRITIC-weighted exposure and vulnerability indices to produce hazard-specific and nine-class bivariate relative-risk maps. S1 achieved higher mean accuracy, precision, recall, F1-score, AUC-ROC, and PR-AUC for both hazards and regions. The largest difference occurred for Nepal flood susceptibility, where AUC-ROC increased from 0.728 under S2 to 0.886 under S1 and PR-AUC from 0.512 to 0.823. S2 produced lower Brier scores for both Nepal hazards and retained zone-specific differences in predictor selection, SHAP rankings, and response patterns, particularly in Kerala. Both strategies reproduced flood-prone lowland and landslide-prone upland patterns but differed in susceptibility and risk classes. Bivariate risk-map agreement was 0.521 in Kerala and 0.711 in Nepal, with allocation disagreement exceeding quantity disagreement in all S1-S2 comparisons. Susceptibility-to-risk correspondence remained below 0.350, showing that exposure and vulnerability changed priority locations. Overall, cross-zone learning strengthens regional discrimination, while zone-constrained learning preserves environmental differences, supporting their integration. --- ### A continually expandable foundation model for brain MRI Authors: Michail Mamalakis, Carmen Jimenez-Mesa, Yonghao Li, Hao Chen, Chao Li, Antonios Mamalakis, John Suckling, Richard Bethlehem, Stephen J. Price, Richard J. Gilbertson, Pietro Lio Published: 2026-08-08 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2608.08319v3 Brain magnetic resonance imaging (MRI) is central to neuroscience and clinical assessment, but models are commonly developed for individual diseases, populations or imaging protocols. Foundation models promise more general representations, yet they are usually pretrained once and can lose earlier capabilities when updated with new data. Here we show that Alcmaeon, a three-dimensional brain MRI foundation model pretrained without manual labels on more than 425,000 volumes and derived imaging maps, can be expanded sequentially across clinical domains. Alcmaeon combines volumetric encoding and latent diffusion generation with Graph-Blueprint Pruning (GBP), which protects network modules important to earlier domains while leaving the remaining capacity trainable. Across expansion from healthy ageing and neurodegeneration to developmental, psychiatric and tumour imaging, GBP showed less forgetting than sequential adaptation and elastic weight consolidation across voxel-level reconstruction measures, with its largest advantage after adaptation to tumour imaging. The blueprints provided an inspectable record of how model capacity was protected and reused. Representations from different model levels supported image synthesis, disease classification, survival modelling and postoperative prediction, although no single representation was optimal for every task. These findings provide a route towards brain MRI foundation models that can grow with emerging data while retaining earlier capabilities. --- ### The Neural Division of Labor: Biologically-Inspired Modular Architectures for Robust Neuromorphic Computing Authors: Maksim Bazhenov, Serafim Grubas, Vakhtang Putkaradze Published: 2026-08-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.08317v1 Biological neural systems achieve high efficiency and robustness through compartmentalized architectures. In contrast, modern artificial neural networks rely on globally entangled structures, which obscure decision logic and suffer from catastrophic forgetting. Here, we report a Decomposable Spiking Neural Network (D-SNN) that eliminates global synaptic entanglement by structurally isolating classification pathways into independent experts. Optimized via a bio-inspired push-pull loss function, the D-SNN achieves competitive accuracies on MNIST, Fashion-MNIST, and CIFAR-10/100 benchmarks. This modular approach matches the performance of fully dense networks while utilizing an order of magnitude fewer parameters. In addition, our networks operate with up to several orders of magnitude lower firing rates and fewer synaptic operations. Furthermore, physically severing connections between experts provides inherent protection against catastrophic forgetting during sequential learning. Crucially, these isolated pathways generate auditable neural signals, increasing decision transparency. This biomimetic, verifiable architecture establishes an efficient foundation for deploying deterministic neuromorphic intelligence in resource-constrained edge environments. --- ### Your VLM Already Knows When: Training-Free Temporal Grounding by Asking Yes or No Authors: Ji Huang, Barry Devereux, Hui Wang Published: 2026-08-08 Categories: cs.CV, cs.MM arXiv: https://arxiv.org/abs/2608.08315v1 Multimodal LLMs that recognise events reliably still fail to say when they happen. Prompted for timestamps, strong VLMs reach as little as $3.8\%$ R@0.5 on Charades-STA, and $77$ to $80\%$ of their wrong predictions carry low output entropy: the models are confidently wrong, and entropy-based error detection stays below a random classifier. We show that this failure lives in the task interface, not in perception. Holding the weights fixed, replacing timestamp regression with a coarse-to-fine scan of binary questions, whose first-token probabilities are consumed only as a ranking, raises R@0.5 by $28$ to $50$ points across four frozen backbones. The residual failures decompose into two measurable axes: a perception axis that moves with the backbone, and a geometry axis that is analytically predictable from the ratio of the output-window and event widths. FV-Action, the training-free method built on this analysis, reaches $56.8\%$ R@0.5 on Charades-STA, above the same backbone's native grounding pipeline and the strongest training-free result on this benchmark; it surpasses every TVG-trained model evaluated zero-shot on TACoS, and improves over direct prediction on ActivityNet Captions and QVHighlights, with no temporal supervision at any stage. --- ### Ouroboros: A Self-Developing Frontier Coding Agent with Reviewed Core Evolution Authors: Anton Razzhigaev, Andrei Gritsaev, Andrei Kaznacheev, Nikita Dragunov, Roman Yampolskiy, Andrei Kuznetsov Published: 2026-08-08 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2608.08311v3 We present Ouroboros, a self-developing agent harness whose tools, prompts, context assembly, and core implementation improve through reviewed commits that become the runtime for later work. Core evolution proceeds in two modes. In recursive free evolution, improvement is itself a task, and completing one evolution cycle can schedule the next. In experience-driven core evolution, ordinary work and social interaction expose bugs, rough edges, and inefficient context construction that lead to reviewed structural changes. On Terminal-Bench 2.1, an Opus 5 run scores 86.74%, the best result reported on the benchmark. On OSWorld-Verified, an Opus 5 run reaches 90.69%, exceeding the best previously reported score. A five-rollout CL-Bench campaign achieves a normalized reward of 0.2301, setting a new state of the art. Hope is the longest-running publicly documented Ouroboros deployment. It is a 161-day living agent experiment in free evolution under governed human communication across seven surfaces. Human interaction surfaces faults and generates proposals, but the agent decides which changes to pursue. Because a self-developing agent may rewrite its own code and select new model APIs, operational safety becomes a primary design problem: guardrails must remain authoritative under evolutionary and public social pressure. Benchmark campaigns use frozen system snapshots, while Hope continues live evolution on a separate lineage. --- ### Three Necessary Principles for Self-Supervised Visual Representation Learning Authors: Nikos Giakoumoglou, Paschalis Giakoumoglou, Tania Stathaki Published: 2026-08-08 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.08309v2 We argue that learning visual representations without labels requires a training signal jointly complete across three non-overlapping objectives: semantic invariance across augmented views, patch-level spatial prediction, and representational non-degeneracy. We formalize these as the observation, prediction, and regularization principles and prove (i) that combining observation and prediction without regularization admits the constant encoder as a global minimizer under negative-free alignment; (ii) that the two objectives are gradient-complementary and structurally non-conflicting at the encoder output; and (iii) that the momentum encoder converges to the same fixed point as the online encoder and provides no collapse guarantee at convergence. Contrastive alignment provides only self-limiting collapse resistance, formalized via an explicit gradient-decay argument. Dropping prediction withholds the spatial training signal by construction; dropping observation forfeits cross-view semantic invariance by construction; at the scale we study, no pair substitutes for the third. Every major self-supervised method is a special case of a single unified energy decomposition. We pair every theoretical claim with a controlled experiment, including a patch-retrieval evaluation for the spatial consequence of prediction. --- ### Open-World Semantic Segmentation with Sensitivity Modeling Authors: Anastasios Romanos Varvarigos, Nikos Giakoumoglou, Tania Stathaki Published: 2026-08-08 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.08308v2 Modern vision systems must operate in "open-world" settings, where models must recognize known categories and detect unseen or anomalous content. Conventional semantic segmentation models operate under a "closed-world" assumption, often producing overconfident misclassifications on novel content. We address open-world semantic segmentation, the joint task of segmenting known classes while detecting and grouping novel or anomalous content without additional supervision, by extending a dual-decoder baseline with a third, complementary decoder within a unified encoder-decoder design. The first decoder performs closed-set segmentation using Gaussian prototypes for known categories. The second uses contrastive feature learning to isolate unknown regions in embedding space. The third, our key contribution, is a sensitivity decoder that captures fine-grained texture irregularities and activation instabilities indicative of semantic uncertainty, which neither semantic prototypes nor contrastive norms can reliably detect. The three decoders provide genuinely complementary signals: class-level OOD distance in logit space, global feature energy in embedding space, and local activation instability across encoder scales. Experiments on Cityscapes and BDD-Anomaly show that our method improves anomaly segmentation and novel-class discovery while maintaining competitive closed-set accuracy, with gains of +2.4% AUROC and a 2.5 pp. reduction in FPR@95TPR on BDD-Anomaly over the baseline. --- ### Frequency-Domain Dual-Branch Fusion for Medical Visual Question Answering Authors: Yusra Tariq, Rakesh Chandra Joshi Published: 2026-08-08 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.08307v1 Medical Visual Question Answering (VQA) requires aligning subtle visual evidence, including lesion texture, boundary sharpness, and diffuse density changes, with clinical language. Existing multimodal fusion approaches operating in the spatial domain may not fully exploit complementary frequency information present in visual and textual representations. We introduce a dual-branch frequency-domain fusion module that conditions spectral filtering on the input question, enabling adaptive selection of global low-frequency structure and fine-grained high-frequency detail before reconstructing the spatial representation for answer generation. To provide a richer spectrum for filtering, we extract complementary features from early texture-sensitive and final semantic layers of a frozen BiomedCLIP encoder and align both with the question representation using a symmetric InfoNCE objective prior to staged joint training with a BioBART decoder. We pretrain the proposed model on PMC-VQA and fine-tune it on the VQA-RAD and SLAKE benchmarks, demonstrating that frequency-aware multimodal fusion improves medical VQA performance while maintaining a lightweight and efficient architecture. --- ### Query-Only Backdoor Attacks on Self-Evolving Skills via Trajectory Poisoning Authors: Yuyang Luo, Haoran Wang, Kai Shu Published: 2026-08-08 Categories: cs.AI, cs.MA arXiv: https://arxiv.org/abs/2608.08303v1 Agentic skills improve large language model (LLM) agents by encoding reusable procedures for complex tasks. However, manually authored skills often adapt poorly to long-horizon tasks and changing environments. To address the limitation, self-evolving skill systems have been developed to automatically construct and update skills from execution trajectories, shifting skill acquisition from external marketplaces to a trusted evolution pipeline. By replacing external skill acquisition with trusted internal construction, self-evolving skill systems reduce exposure to skill injection attacks that rely on direct skill manipulation. However, this skill evolution pipeline may introduce a new attack surface in which an attacker can indirectly steer skill evolution by inducing compromised trajectories through agent interactions. To demonstrate the threat, we propose Trajectory Backdoor Attack (TBA), a query-only attack that steers a trusted skill-evolution pipeline toward producing a backdoored skill. Specifically, we craft attacker-submitted queries to lead the agent to perform the target action and explicitly state the corresponding activation condition in the trajectory. We repeat the same condition-action pattern across diverse triggered tasks, while leaving clean queries unchanged, encouraging the evolver to consolidate the pattern as a reusable trigger-dependent rule into the evolved skill. Experiments on three benchmarks across two skill-evolution systems using four open- and closed-source backbone models demonstrate that TBA reliably implants conditional backdoors while preserving clean-task utility, matching or even surpassing direct skill injection. The results reveal a critical vulnerability in trajectory-driven skill evolution. --- ### Machine-Learning-Based Diagnostic Framework for Passive Ultrasonic Detection of Railway Wheel Defects Authors: Aashish Shaju, Steve Southward, Mehdi Ahmadian Published: 2026-08-08 Categories: cs.LG, cs.RO, eess.SP arXiv: https://arxiv.org/abs/2608.08301v1 Reliable identification of railway wheel defects is important for safety and maintenance. This study develops a machine-learning-based diagnostic framework for multi-class defect identification using passive air-coupled ultrasonic acoustic emission signals. Data were collected from eleven full-scale railway wheelsets representing nine health states. Time- and frequency-domain features were evaluated using Kruskal-Wallis statistical testing and mutual-information analysis to identify the most discriminative indicators. A Random Forest classifier was then trained using the selected features with stratified 5-fold cross-validation. The model achieved a balanced accuracy of approximately 0.66 and a Macro-F1 score of 0.65 across the nine classes. Decay rate, kurtosis, skewness, and envelope low-frequency power emerged as the most influential features, while a compact subset of features retained most of the classification performance. The results demonstrate the feasibility of combining passive ultrasonic sensing, statistical feature selection, and supervised machine learning for non-contact railway wheel defect classification and provide a foundation for future field-deployable inspection systems. --- ### Mitigating Over-Personalization in LLMs via Structured Memory Authors: Hakeem Hannoon, Andrew Zhao, Mihir Narayan, Sharvin Goyal, Ivaxi Sheth Published: 2026-08-08 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.08300v1 Conversational assistants increasingly rely on persistent long-term memory to personalize responses across sessions. However, when stored user information is reintroduced into the model context, it can also influence responses in inappropriate or unrelated settings. We study two such failure modes in memory-augmented LLMs: cross-domain leakage, where memories from one life domain affect responses in another, and memory-induced sycophancy, where stored user beliefs make models more likely to agree with the user rather than respond truthfully. We apply a simple inference-time modification to how memories are presented to the model, without changing the model or the memory contents. Across seven models on PersistBench, we compare the commonly used all-in context format, where memories are injected as an unstructured list, with structured formats that partition memories by domain. This simple modification consistently reduces cross-domain leakage while preserving utility, with our strongest method reducing leakage by $8.8\%$ on average relative to the baseline. --- ### A Controlled Study of Feature-Based Knowledge Distillation Across Student Designs Authors: Abhinand Balachandran, Praveen Prashant Published: 2026-08-08 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2608.08294v1 Knowledge distillation trains a smaller student to match the outputs of a larger teacher. Feature-based methods also align intermediate representations, but this extra constraint may affect students differently. We study this question on CIFAR-100 using a ResNet-50 teacher, a width-controlled CustomResNet family and MobileNetV2 as a cross-design comparison. For each student, we evaluate each feature method against a matched logit-KD run using the same teacher, optimizer settings, training schedule and seed. We repeat the main comparisons across multiple seeds. Logit KD improved every tested student over its scratch baseline. Attention Transfer showed no clear relationship with size inside the CustomResNet family, but its average effect was negative for that family and positive for MobileNetV2. FitNets was below logit KD in all 15 paired runs. Within the constant-depth width sweep, its gap increased for wider students, although the different-depth w=48 student did not follow this trend. Finally, the same auxiliary coefficient produced different gradient scales across students, showing that a fixed coefficient does not create a uniform training condition. --- ### Test-Time Prototype Adaptation for Open-Vocabulary Semantic Segmentation Authors: Haozhe Wang, Jintao Cheng, Weibin Li, Xiaoyu Tang Published: 2026-08-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.08290v1 Open-vocabulary semantic segmentation (OVSS) repurposes a pretrained CLIP encoder for dense prediction without additional labeled supervision. Existing methods improve CLIP's spatial behavior either by redesigning its internal attention or by injecting features from auxiliary vision foundation models; both require access to the host's internal computation and are tailored to its specific forward pass. In this work, we propose Test-time Prototype Adaptation (TPA), a training-free plug-in that operates at the output level, leaving the host's forward pass and weights unmodified. By leveraging a lightweight transductive adaptation phase, TPA identifies confident anchor patches from the host's own output predictions on a small pool of unlabeled deployment-domain images, and aggregates their frozen DINO features into per-class prototypes; at inference, a single cosine similarity lookup against this frozen bank provides an auxiliary score fused linearly with the host's logits. TPA composes with five representative OVSS hosts spanning attention-redesign and VFM-injection designs, across three CLIP backbones, eight benchmarks, and multiple internal VFM choices. Under a single set of hyper-parameters and without per-host tuning or parameter updates, TPA consistently improves segmentation accuracy, with as few as approximately 10% of unlabeled deployment-domain images sufficing for effective bank construction on most benchmarks. --- ### Causal State-Space Model for Causal Inference: Estimating Longitudinal Individual Treatment Effects Authors: Abisoye Abidakun, Mingjun Zhong, Georgios Leontidis Published: 2026-08-08 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2608.08288v2 Estimating counterfactual outcomes over time from longitudinal observational data is central to clinical decision support. Existing methods rely on domain confusion -- adversarial training that renders representations invariant to treatment assignment -- yet this invariance creates a mutual information conflict: it suppresses treatment-correlated covariate signals necessary for accurate outcome prediction. We formalise this tension via a Jensen-Shannon divergence bound on counterfactual prediction error and develop two complementary models. CSSD (Causal State-Space model with Direct decoder) adapts selective State Space Models with a parallel multi-step decoder that eliminates accumulated rollout error by producing all prediction horizons simultaneously in a single forward pass. CSSPD (Causal State-Space model with Predictive regularisation and Direct decoder) augments CSSD with Contrastive Predictive Coding and Local Information Maximisation to reinforce temporal predictability in the balancing representation and recover local covariate information destroyed by domain confusion. On MIMIC-III, CSSPD achieves lower counterfactual RMSE than the Causal Transformer at every horizon tau >= 2 at O(T) encoder cost, with gains from 0.02 (2-step) to 0.07 (6-step). On Cancer Simulation across confounding strengths gamma in {0,1,2,3,4}, CSSPD outperforms CT at gamma <= 3 (margins 25.9%--37.0%), and CSSD achieves the lowest overall average RMSE (12.7% reduction over CT), confirming the MI conflict analysis. To our knowledge, this is the first work to formalise the balancing-prediction MI conflict and propose a structured resolution through complementary predictive and information-theoretic training objectives. --- ### What Irregularity Costs: CUDA C++, Rust, and Triton on a Hash-Blocked GPU Workload Authors: Petr Korolev Published: 2026-08-08 Categories: cs.CV, cs.DC, cs.PF, cs.PL arXiv: https://arxiv.org/abs/2608.08287v1 GPU language comparisons are almost always run on tiled dense linear algebra, where every toolchain is good and the differences are small. We implement the same hash-blocked TSDF fusion kernel in CUDA C++, in Rust through NVIDIA's cuda-oxide, and in Triton, and measure it on a workload with the opposite character: an open-addressed hash table with compare-exchange insertion, data-dependent per-lane probe depth, and contended scatter. The result is a split. On the regular stage, which walks a truncation band and accumulates, all three languages land within a small factor of each other. On the irregular stage, which probes and inserts, Rust stays close to hand-written CUDA C++ while Triton is more than an order of magnitude slower. Language choice is nearly free on the work that is usually benchmarked and expensive on the work that is not. We attribute both gaps to specific things the languages cannot express, not to ratios. Triton's cost follows from a probe loop that must run to a compile-time bound and from tl.atomic_cas taking no mask, which forces a scratch structure with no counterpart in CUDA. Rust's cost was invisible in every instruction count: its kernel issues fewer instructions, fewer compare-exchanges and fewer registers at identical occupancy, yet was slower. Hardware counters located it in L1 residency. A GPU-scope atomic load must be coherent across SMs, no NVIDIA L1 is, so the type-correct way to read a shared location bypasses the cache on every access. Triton's bounded probe is also a correctness problem for fusion: at load factors an ordinary depth trajectory reaches, it silently discards blocks and the reconstruction loses patches of surface with nothing reported. We also report a defect found and fixed in cuda-oxide itself, now merged upstream: its scoped atomic load and store could not be called at all in the build mode that produces real kernels. --- ### Ego-OSCAR: Egocentric Open source Stereo CAptuRe System Authors: Gunjan Paul, Senthil Palanisamy, Satpal Singh Rathore, Pratyush Kumar Patnaik, Shubhanshu Khatana, Abhishek Anand Published: 2026-08-08 Categories: cs.CV, cs.AR, cs.RO arXiv: https://arxiv.org/abs/2608.08285v2 We present Ego-OSCAR, an open-hardware, low-cost, head-mounted stereo-inertial capture device for egocentric data collection in the wild. EgoOSCAR pairs a hardware-synchronized global-shutter stereo camera with a 6- axis IMU, an embedded Linux SBC for on-device video encoding, and a realtime microcontroller for user feedback and watchdog functions. The complete bill of materials is under USD 200 per unit, using only commercially available components and 3D-printed parts. Alongside the device, we release a complete software stack (hardware-accelerated recording pipeline, IMU sampling daemon, time-synchronization tooling, and watchdog firmware) and roughly 550 hours of egocentric stereo video per camera with synchronized IMU, collected by a distributed contributor network across everyday indoor environments. The release is annotated rather than raw: free-form action captions cover essentially the entire recorded timeline with an open vocabulary, and per-frame 3D hand reconstructions ship alongside per-session stereo calibration. Ego-OSCAR does not aim to match the per-unit fidelity of research-grade systems such as Project Aria; it aims to be the cheapest defensible substrate for crowdsourced egocentric capture, and to lower the activation energy for any team that wants to collect egocentric data at scale. All hardware designs, software, and the dataset are open-sourced --- ### Fair on the Surface? Benchmarking Hidden-Output Fairness Gaps in LLM Recommenders Authors: Chan Aristella Lu, Arya Fayyazi, Junhao Zhang, Saeid Shokoufa, Yue Xing, Zhen Xiang, Kyu Hyung Lee, Mehdi Kamal, Massoud Pedram Published: 2026-08-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.08284v1 Fairness audits for LLM-based recommenders have largely focused on observable outputs, implicitly assuming that stable recommendations reflect stable internal processing. We challenge this assumption with FairGap, the first benchmark to jointly evaluate recommendation fairness at two levels: observable output shift (OBS) and hidden representation shift (IBS), measured through controlled counterfactual identity probes across gender, age, and race. Their relationship is summarized via Representation-Output Alignment (ROA), with quadrant diagnostics for identifying user-level hidden-output mismatch. Applied to six open-weight LLM families across three domains, FairGap reveals pervasive hidden-output decoupling: ROA rarely exceeds 0.22, and a non-negligible user population shows stable outputs despite substantial internal shifts, a mode that output-only audits cannot detect by design. Further, activation steering that reduces IBS by up to 8x simultaneously worsens OBS, demonstrating a fundamental tension between internal and output-level fairness that existing frameworks are unequipped to diagnose. --- ### Do Evaluation Metrics Detect Errors in Classical Chinese to English Translations? Authors: Osvaldo Quinjica, Eric Bennett, Xinchen Yang, Andrew Schonebaum, Marine Carpuat Published: 2026-08-08 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.08283v2 Although large language models can translate some historical languages surprisingly well, their usefulness in digital humanities workflows is limited by the lack of reliable evaluation. We investigate whether existing automatic evaluation metrics developed for modern languages are reliable in this setting, using translation from Classical Chinese to English as a test case. We introduce a diagnostic framework based on minimal pairs capturing error types salient in scholarly use, probing both reference-based and reference-free metrics for error sensitivity and tolerance to valid variation. We find that all metrics exhibit blind spots, however MetricX24 performs best overall. Our findings highlight the need for more robust and interpretable metrics for historically and culturally distinct translation settings. --- ### Stateful CARS: Exact Cross-History Reuse for Policy-Constrained LLM Agents Authors: Ibne Farabi Shihab, Md Najmus Swaqeeb, Abu Sa-Adat Mohamed Moon-Im Al Ahsan Published: 2026-08-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.08282v1 Tool-using language-model agents face constraints whose meaning changes with observations and prior actions. We study exact sampling from the model distribution conditioned on a hard stateful validator while reusing invalidity certificates across histories. Stateful CARS freezes a bank of sound state--continuation schemas within each attempt and removes every trajectory containing a certified continuation at a matching abstract state. An exact residual Doob transform samples from the resulting proposal. We give a checkable future-validity bisimulation condition, prove schema soundness, adaptive exactness, i.i.d.\ outputs, almost-sure termination, monotone acceptance, and compression invariance, and characterize computation by the number of reachable full-history product states. This number can be exponential for a history-dependent language model; the evaluated method therefore makes no generic finite-trie scalability claim. On enumerable workflows, its analytic law matches the valid conditional to $10^{-16}$ at validity probability $6\times10^{-8}$, whereas state-aware local decoding can be $0.97$ away. A matched comparison is negative: observation-keyed official CARS is cheaper in sampler steps (root/Stateful ratio $0.942$ $(0.934,0.951)$), and the Qwen comparison is null ($0.99$ $(0.90,1.08)$). Cross-history transfer helps only in an internal matched-key ablation ($1.27\times$). Thus the evidence supports exact schema-induced conditioning, not a systems advantage over CARS. --- ### Exploring LLM Capabilities for Situational Understanding and COLREG compliance on real-world maritime navigation scenarios Authors: Julius Wirbel, P. Nicholas Hansen, Line K. H. Clemmensen, Roberto Galeazzi Published: 2026-08-08 Categories: cs.AI, cs.RO arXiv: https://arxiv.org/abs/2608.08281v1 Recently, Large Language Models (LLMs) have shown considerable capability for situational understanding, reasoning, and decision making in different domains, most notable in the automotive sector. Therefore, we explore current state-of-the-art LLMs as a tool for maritime navigation, which includes both codified rules in the Collision Regulations (COLREGs) and uncodified best practices summarized in the concept of ``Good Seamanship''. We construct a dataset consisting of 50 diverse, real-world navigation scenarios from AIS data, label scenarios with applicable COLREG rules, recommended actions, and the reasoning for the action. We explore a variety of different LLM architectures and sizes to determine their understanding of maritime navigation tasks as well as evaluate their reasoning capabilities in this domain. The results obtained indicate that the maritime navigation task remains difficult to solve without fine-tuning, even for larger online models. --- ### Action- and Language-Conditioned Video Assessment for Embodied Control Authors: Hwanhee Kim, Jaehyun Jang, Seungmin Cha, Hyeonseo Yun, Donghoon Lee, Chang D. Yoo Published: 2026-08-08 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2608.08273v2 Vision-based embodied agents executing multi-step natural language instructions require feedback mechanisms that assess task progress over complete trajectories. Conventional approaches based on final-frame matching or continuous embedding similarity may overlook intermediate transitions that are necessary for determining whether an instruction has been completed. We propose ALVA (Action- and Language-Conditioned Video Assessment), a trajectory evaluator that conditions its assessment on visual observations, the executed action sequence, and the natural language instruction. The method uses a pre-trained vision-language model (VLM) in two stages: it first summarizes frame-to-frame visual transitions conditioned on the executed actions and then assesses the generated summary with respect to the instruction to produce a discrete trajectory-level progress score. In simulated 3D household environments, ALVA exhibits a conservative assessment pattern with near-zero false-positive rates. When used as terminal feedback for closed-loop policy optimization, it provides more effective feedback than the evaluated static image and embedding-based visual baselines and reduces the performance gap to a ground-truth oracle. These results support action- and language-conditioned video assessment as an interpretable feedback mechanism for the evaluated simulated embodied-control tasks. --- ### Learning under Opponent Unawareness in Linear-Quadratic Stochastic Games Authors: Dantong Chu, Xuefeng Gao, Yufei Zhang Published: 2026-08-08 Categories: math.OC, cs.GT, cs.LG, econ.TH arXiv: https://arxiv.org/abs/2608.08268v1 As firms increasingly deploy machine learning for strategic decision-making, understanding algorithmic interactions has become central to operations research and economics. This paper studies learning in infinite-horizon, nonzero-sum linear-quadratic stochastic games under a radically uncoupled information structure, where players are either unaware of opponents or strategically oblivious, observing only a common state and their own action history. Under this minimal information, we analyze an asynchronous decentralized learning process in which each player independently runs a single-agent $ε$-greedy iterated least-squares algorithm. We prove that, despite being unable to identify the system parameters, players' learning dynamics converge almost surely to the complete-information Nash equilibrium and characterize the convergence rate. We then apply the framework to a dynamic Cournot competition with sticky prices. Numerical experiments validate the theoretical results and show that learning under limited information reduces firm profits under both low and high price stickiness, while total surplus declines and market concentration increases when price stickiness is high. Publicly revealing aggregate market output substantially accelerates convergence and mitigates these welfare losses. --- ### On the Robustness of LLMs' Internal Representation of Code Correctness Authors: Francisco Ribeiro, Sohaila Abdulsattar, Renata Gonzalez, Mahmoud Kassem, Sarah Nadi Published: 2026-08-08 Categories: cs.SE, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.08266v1 Code generated by modern language models often reads naturally. Yet, it also often fails to implement what was asked. This should be no surprise, as research shows the models' own confidence signals are poorly calibrated with actual correctness. A promising way to assess correctness looks inside the model: by contrasting the hidden states of correct and incorrect programs, recent work captured an internal signal of code correctness that is able to judge candidate solutions better than the model's token-level or stated confidence, with no test execution. However, this signal was captured under one particular way, leaving open an important question: whether it reflects a robust property of the model or an artifact of that choice. We study this question systematically, varying how the signal is extracted from the model internals. Besides this, we also ask if the signal's quality is limited by the data used to extract it, by constructing program pairs that differ only in the fault that makes them incorrect. Our results show that no single configuration is best, and that isolating the fault does not help. --- ### Opportunity Is Not Realizability: Selection-Valid Diagnostics for Multi-LLM Routing Authors: Ibne Farabi Shihab, Abu Sa-Adat Mohamed Moon-Im Al Ahsan, Md Najmus Swaqeeb Published: 2026-08-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.08265v1 Oracle routing measures how much a pool of language models could gain from per-query selection, but the diagnostic has two flaws: testing against a best fixed model selected on the same examples invalidates paired inference, and a full-information oracle sees outcomes no deployable router observes. We separate three estimands (outcome-oracle opportunity, the Bayes-optimal gain from a declared pre-answer signal, and the held-out gain of a learned router) and prove selection-valid confidence intervals that survive choosing the best fixed model or the best member of a router family, a signal-information sandwich, and a $(1-1/e)$ greedy guarantee for building compact pools from submodular complementary coverage. On eight checkpoints from six families over four benchmarks, selection-valid intervals certify a population oracle gap of $9.7$--$30.7$ points on every task, yet the strongest deployable prompt router recovers only $7.5$--$14.4\%$ of it, and the simultaneous interval for the best of eleven tested policies has lower limit zero throughout. The realizable share of oracle opportunity is small and certifiable: strong routers beat the best fixed model, and most of the gap remains. --- ### OBLIVION: Workflow-Level Operational Skill Unlearning for Deployed Agents Authors: Zhengyang Shan, Xu Qian, Jiayun Xin, Kun Li, Yue Zhang, Minghui Xu Published: 2026-08-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.08264v1 Large language model agents are becoming operational interfaces to files, memories, registries, and external tools. This deployment shift creates a new skill revocation problem: after a skill is removed from an explicit registry, an agent may still reconstruct it from residual carriers such as archives, transcripts, schemas, or memory entries. We study this problem as operational skill unlearning, where the goal is not parameter-level forgetting, but preventing a deployed agent from rebuilding a revoked skill through primitive tools. We introduce OBLIVION, a controlled benchmark and defense harness for revoked-skill resurrection. OBLIVION models each episode as a source-to-sink workflow, applies Cross-Surface Coherent Erasure to reduce residual carriers, and uses frozen workflow remediation near dangerous sinks. On the locked 88 attack episodes, the no-defense arm reaches formal attack success rate 1.0. OBLIVION reduces the rate to 0.114 and impact-weighted exposure to 0.115 while keeping locked utility at 1.0 and benign block rate at 0. In a separate skill-attack-derived sandbox, OBLIVION reduces attack success from 1.0 to 0.2 and impact-weighted exposure from 1.0 to 0.213 while preserving all utility controls. These results support workflow-level evaluation beyond checking explicit skill entries. --- ### AraSSM: A bidirectional state-space encoder for Arabic masked language modeling Authors: Ahmed Amine Aliane, Hassina Aliane, Nasredine Semmar Published: 2026-08-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.08256v1 Pretrained Transformer encoders such as AraBERT, MARBERT, and CAMeLBERT have become the standard backbone for Arabic natural language understanding, but their self-attention mechanism scales quadratically with sequence length, which limits efficiency on long documents. Mamba, a selective state-space model (SSM), offers linear-time sequence modeling as a competitive alternative to attention, yet no dedicated bidirectional Mamba encoder pretrained specifically for Arabic currently exists. We introduce AraSSM, a bidirectional Mamba encoder pretrained via masked language modeling on a corpus combining Arabic Wikipedia and CulturaX text, trained end-to-end on four consumer-grade NVIDIA RTX 2080Ti GPUs (11GB) over approximately ten days. We evaluate AraSSM by fine-tuning on four established Arabic NLU benchmarks covering sentiment classification (HARD), named entity recognition (ANERcorp), extractive question answering (ARCD), and natural language inference (XNLI-ar), following the per-task evaluation protocol introduced by AraBERT, and report results as mean +/- standard deviation across three fine-tuning seeds. AraSSM matches or exceeds published base-sized Transformer baselines on sentiment classification (96.37 +/- 0.03% accuracy on HARD), is competitive on extractive QA (32.19 +/- 1.07 EM, 63.79 +/- 0.25 F1 on ARCD) and named entity recognition (81.54 +/- 0.30 entity-level F1 on ANERcorp), and trails the base-sized Transformer range on natural language inference (72.83 +/- 0.07% accuracy on XNLI-ar), despite being trained entirely from scratch on consumer hardware rather than large-scale accelerator clusters. --- ### Learning from Environmental Feedback: Credit Assignment across Multiple Timescales for Agentic Reinforcement Learning Authors: Yifu Huo, Shunjie Xing, Chenglong Wang, Peinan Feng, Qiaozhi He, Yan Ding, Anxiang Ma, Yuxin Gao, Tongran Liu, Tong Xiao, Jingbo Zhu Published: 2026-08-08 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2608.08255v1 Agentic reinforcement learning (RL) often suffers from delayed and sparse rewards in real-world environments. A promising solution to this challenge is credit assignment, which aims to decompose trajectory-level rewards and provide more fine-grained supervision for intermediate decisions. However, existing credit assignment approaches ignore the rich process information naturally generated during environment interaction, e.g., interaction history. We argue that such information provides valuable supervision for identifying the contribution of individual actions. To this end, we propose Environmental Feedback-based Credit Assignment (EFCA), a multi-timescale credit assignment approach for long-horizon agentic RL. EFCA complements the long-term outcome signal with two environment-grounded process signals: a short-term feedback signal that captures the immediate effect of the current action and a medium-term state-history signal that identifies ineffective patterns from recent interactions. Both signals are directly extracted from environment feedback and integrated through a return reweighting mechanism. Experiments on ALFWorld and WebShop demonstrate that EFCA consistently improves both task success and task quality over strong baselines, highlighting the effectiveness of environment-grounded multi-timescale credit assignment for long-horizon agentic RL. --- ### Your Prompt Is Not the Only Prompt: How Much Do LLMs Weight Structured-Output Schema Descriptions? Authors: Sin-Ying Lin Published: 2026-08-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.08254v1 Structured output, where an LLM populates a predefined JSON schema, has become a default mechanism for data labeling and information extraction, but it also introduces a second instruction channel through schema descriptions. We tested whether classification-label definitions are better placed in the system prompt, user prompt, or schema description using a single-field classification task with nonce labels across ten model configurations from two vendors. Schema descriptions did not consistently outperform prompt-based placement; for GPT-4.1 and GPT-5.4 without reasoning, schema placement underperformed system prompts by 11-13 percentage points. Yet schemas are not inert metadata: when prompts and schemas conflicted, incorrect schema instructions caused accuracy drops of 5-45 points, with Claude Haiku 4.5 falling from 52.5% to 7%, indicating that schema instructions can override prompt instructions, and GPT-5.5 falling from 100% to 73%. Further, adding a required intermediate reasoning field before the label field improved schema-only accuracy by 15-24 points when headroom existed, exceeding system-prompt-only performance in every case tested. The effect held even for Claude Sonnet 4.6 at medium reasoning, where extended thinking alone did not produce a comparable gain. This suggests that schema design can affect how effectively models use information encoded in field descriptions. Overall, these results indicate that schema influence is model-dependent. In practice, the system prompt remains a safe default for definitions, but the bigger discipline is maintaining a single source of truth and preventing prompt/schema drift. More importantly, schema design itself may be a stronger lever than instruction placement. Practitioners should treat prompts and schemas as a unified instruction surface and empirically validate both placement and field design for their target model. --- ### SuperLocalMemory 4.0: The Governed Memory Operating System for AI Agents Authors: Varun Pratap Bhardwaj, Garima Singh, Arun Pratap Bhardwaj Published: 2026-08-08 Categories: cs.AI, cs.IR arXiv: https://arxiv.org/abs/2608.08253v2 We present SuperLocalMemory 4.0, a governed, local-first memory operating system for AI agents, unifying multi-channel retrieval under reciprocal-rank fusion, bi-temporal recall, multi-scope isolation, role-based access, verified erasure, and a hash-chained audit trail. A reliability spine governs the primary write path: generation-fenced admission, verifiable memory transactions with per-projection apply, verify, compensate and erase owners, and hash-checkable completion manifests. Eleven fault-injection scenarios, each repeated 200 times, upheld 2,199 of 2,200 scoped component properties. This version leads with a negative result. Ten mechanisms here were implemented, reachable on a live call path, and ineffective at their final connection. Implemented, reachable and effective are three different questions, and the third requires an oracle independent of the mechanism under test. We contribute two mechanical invariants that supply one: a prior-distance assertion over Bayesian learners, and a join-liveness assertion over schema-guarded paths that reports where a guard's missing data resides. A three-arm ablation varying only the recall session-identifier namespace moves no posterior with the defect present and every instantiated arm with it absent, while a negative control that writes every ticket but supplies no engagement settles nothing. We withdraw the previous version's governed write-envelope overhead figure: the two paths it differenced are not comparable. Timing the envelope in place gives an 11.0 ms governed write of which the envelope is 70.6 percent, but the generation fence costs 1.9 microseconds and the obligation ledger 42 microseconds. The cost is durability, not governance. --- ### Conditional Distribution Estimation for Functional Responses with Random Forests Authors: Poorbita Kundu, Antonio R. Linero Published: 2026-08-08 Categories: stat.ME, stat.ML arXiv: https://arxiv.org/abs/2608.08247v1 Many functional data analyses reduce random functions to scalar summaries or conditional mean curves. This is limiting when we wish to understand how covariates affect the distribution of entire functional responses, including their shape, timing, or variability. We study the problem of estimating conditional laws of functional outcomes and show that these objects can be estimated and evaluated in a practical nonparametric framework. To do this, we introduce functional distributional random forests, which estimate each conditional law as a covariate-dependent distribution over sampled functions by training a random forest to minimize a kernel-based maximum mean discrepancy within the leaf nodes of the decision tree. This supports inference on arbitrary functionals of the conditional distribution while keeping predictive samples tied to realistic curves. We consider a variety of kernels defined on function spaces, including Sobolev and operator-induced kernels. We also provide conditions for consistency of our estimator and develop scoring rules for comparing it to baseline estimators. In simulations, our method recovers distributional changes that are missed by baseline methods. In an application to NHANES accelerometer data, it identifies interesting covariate-associated changes in both median activity profiles and predictive dispersion. --- ### Privacy-Preserving Data Drift Detection and Recovery for Large-Scale LLM Applications via Proxy Representations Authors: Michael Levit, Josh Ledgard, Haoyu Dong, Vishwas Suryanarayanan, Eyal Kolman, Sharon Tan, Qiang Gan, Vishal Chowdhary Published: 2026-08-08 Categories: cs.CR, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2608.08245v1 LLM applications deployed at scale face a fundamental challenge: privacy constraints prevent direct inspection of user interactions, making it difficult to obtain any representative evaluation dataset or to track the ongoing evolution of production traffic. We present ProxyDrift, a framework that (i) identifies and measures drift between production traffic and offline evaluation sets, and (ii) constructs and refreshes those evaluation sets accordingly; all without access to raw user data. Our approach operates entirely on non-PII proxy representations: structured, multi-dimensional descriptors derived from LLM-based classification of user interactions. We introduce (1) a chance-calibrated, redundancy-aware (RA) alignment score that aggregates per-dimension drift measurements via mutual information; (2) a conditional sampler that generates synthetic proxies respecting inter-dimensional dependencies; (3) a roundtrip consistency analysis that exposes generator/classifier disagreements and guides proxy taxonomy refinement; and (4) a feedback-linkage analysis that ties per-dimension and per-value proxy distributions to user satisfaction, surfacing actionable failure and success modes. Serving hundreds of millions of users, ProxyDrift enables continuous drift monitoring and targeted synthetic data generation without exposing sensitive user data. Experiments confirm strong roundtrip consistency, discriminator-level indistinguishability of synthetic queries from human queries, and tight end-to-end alignment (RA~0.9) with production. --- ### FemWear: A Parameter-Efficient Wearable Foundation Model for Women's Health Authors: Yifan Wang, Chenzhong Li Published: 2026-08-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.08244v2 General-purpose wearable foundation models are pretrained on broad sensor streams and populations, but their representations are not organized around women's health. FemWear is a women's wearable foundation model, obtained by parameter-efficiently repurposing a pretrained general multimodal wearable backbone into a specialized representation for women's health. It keeps the pretrained patch projection and Transformer encoder frozen and trains 239,236 encoder parameters - 1.11% of a 21.54M-parameter encoder - through low-rank residual adapters and causal task-family heads, producing one shared longitudinal representation for menstrual, symptom, affective, sleep/recovery, autonomic, activity, and pregnancy outcomes. We evaluate six cohorts with 63 comparable primary metrics, 33 from women's-health cohorts, while retaining the 32-task OpenMHC ability-retention benchmark. On a fixed participant split over three seeds, FemWear improved cycle-phase macro-F1 by 8.15% and reduced mean absolute error for cramps, mood symptoms, and sleep problems by 9.32%, 5.80%, and 9.43%; 24-hour onset AUPRC decreased by 3.40%. A stricter 42-participant nested leave-one-participant-out audit retained positive changes for 24-hour onset (+2.87%), 72-hour onset (+6.35%), and cramps (+2.19%), while phase, mood, and sleep changes were neutral or negative and no endpoint had a strictly positive corrected confidence interval. Capacity-matched experiments beat a latest-day multilayer perceptron but not shared-GRU or multi-gate mixture-of-experts baselines. Train-only calibration reduced onset expected calibration error by 84.2-88.2% with zero temporal-nesting violations. FemWear is therefore a women's wearable foundation model: a reproducible, parameter-efficient specialization delivering targeted transfer across women's-health tasks and coherent probability outputs. --- ### A Fair Objective for Human-Empowerment-Preserving AI: Desiderata, Design, and Likely Behavioral Consequences Authors: Jobst Heitzig, Ram Potham Published: 2026-08-08 Categories: cs.AI, cs.GT, cs.MA, econ.TH arXiv: https://arxiv.org/abs/2608.08240v1 This paper explores the idea of promoting well-being and safety in human-AI interactions by forcing AI agents explicitly to empower humans and to manage the power balance between humans and AI agents in a desirable way. Using a principled, partially axiomatic approach based on desirable properties, we design a parametrizable and decomposable objective function for AI systems that represents an inequality- and risk-averse long-term aggregate of human power. It can take into account models of human bounded rationality and social norms, and crucially, considers a wide variety of possible human goals. We prove how certain desiderata enforce particular functional forms and restrict parameter ranges. We exemplify the consequences of softly maximizing this metric in several paradigmatic situations and describe what instrumental sub-goals it will likely imply. --- ### The Replay Gap: Static Evaluation of Model Switching in LLM Agents Scores the Wrong World Authors: Ashritha Gonuguntla Published: 2026-08-08 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2608.08239v1 LLM routers promise efficiency by matching each request to the cheapest adequate model, and are increasingly applied per step inside multi-step agents. Yet agentic routers are evaluated like single-turn routers: by replaying logged trajectories and substituting another model's recorded outputs, assuming the rest of the trajectory is unaffected. We test this assumption with branching rollouts: we fork live SWE-bench agent trajectories at controlled points, rebuild the environment, continue each fork with a different model, and compare against same-model control forks that isolate sampling and replay noise. Across six paired runs (~900 rollouts), swaps exceed their matched control floors by +0.25 to +0.66 normalized edit distance (multiplicity-corrected CIs exclude zero), rewriting 61-94% of post-fork actions; 74-77% of early swaps diverge at the first post-fork action, versus 6-35% of controls, leaving only 3% of replayed states valid. Divergence decreases with fork depth in both directions. All five outcome flips we observe occur in swap arms, upgrades rescuing unsolved instances and a downgrade losing the sole solve, and zero occur across 359 control forks. Scoring these same swaps with a log-stitching replay evaluator, replay mispredicts every success-relevant outcome call and predicts patches with 0.00-0.11 similarity to reality. Auditing the noise floor, temperature-0 "determinism" is configuration-dependent: FP8-served controls diverge on over 90% of forks while AWQ-served ones remain near-identical; and under tight budgets the stronger model more often exhausts its steps without submitting. Replay-based benchmarks score the wrong world for agentic routing; we release our harness and all trajectories. --- ### SAGE: SLO-Aware Adaptive Retrieval for Production RAG Systems Authors: Muhammad Faizan Raza, Shuo, Yang, Satish Mahadevan Srinivasan Published: 2026-08-08 Categories: cs.LG, cs.CL, cs.DC, cs.IR arXiv: https://arxiv.org/abs/2608.08237v1 Retrieval-Augmented Generation (RAG) systems in production operate under strict service level objectives (SLOs) on tail latency and infrastructure cost. However, standard retrieval pipelines rely on fixed retrieval budgets that ignore query difficulty, over-retrieving for easy queries and under-serving hard ones, forcing operators to trade answer quality against SLO compliance. This paper proposes SAGE, a learned SLO-aware adaptive retrieval policy that dynamically selects the number of passages k per query. SAGE uses lightweight features derived from initial retrieval (e.g., score distributions, rank gaps, lexical signals) and is trained offline via imitation learning from an oracle that approximates optimal latency-quality trade-offs. At inference, it adds no LLM calls and minimal overhead. On Natural Questions, under a 5s P95 latency SLO, SAGE achieves 95% SLO compliance versus 30% for the best static baseline (k=20), reduces P95 latency by 36% and retrieval cost by 51% with only 2 percentage points Exact Match (EM) loss. A single policy trained on Natural Questions generalizes across HotpotQA, UnSeenTimeQA, and four LLM families (Llama, Qwen, Mistral, Gemma), consistently yielding +45-52 point SLO improvements without quality degradation. --- ### LatticeMind: A Conflict-Aware Memory Primitive for Multi-Agent Systems Authors: Heng Zhou, Lian Zhang, Yutao Fan, Tiancheng He, Siki Chen, Hejia Geng, Philip Torr, Zhenfei Yin Published: 2026-08-08 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.08236v1 Multi-agent LLM systems often fail not for lack of candidate answers, but because they have no persistent mechanism for deciding which incompatible claim should currently be trusted. Majority vote, debate, and judge-based selection choose an output without recording which claim wins, which is contested, or why a later update supersedes it. We present \term{LatticeMind}, a conflict-aware structured memory that handles contradiction at write time. It maintains explicit item status, applies cheap symbolic conflict checks, and invokes LLM reconciliation only for unresolved semantic cases. On a label-blind ConflictBank evaluation that removes source-name hints, LatticeMind reaches 0.97 accuracy versus 0.61 for the strongest aggregation baseline, with the gap significant at $p<10^{-6}$ by paired McNemar test. Ablations show that removing the checker or the reconciler costs 12 to 14 points. On four secondary planning benchmarks the picture is mixed: LatticeMind beats naive merge on three of four, but does not replace deliberation methods on tasks rewarding iterative search. --- ### Focus particles and scalar inferences across humans and language models Authors: Catherine M. Brousse, Nelu D. Radpour Published: 2026-08-08 Categories: cs.CL, cs.HC arXiv: https://arxiv.org/abs/2608.08227v1 Focus particles such as "even" and "only" are central to formal semantic theories that posit structured representations over sets of alternatives. "Even" highlights unexpected or extreme alternatives, while "only" enforces exclusivity. If such scalar representations are robust and generalizable, they should give rise to consistent judgments across contexts and systems. In this work, we test whether humans and large language models (LLMs) construct stable scalar representations from sentences containing these particles. Using a dataset of approximately 100 items, participants and models were asked to make scalar judgments. Preliminary results suggest that similar outputs across humans and LLMs may arise from different underlying mechanisms. --- ### High-Capacity Generalized Hopfield Networks Authors: Victor Galitski Published: 2026-08-08 Categories: cond-mat.stat-mech, cs.CV, cs.NE, quant-ph arXiv: https://arxiv.org/abs/2608.08226v1 Generalized Hopfield networks are introduced where memories and neurons are continuous variables that lie on a Riemannian manifold. We explicitly focus on symmetric spaces associated with the special unitary groups SU(d), and use both numerical and analytical (replica) techniques to demonstrate an almost order of magnitude enhancement in critical capacity over the vector networks starting with d=3 and further rapidly growing with d. To circumvent the non-linear geometric constraints, we use a Lie algebraic method (following V. Galitski, Phys. Rev. A 84, 012118 (2011)) to exactly describe the classical neural network in terms of linear algebra in an auxiliary Hilbert space. It is shown that in contrast to the traditional Hopfield networks, memory recall in SU(d) Hopfields corresponds to neuron alignment along a top eigenvector of a spiked matrix, which is less susceptible to random matrix crosstalk than other models with continuous neuron variables. Physical platforms to realize SU(d) Hopfields are briefly discussed and physical (in addition to algorithmic) recall mechanism is demonstrated, where memory recovery occurs naturally through generalized Landau-Lifshitz-Gilbert dynamics. To illustrate SU(3) memory recall, we introduce a color (RGB) image encoding/decoding protocol and explicitly run image recovery on corrupted cues. Finally, we quantize the generalized Hopfields which are shown to reduce to Sachdev-Ye glassy type of models. Their many-body spectra generally feature two types of dark and memory bands, where the latter exhibits chaotic Wigner-Dyson level statistics that hides Hebbian data. --- ### Control-Diverse Reinforcement Fine-Tuning: Decoupling the Shared Control Bottleneck of RL Post-Training Authors: Binwen Tan, Jingchao Wang, Dengzhe Hou, Lingyu Jiang, Zeyuan Wu, Yunhan Shen, Fangzhou Lin, Kazunori Yamada, Atsushi Koike Published: 2026-08-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.08224v1 Reinforcement learning post-training unlocks complex reasoning in LLMs. Yet benchmark scores reveal only whether a model improved, not what changed inside it, nor how it splits finite capability across tasks. A representative interpretability line attributes the success of RL fine-tuning to stronger and more diverse circuit activation. We challenge this activation-centered account by separating activation from control: an activated circuit need not control the post-training reward gain. Adapting Metabolic Control Analysis, we define the Post-training Control Coefficient to measure component control over reward gain and arrange these coefficients by task family into a control matrix, paired with an activation-magnitude matrix. We call cross-task control concentration the Shared Control Bottleneck and the difference between activation and control concentration the Activation-Control Gap. This reveals that highly shared activations can coexist with task-specific control, while a small gap indicates that control has collapsed onto a shared direction and lost task specificity. To reduce this collapse, we regularize the post-training loss with the Shared Control Bottleneck and propose Control-Diverse Reinforcement Fine-Tuning (CD-RFT). The exact regularizer gradient requires second-order automatic differentiation incompatible with flash attention, so we derive a first-order proxy with worst-case overhead below eight percent. On Qwen2.5-7B, CD-RFT achieves the largest control decoupling and improves multi-task capability over matched GRPO across mathematics, code, and logic. The no-KL variant leads on pass@1, and the KL-penalized variant leads on large-k pass@k coverage that KL otherwise degrades. Together, these results show that the Shared Control Bottleneck is both a mechanistic diagnostic and a training regularizer, and that control decoupling and capability gains transfer to Llama-3.2-3B. --- ### Metanormative Theory for RL-Based Moral Agents Authors: Aleks Knoks, Marija Slavkovik Published: 2026-08-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.08220v1 The overlapping disciplines of machine ethics and value alignment are concerned with designing artificial agents that are aligned with human values and that act in ethically acceptable ways. A recent trend in these disciplines is the use of reinforcement learning (RL) to design such agents, sidelining the philosophical literature that used to play a more central role. Against this backdrop, this paper pursues two goals. The first is to draw out ideas from recent work in metanormative theory that can be useful for designing artificial moral and value-aligned agents. The second is to examine the RL architecture through the lens of these ideas. This will give us clearer criteria for when an RL agent's behavior can be classified as moral, as well as a basis for evaluating and comparing different RL-based approaches to machine ethics and value alignment. --- ### VTO: Visual Tool Orchestration for Video Anomaly Detection Authors: Rui Wang, Yeteng Wu, Xianling Zhang, Mengshi Qi Published: 2026-08-08 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.08219v1 Video anomaly detection (VAD) is a critical yet challenging task due to the complex and diverse nature of real-world scenarios. Traditional deep learning approaches are fundamentally limited by poor generalization across diverse scenarios. While multimodal agents offer a promising tool-learning paradigm for VAD, current systems relying on supervised fine-tuning struggle with complex orchestration, and standard reinforcement learning often causes premature termination due to coarse-grained outcome rewards. To address these challenges, we propose VTO, a process-supervised reinforcement learning framework. Moving beyond static tool usage, VTO enables the agent to dynamically explore and interact with the environment. Specifically, we introduce a foundation model-driven cognitive evaluator to provide context-aware semantic feedback, which is seamlessly integrated into a Process-Supervised Cognitive Alignment that delivers fine-grained, step-wise supervision. By explicitly penalizing logical truncation and rewarding complete causal chains, the agent optimizes its multi-step reasoning policy for interrelated tool orchestration. To support our proposed framework, we meticulously crafted VAD-Tool, a hierarchical visual tool set comprising 12 specialized vision tools spanning from entity tracking to high-stakes hazard detection, and established the corresponding benchmark for rigorous multi-step reasoning evaluation. Extensive experiments on VAD-Tool demonstrate that VTO significantly outperforms baselines, achieving up to a 10.2\% absolute accuracy improvement in tool scheduling. Code and data are available at https://github.com/MICLAB-BUPT/VTO. --- ### Harmful Content Is Not Enough: Continuation Framing Moderates In-Context Emergent Misalignment Authors: Peiyang Liu, Xi Wang, Ziqiang Cui, Di Liang, Wei Ye Published: 2026-08-08 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.08212v1 In-context learning (ICL) can induce emergent misalignment (EM), where narrow misaligned examples alter answers to unrelated questions. Existing prompts, however, conflate harmful-text exposure with an invitation to continue assistant behavior. We hold harmful answers fixed while varying their delivery as demonstrations, evidence, assistant history, or tool output. Across ten independently sampled contexts, demonstration framing raises broad EM by $30$--$32$ percentage points on a susceptible Gemini model; the gap survives domain exclusion, semantic clustering, unseen questions, and four prompt templates. Format and length-matched controls show that harmful content is necessary but insufficient. A role times continuation factorial further reveals model-dependent provenance effects: Gemini follows both assistant and tool histories, whereas Grok largely resists tool-framed continuation. Several other frontier and open-weight models show no gap. Blinded human audits confirm every main contrast and show that the model judge underestimates active-condition failures. Thus continuation framing is a strong, model-dependent moderator of ICL-EM, not a universal consequence of harmful context. --- ### Retrieval-Augmented Generation-Based Color Restoration for Low-Light Image Enhancement Authors: Li-Wei Lu, Shaou-Gang Miaou Published: 2026-08-08 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2608.08211v1 Recent low-light image enhancement (LLIE) methods have driven brightness and structural fidelity close to that of normally-exposed images, yet their outputs still exhibit systematic color shifts such as greenish skies, yellowish faces, and warm-tinted white objects. We attribute this to end-to-end LLIE training coupling brightness, structure, and color within a single network, leaving the color channels weakly supervised. We recast color restoration as an independent sub-problem and decouple it from brightness enhancement, realizing it as a general-purpose post-processing module built on retrieval-augmented generation (RAG). Rather than relying solely on parametric color priors learned during training, the module dynamically retrieves a reference image from an external high-quality color knowledge base and injects its color distribution into a color-restoration network to correct residual bias. The design has three components: (i) a dual-index FAISS retriever built on intermediate VGG19 features, capturing textural and structural similarity through global mean and variance statistics; (ii) GlobalSPHistAdaIN, which reduces the reference spatial-preserving color histogram to a global color vector and modulates network features via adaptive instance normalization, removing dependence on pixel-level correspondence; and (iii) a residual formulation that predicts a color correction over the front-end output. Across LOLv1, LOLv2-Real, and LOLv2-Synthetic, the module consistently improves color-specific metrics, and it remains effective when the front end is swapped among CPGA-Net++, LLFormer, FLIGHTNet, and IAT, confirming cross-front-end generality. Ablations show that a VGG19 dual index outperforms CLIP-based retrieval, indicating that color restoration depends on textural and structural similarity rather than high-level semantics. --- ### Illusion of Alignment: Detecting Hidden Disagreement in Collaborative Dialogue Authors: Kaiming Liu, Fuwen Luo, Ziyue Wang, Jinrui Ju, Yuxuan Liu, Xuanyu Lei, Yunghwei Lai, Peng Li, Yang Liu Published: 2026-08-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.08210v1 Collaborative dialogue can end with apparent agreement while participants still differ on goals, assumptions, or execution plans, creating an \textbf{illusion of alignment (IoA)}. A real-user study across 18 meetings confirms that IoA arises routinely in human collaboration. Yet IoA poses a paradox: if participants were aware of such disagreements, they would already be explicit; if not, they cannot articulate them when asked, leaving IoA invisible to both participants and observers. In this work, we make IoA detectable by generating diagnostic multiple-choice questions whose divergent answers across participants provide direct behavioral evidence of hidden disagreement. We construct \textbf{IoA-Suite}, a dataset and evaluation protocol for detecting hidden disagreement, spanning five task types and six domains. We find that even the best model attains only 49.5\% F1, with the bottleneck traced to private context that the dialogue does not surface. We then train \textbf{IoA-Prober-8B} based on IoA-Suite, reaching 51.8\% F1 on IoA-Suite. Across the aforementioned 18 real meetings, it surfaces 2.89 hidden disagreements per meeting that participants confirm they had not voiced, transferring to live human dialogue. Further, in multi-agent collaboration, pairing IoA-Prober-8B with LLM agents improves downstream task performance on BigCodeBench-Hard and HiddenBench. --- ### FreSH: Frequency-Segmented Hierarchical Multi-Expert Framework for Multivariate Time Series Classification Authors: Pingping Liu, Muyao Wang, Zijian Zhang, Tongshun Zhang, Hao Miao, Guorui Xie, Qingliang Li, Qiuzhan Zhou Published: 2026-08-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.08207v1 Multivariate Time Series Classification (MTSC) demands models that can effectively capture complex temporal patterns across multiple scales while remaining computationally efficient. However, existing approaches generally struggle to reconcile fine-grained representation learning, especially under class imbalance and real-world constraints. In this paper, we present FreSH, a Frequency-Segmented Hierarchical Multi-Expert Framework designed to address these challenges. FreSH introduces a new perspective for MTSC by enabling adaptive, multi-scale analysis of temporal signals, allowing different aspects of the data to be modeled in a complementary and coordinated manner. By combining localized specialization with holistic context modeling, FreSH achieves strong representational capacity without incurring excessive computational overhead. An adaptive fusion strategy further enhances flexibility, enabling the model to dynamically emphasize the most informative components of the input. In addition, we incorporate a more robust optimization objective that improves learning stability across varying sample difficulties and class distributions. Extensive evaluations on 30 UEA benchmark datasets and real-world vibration data demonstrate that FreSH consistently outperforms state-of-the-art methods in classification accuracy, while substantially reducing model size and efficiency. ---