# Research Papers — 2026-08 · bundle page 159 of 240 > Full-text markdown of papers #7901–7950 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/159 [← Previous bundle](https://aipod.dev/papers/archive/2026-08/page/158.md) [Next bundle →](https://aipod.dev/papers/archive/2026-08/page/160.md) ### A Convolutional Layer Activation Dimensionality Reduction for Out-of-Distribution and Adversarial Attack Detection Methods Authors: Leandro de Souza Rosa, Lorenzo Capelli, Clara Nunes Barrancos, Mauro Mangia, Riccardo Rovatti Published: 2026-08-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.10203v1 Despite the success of convolutional neural networks in image classification tasks and their general application in multi-modal models, their susceptibility to out-of-distribution and adversarial attack samples raises concerns regarding trustworthiness and safety. Among the approaches to tackle such issues, detection methods that analyze the model's intermediate activations to estimate a confidence score are a promising family that evaluates the decision process, relying on a dimensionality reduction step to enable efficient downstream processing of the high-dimensional activations. However, when considering convolutional layers, the dimensionality reduction methods in the literature either lack a mechanism to control the compression/information-loss trade-off or yield large representations. In this paper, we carefully analyze two state-of-the-art detection methods and their dimensionality reductions for convolutional layers and develop a novel reduction method with a controllable high-compression level. We extend these two state-of-the-art detection methods, enabling the usage of any dimensionality reduction, and evaluate their performance on out-of-distribution and adversarial attack detection. Results show that the detection methods with the proposed dimensionality reduction consistently perform better than, or comparable to, the strongest alternative. Furthermore, the proposed method is shown to reduce computation and memory footprints, given that it has the highest compression among the compared methods. --- ### Post-Hoc Sparse Coding of Latent Communication Between Vision-Language Model Agents Authors: Di Wu, Xiaohui Zhu Published: 2026-08-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.10198v1 Latent-space communication allows heterogeneous vision-language model agents to exchange continuous representations without serializing visual and reasoning states into text. Vision Wormhole realizes this approach by translating visual features into a universal latent representation that can be consumed by another model, but every message is transported as a dense tensor of the same size regardless of its content. A fixed-capacity dense tensor therefore need not have a fixed effective information density: some messages may use only a small fraction of the available representational degrees of freedom. This observation suggests that the communication channel may be substantially compressible. We study its redundancy by fitting a post-hoc sparse autoencoder to frozen Vision Wormhole activations and measuring reconstruction, downstream utility, feature reuse, and token-level interventions across nine reasoning benchmarks. Relative to the original float32 transport, a uint16-index/float16-value sparse payload with k=4 active coefficients per token reduces the transmitted bytes by 128x. In a single-run evaluation, the seven-task non-AIME mean accuracy changes from 49.85% to 49.77%. The fitted 4096-element dictionary uses only 50 features, and task-level active sets have a mean pairwise Jaccard similarity of 0.906. These measurements establish strong post-hoc compressibility relative to the original transport, but do not yet isolate the incremental contribution of sparse coding from position selection, reduced precision, low-rank structure, or SAE optimization effects. The results motivate matched-payload comparisons and communication mechanisms whose payload adapts to the information used by each message. --- ### ELMER: Evolutionary Language Model that Explores and Refines Authors: Matthew Siper, Ahmed Khalifa, Julian Togelius Published: 2026-08-10 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.10196v1 Program evolution can measure whether a mutation helped, but it rarely controls how far the mutation moves in behavior space. Syntactic edit size is an unreliable proxy: a small code change can alter nearly every action, while a larger rewrite can preserve the same execution trace. We introduce an Evolutionary Language Model that searches over natural-language policy descriptions and compiles typed programs for execution. A fully fine-tuned Qwen3-8B model learns three task-conditioned operations: conditional semantic mutation, natural language to domain-specific language (GPTL) compilation, and GPTL to natural language translation. The model is fine-tuned with conditional input on the mutation strength (low, medium, high) using Direct Preference Optimization (oDPO). Across 252 fixed-budget evolutionary searches, oDPO improves both behavioral calibration and finite-budget search efficiency. Natural-language attains the highest observed held-out fitness. Our analysis shows that the condition input (mutation strength) systematically changes semantic edit composition and that language mutations preserve more parent fitness at matched small-to-moderate behavioral displacement. These results show that language can serve as a steerable, execution-grounded search representation over executable program space. --- ### More Accurate, Less Human: Gestalt Grouping in Vision Models Authors: Sudhanva Manjunath Athreya, Sai Phani Kumar Malladi Published: 2026-08-10 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2608.10195v1 Human vision organizes what it sees into wholes: same-colored points group into series, similar marks cohere into categories, and shapes complete into recognizable objects. These are the Gestalt operations that visualization design builds on. Whether vision models organize visual content this way has not been systematically tested. We introduce a behavioral battery that scores models against human data from prior perception studies on four grouping tasks: mark-color odd-one-out, color-series counting, silhouette recognition, and object odd-one-out. We apply it to 45 models across five training families: supervised, self-supervised, and contrastive vision-language encoders, open-weight VLMs, and closed foundation models. The battery reveals that agreement with human responses captures aspects of perceptual organization that conventional performance metrics fail to distinguish, with several closed models exhibiting substantially lower alignment than their benchmark accuracy would suggest. Scoring against published perception data therefore gives visualization research a reusable yardstick, requiring no new user study, for auditing whether the models now entering visualization pipelines organize what they see the way their human audience does. --- ### The Deliberative Deficit: An Empirical Critique of LLMs in Democratic Discourse Authors: Maurice Flechtner Published: 2026-08-10 Categories: cs.MA, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2608.10186v1 LLMs are increasingly deployed in settings that require collective reasoning on complex, value-laden problems. Confidence in these deployments rests largely on benchmarks for verifiable tasks (mathematics, coding, coordination games), yet many of these applications concern problems where no objectively correct answer exists and where decision quality instead depends on integrating pluralistic perspectives to find mutually acceptable solutions. We argue that LLM reasoning capacity on this class of problems cannot be fully inferred from verifiable-task benchmarks, and that procedural evaluations of LLM discourse (respectfulness, justification, engagement) are systematically insufficient. We apply the Deliberative Reason Index (DRI), a measure developed in political science and validated across citizen assemblies, as a tool for evaluating reliable group-level reasoning on pluralistic, non-verifiable problems. Synthesizing recent evidence across 1,980 five-agent LLM runs on 12 citizen-assembly topics across 11 frontier model configurations, we find that LLM groups produce discourse with procedural quality comparable to human deliberation, while gains in intersubjective consistency are small, topic-dependent, and concentrated on tractable rather than ethically contested questions. LLM groups exhibit roughly one-third the perspective diversity of human assemblies and reverse the human convergence pattern: human deliberation decreases dispersion as diverse views synthesise, whereas LLM deliberation increases it. Engineering diversity through persona prompting does not restore the human dynamic but inverts which component of deliberative reasoning is updated. Our conclusion is constraining rather than prohibitive: LLMs can function as tools supporting human reasoning on pluralistic problems, but current evidence does not license treating them as autonomous deliberative agents. --- ### The sharp SAT/UNSAT phase transition in random ellipsoid fitting Authors: Theodor Misiakiewicz, Garrett G. Wen Published: 2026-08-10 Categories: math.PR, cond-mat.dis-nn, cs.DS, math.ST, stat.ML arXiv: https://arxiv.org/abs/2608.10184v1 Let $x_1,\ldots,x_n$ be independent standard Gaussian vectors in $\mathbb{R}^d$. An \emph{ellipsoid fit} is a matrix $S \succeq 0$ such that $x_i^\top S x_i =d$ for every $i$, so that all the points lie on the boundary of the centered ellipsoid $\{ x : x^\top S x = d\}$. Saunderson, Parrilo and Willsky conjectured that, as $n,d \to \infty$, this semidefinite feasibility problem undergoes a sharp transition at $n \sim d^2/4$. We prove this conjecture. If $\lim \sup n/d^2 = α^* <1/4$, then, with probability tending to one, an ellipsoid fit exists; moreover, one can choose $S$ with all eigenvalues in a fixed interval $(λ_- , λ_+) \subset (0,\infty)$ depending only on $α^*$. Conversely, if $\lim \inf n/d^2 > 1/4$, then, with probability tending to one, no ellipsoid fit exists, without any spectral restriction. Our proof builds on the Gaussian-equivalence framework developed by Bandeira and Maillard (2025) and closes the two gaps left open in their work: establishing exact fitting and removing the operator-norm constraint. On the satisfiable side, the new ingredients are a head-tail decomposition of the dual vector, exact correction of the sparse head constraints, and a Gaussian comparison principle for the low-influence tail. On the unsatisfiable side, we split a candidate into a low-rank spectral head and a Schatten-3 diffuse bulk, Gaussianize the bulk conditionally on the head, and apply a projected Gordon escape argument. The threshold is governed by the statistical dimension $d(d+1)/4$ of the positive semidefinite cone. --- ### From Prediction to Incrementality: Causal Optimization for Large-Scale Targeting and Recommendation Authors: Changshuai Wei, John Bencina, Phuc Nguyen, Andre Assuncao Silva T Ribeiro, Benjamin Zelditch Published: 2026-08-10 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.10182v1 Large-scale targeting and recommendation systems are typically built around predictive scores fed into heuristic or local allocation. When the business goal is incremental impact, as in marketing campaigns, incentives, and notifications, this paradigm systematically misallocates resources toward users who would have acted anyway. We present a decision-centric framework that instead optimizes causal effects under global constraints, aligning three components under a single objective: a causal neural network with a Transformer backbone for individual treatment-effect estimation, a Bayesian neural-bandit layer for uncertainty-aware exploration, and a dual-based large-scale linear-programming layer for constrained allocation. The framework also supports sequential context and multi-outcome, attribute-conditioned scoring through a Transformer encoder and outcome embeddings. We evaluate it with offline simulations on a public bandit dataset, targeted architectural ablations, and an online A/B test on LinkedIn Feed marketing traffic. We also distill production lessons on causal training-data construction and cost and delivery control, which were critical to successful deployment. The end-to-end treatment policy delivered a statistically significant $+7.20\%$ lift in the primary long-term-value metric, demonstrating the feasibility of production-scale causal optimization under business constraints. --- ### Human versus Computer Vision Authors: Elena Sirotkina Published: 2026-08-10 Categories: cs.CV, cs.CY arXiv: https://arxiv.org/abs/2608.10181v1 Computer vision saliency models predict where people will look, one map per image, and a billion-dollar predicted-attention industry sells those maps in place of measuring real viewers. I test the leading models from the audience side, against 11.4 million webcam gaze points from 3,023 US adults recruited to national quotas, viewing circulating news photographs. I show that an untrained central marker outperforms every trained network, because the content the networks add on top of the center falls where these audiences never look. What accuracy remains is systematically biased, favoring younger, White, and moderate viewers over older, Black, and ideologically extreme ones. I propose a way forward and build on what a group's own gaze reveals about whether a model can learn that group at all, and I apply it across every demographic axis this sample supports. Ultimately, I show how systems that decide what people see can learn to see everyone, and this study supplies the standard by which such a claim should be judged. --- ### TRACE: Trustworthy Retrieval-Augmented Conversational Engine Authors: Touseef Hasan, Laila Cure, Souvika Sarkar Published: 2026-08-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.10176v1 Public service chatbots are expected to deliver recommendations from an underlying public service directory, while also making sure that the recommendations respect explicit user constraints. In practice, public service directories are noisy and inconsistent, and general-purpose large language model (LLM) or AI-based chatbots frequently generate unreliable recommendations, citing unverified sources from the web. We investigate the impact of retrieval quality on constraint-aware recommendation in public service conversational systems built over noisy and heterogeneous service directories. We propose TRACE (Trustworthy Retrieval-Augmented Conversational Engine), a retrieval-based, constraint-aware framework that parses input user queries into structural and semantic constraints for downstream retrieval, with the help of a dual data representation schema. Using a curated statewide pantry directory and a synthetic query benchmark, we evaluate multiple knowledge-representation variants with and without knowledge graphs (KGs). We experiment with several open-source LLMs and a proprietary model, showing that strengthening retrieval substantially improves user constraint satisfaction while reducing hallucinated recommendations. Performance differences across LLMs narrowed in our experiments as retrieval quality improved, making results less sensitive to model size. These findings suggest that the quality of retrieval is key for robust public service conversational systems. --- ### DoseBridge: Denoising Diffusion Bridge Model for Dose Prediction in Lung Intensity-Modulated Proton Therapy Authors: Zerun Zhang, Xiaoda Cong, Xiangkun Xu, Peter Y. Chen, Xuanfeng Ding Published: 2026-08-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.10173v1 Most radiotherapy dose-prediction models use only CT images and anatomical structures, although intensity-modulated proton therapy (IMPT) dose also depends strongly on beam geometry and available clinical datasets are often small. We present DoseBridge, a denoising diffusion bridge model that uses the patient CT as a structured bridge endpoint and encodes plan-specific beam geometry in a spatially aligned beam mask. Multiscale fusion combines CT, target, organ-at-risk, and beam-mask representations with 1.95% additional parameters. DoseBridge was retrospectively evaluated on single-institution CT images and treatment plans from 52 patients with advanced-stage lung cancer treated with 60 Gy in 30 fractions; 42 cases were used for training and 10 for testing. Performance was assessed using image-similarity, dose-volume, and Lyman-Kutcher-Burman normal-tissue complication probability (NTCP) metrics and compared with two deep-learning models. On the test cohort, DoseBridge achieved a mean absolute error of 4.170 Gy, peak signal-to-noise ratio of 23.06 dB, and structural similarity index of 0.798, outperforming both comparison models on these metrics. Clinical target volume D95 differed from the reference dose by 0.62 +/- 1.6 Gy; signed organ-at-risk mean-dose differences ranged from -0.32 to 0.24 Gy, and NTCP differences were -0.40 +/- 2.2 and 0.52 +/- 3.4 percentage points for acute esophagitis and radiation pneumonitis, respectively. Changing only the beam mask redirected predicted low-dose entrance regions while preserving the high-dose target region. To our knowledge, DoseBridge is the first denoising diffusion bridge model for radiotherapy dose prediction. These results support its feasibility as a beam-aware planning prior for lung IMPT, pending evaluation in larger external cohorts. --- ### Intrinsic Structure: Spectral Identifiability for Mechanistic Interpretability Authors: Ashim Dhor, Pin-Yu Chen Published: 2026-08-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.10172v1 Mechanistic interpretability explains models by identifying circuits inside them, but has no way to tell whether a circuit is a property of the model or an artifact of the method that found it. Sparse autoencoders illustrate the problem: different seeds and widths recover materially different features from the same activations, and no theory says whether that variability is incidental or structural. We put dictionary learning for interpretability on an identifiability footing. Treating the forward pass as a controlled dynamical system with depth as time and lifting it with the Koopman operator yields a finite linear realisation whose \emph{spectrum} is a coordinate-free property of the model. We prove the spectrum is recoverable from $M$ calibration samples at rate $M^{-1/2}$ up to permutation - to our knowledge the first identifiability theorem for a mechanistic-interpretability primitive, with a matching minimax lower bound, a median-of-means variant for heavy-tailed activations, and a dissociation theorem: whenever the realisation is non-normal, the directions carrying activation variance and the directions carrying information across depth cannot coincide. The identifiable object and the legible object are not the same object. On GPT-2 small, Gemma-2-2B and Qwen3-8B-Base the spectrum converges everywhere and attains the predicted exponent on Qwen3-8B-Base ($0.506 \pm 0.031$); shortfalls collapse onto one curve against each cell's sample threshold. Koopman modes beat random directions but lose to principal components on indirect-object identification, with the gap decaying $4.1\times$ in depth-distance, as the theorem predicts. The Koopman spectrum is an identifiable, model-intrinsic fingerprint with a stated error bar, not a legible decomposition. --- ### Generating Attacks for LLMs with GFlowNets Authors: Berkay Ozcam, Irem Onen, Mehmet Fatih Amasyali, Emin Islam Tatli Published: 2026-08-10 Categories: cs.AI, cs.CR arXiv: https://arxiv.org/abs/2608.10171v1 The rapid advancement of Large Language Models (LLMs) has facilitated their ubiquitous integration into various domains, leading to widespread adoption. However, this escalating trend has introduced significant security vulnerabilities, necessitating the identification and mitigation of flaws arising from malicious exploitation. Red teaming assessments, conducted to evaluate model robustness through diverse adversarial inputs, are essential for exposing security risks and implementing countermeasures. Currently, red teaming is performed either manually by experts or automatically using predefined attack datasets. Nevertheless, manual testing remains time-consuming, while existing automated methods suffer from limited creativity due to their inherent dependency on fixed datasets. In this study, we propose an automated, human-independent, and adaptive approach leveraging GFlowNets to identify LLM vulnerabilities by utilizing one large language model to test another. Within this framework, an attacker model is trained against a specified victim model to perform automated red teaming and provide a quantitative robustness score. This research aims to generate more effective adversarial attacks in English compared to existing benchmarks and, as a novel contribution to the literature, introduces a model capable of generating attack inputs in the Turkish language. --- ### Motion Artifact-Aware Self-Supervised Representation Learning for 3D Brain MRI Motion Artifact Reduction Authors: Mojtaba Safari, Shansong Wang, Zach Eidex, Matthew Goette, Tonghe Wang, Zhen Tian, Xiaofeng Yang Published: 2026-08-10 Categories: cs.CV, physics.med-ph arXiv: https://arxiv.org/abs/2608.10170v1 Patient motion remains a source of image degradation in brain MRI, leading to signal loss, blurring, and geometric distortion that compromise quantitative analysis. Existing deep learning methods for motion correction typically rely on paired clean-corrupted data or k-space acquisitions, which are rarely available in clinical settings. We propose SSRL-MAR, a motion artifact-aware unpaired representation learning framework for motion artifact reduction that requires neither paired training data nor explicit motion labels. SSRL-MAR employed a three-stage training strategy: (1) contrastive learning on 3D patches to extract motion representations by contrasting clean and synthetically corrupted images, (2) a motion artifact-aware synthesis network to generate motion artifacts from clean scans, and (3) a motion artifact-aware generator to restore clean volumes using the learned degrader for self-supervised supervision. On in-silico dataset, SSRL-MAR achieved PSNR 23.81dB, SSIM 91.55%, and NMSE 0.79%. On in-vivo MR-ART dataset, the pretrained model reduced motion distortion, and unsupervised domain adaptation further improved anatomical fidelity. Against a source-only supervised model trained on the same simulated pairs, SSRL-MAR improved PSNR by up to 2.0 dB on MR-ART after unsupervised domain adaptation, and remained within 0.25-0.47 dB of an oracle supervised model that requires real paired data unavailable in practice. At the milder motion level, volumetric error in structures such as the corpus callosum and ventricular system decreased by more than 50%, confirming improved neuroanatomical consistency. These results indicate that SSRL-MAR provides a robust and scalable image-domain solution for 3D brain MRI motion correction, enabling reliable structural quantification in large-scale neuroimaging studies without requiring prospectively acquired pairs or acquisition-specific calibration. --- ### MarkNull: Model-Agnostic Watermark Removal in AI-Generated Images via On-Manifold Latent Manipulation Authors: Jie Cao, Qi Li, Zelin Zhang, Xiaodong Wu, Lingshuang Liu, Xiangman Li, Jianbing Ni Published: 2026-08-10 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2608.10166v1 Digital watermarking has emerged as a critical technique for provenance and copyright attribution in AI-generated imagery, yet its robustness against realistic, model-agnostic removal attacks remains poorly explored. Existing attacks either succeed only against specific generative models or achieve removal at the cost of severe visual degradation. In this paper, we propose MarkNull, a model-agnostic watermark removal attack via on-manifold latent manipulation. MarkNull is grounded in a key observation: watermarked images exhibit a strong statistical dependency between the generated latent representation and the embedded initial noise. To quantify this dependency, we introduce the Noise-Latent Alignment Score (NLAS) and formulate an optimization objective that selectively decorrelates the latent representation from the embedded watermark while preserving semantic fidelity. Extensive evaluations across different categories of watermarking paradigms, including post-hoc, fine-tuning-based, and initial-noise-based schemes, demonstrate that MarkNull reduces average bit accuracy to 53.14%, approaching random-guessing (50%), without perceptible image degradation. To further improve scalability, we propose MarkNull-A, an amortized, optimization-free variant that distills the attack into a single forward pass, achieving 0.50 s/image with modest computational overhead. Notably, our attacks successfully compromise Google's SynthID-Image system while preserving high visual quality and transfer effectively to video watermarking. Finally, we present an attack detection mechanism as a defensive counterpart to MarkNull and MarkNull-A, highlighting the necessity of developing watermark designs resilient to model-agnostic latent-space attacks. --- ### MAD-HOI: Masked Autoregressive Diffusion for Generating Articulated Hand Object Interactions from Text Authors: Ananya Bal, Kartik Sharma, Ethan Lai, Samyak Tiwari, Liza Dahiya, Chaitanya Chawla, Laszlo A. Jeni Published: 2026-08-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.10162v2 Methods for text-based generation of hand-object interaction (HOI) sequences primarily focus on producing smooth, physically plausible trajectories. A truly utilitarian method should additionally support variable-length generation, composite motion sequences, motion completion and infilling, and reliable termination without compromising physical plausibility. Standard diffusion models for HOI generation are typically trained only for text-to-motion generation on atomic motions and require the motion length to be specified a-priori. Autoregressive (AR) methods provide greater sequence-level flexibility, but commonly depend on discrete motion codes, which can lose contact-sensitive motion detail. To address these key limitations, we present a model performing Masked Autoregression with Diffusion for HOI generation (MAD-HOI). Our method starts by encoding hand and object motions in a continuous latent space while keeping them disentangled to maintain stream-wise control. This is followed by a masked autoregressive transformer to predict context features that condition a flow-matching head. MAD-HOI is capable of motion generation for atomic and composite articulated sequences, conditioned motion completion and infilling, as well as EOM (End of Motion) prediction from a single training objective. We provide comprehensive evaluations for these capabilities and benchmark our method on the ARCTIC and GRAB datasets. Our experiments demonstrate that our method generates more diverse and physically plausible interactions compared to other open-sourced baseline methods. --- ### SBCO: Self-Supervised, Verifier-Grounded Harness Optimization For Planning Agents Authors: Vivek Kulkarni, Sudipta Paul, Aounon Kumar, Nicholas Tzou, Srinivas Chappidi Published: 2026-08-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.10157v1 Self-improving agents seek to reduce the human engineering effort behind AI systems by enabling them to evolve and self-improve their performance over time. Recently, methods like the Darwin Gödel Machine and the Huxley Gödel Machine have been proposed which enable open-ended, recursive self-improvement through self-reference where a coding agent edits its own code. Such self-referential self-improvement methods require that the competence required to perform the task coincides or aligns well with the competence required for self-modification which is the case for coding tasks. For domains or tasks, which do not satisfy the alignment needed, self-referential self-improvement is not available. In such cases, it is possible to adapt the above algorithms to other tasks by removing the self-referential aspect or introducing explicit self-modification of a meta-agent -- both computationally expensive, relying on population or self-modification search over many candidate agents. For planning tasks with explicit constraints, we propose a far cheaper alternative. We introduce SBCO (Self-supervised Block Coordinate Optimizer), a verifier-grounded harness optimizer in the same closed-loop, improve-from-experience family as the Gödel-machine methods, but self-supervised rather than self-referential. Given an agentic harness, SBCO learns a decomposed bank of verifiers and a harness policy via approximate block coordinate ascent, improving the agent's outputs from its own graded feedback---with a fixed meta-agent and no human labels. Across two domains SBCO matches or exceeds a customized self-modifying baseline while using 4-5.5 times less compute budget. --- ### Optimal Inference with Black-box Predictions Authors: Lucas Kania, Abhinav Chakraborty, Edward Kennedy, Larry Wasserman, Sivaraman Balakrishnan Published: 2026-08-10 Categories: math.ST, stat.ME, stat.ML arXiv: https://arxiv.org/abs/2608.10155v1 Powerful black-box predictive models have motivated many proposals for combining observed data with predictions to perform valid statistical inference. Despite this progress, the field lacks a unifying principle that explains how hypothesis tests should integrate data and predictions in a way that is both valid and efficient. In this work, we address this gap in the high-dimensional Gaussian sequence model. We characterize the information-theoretic limits of inference with black-box predictions when their accuracies are known and, for orthogonal predictions, when they are unknown. Building on these characterizations, we develop practical hypothesis tests that adapt to the unknown accuracies of the predictions while benefiting from strong alignment among them. --- ### Multimodal Item Parameter Estimation using Simulated Response Probabilitie Authors: Christopher Ormerod, YoungKoung Kim Published: 2026-08-10 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.10154v1 We present results from reconstructing multiple-choice model (MCM) and three-parameter logistic (3PL) model curves using a fine-tuned multimodal large language model (LLM) based on Qwen3.5. The model is prompted and fine-tuned to replicate choice probabilities across a large training corpus of multiple-choice items containing both image and text stimuli, conditioned on a labeled set of student ability levels. By learning to reproduce the systematic error patterns of students across a discrete range of abilities, the LLM implicitly captures the underlying response probabilities encoded in the 3PL and MCM curves. This allows us to accurately approximate item difficulty on a held-out test set directly from the model's predicted option probabilities. --- ### The CASE Framework: A Multi-Disciplinary Control Architecture for Governing Enterprise Agentic AI Authors: Srinivas Telukunta, Georgios Nektarios Lilis, Lucio Baron Published: 2026-08-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.10153v1 Enterprises are deploying autonomous AI agents faster than they can govern them, and prevailing approaches stretch a single discipline, typically DevSecOps built for deterministic automation, across every scale of agency. We argue that agentic AI governance is four problems, not one, each with a mature governing science. The CASE framework assigns Control theory to the individual agent (intent as setpoint, guardrails as feedback, evaluation as observation), complex Adaptive systems theory to agent collectives (where emergence makes single-agent assurance non-compositional), Supervisory cybernetics to human-agent teams (where the Law of Requisite Variety shows unaided human oversight fails structurally), and Engineering operations to fleets (extending error budgets to decision quality so autonomy becomes a controlled variable). We formalize each layer, derive cross-layer coupling conditions, including a zero-touch deployment paradox where excellence at one-layer strains the others, and trace twenty-plus enterprise controls to their classical constructs. Three empirical studies validate the thesis: 82 percent of documented production agent failures are multi-layer trajectories; none of 22 ecosystem tools offers full Layer 2 (emergence) coverage; and all 35 scored public deployments fall in the lowest maturity band. We name this mismatch, risk realized at the emergence layer against capability barely offered and practice absent, the Emergence Gap. A five-level maturity model with a non-compensatory bottleneck-weighted index and assessment instrument operationalizes CASE as a scientific rather than process maturity model, grounded in production enterprise agentic platforms. As EU AI Act Article 14 makes effective human oversight a legal requirement, only architectures satisfying requisite variety can make oversight real rather than ceremonial. --- ### REATS: LLM Reasoning-based Ensemble Learning for Adaptive Time Series Forecasting Authors: Xu Zhang, Chang Xu, Hui Sun, Nan Ma, Zijian Zhang, Peng Wang, Wei Wang, Li Zhao Published: 2026-08-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.10149v1 Due to the diversity of real-world time series, no single forecasting model consistently dominates across all samples. Ensemble learning addresses this by combining complementary model strengths, yet existing methods rely on fixed rules or black-box models based solely on numerical inputs, failing to leverage LLM reasoning for interpretable weighting decisions. We propose REATS, which leverages LLM reasoning capabilities as an intelligent ensemble router that jointly processes textual temporal pattern descriptions and numerical features to produce interpretable, sample-adaptive ensemble weights through chain-of-thought reasoning. To enable effective LLM-based ensembling, we study its key design choices and propose: (i) a structured input pipeline that transforms raw time series into hybrid textual--numerical representations with fixed token cost, enabling rule-based chain-of-thought construction without API dependency, augmented with retrieved similar-sample priors; (ii) a diverse multi-row weight supervision scheme coupled with a token-efficient percentage-table format that reduces numerical complexity and mitigates LLM hallucinations; and (iii) a two-stage fine-tuning framework combining SFT with GRPO, where a reciprocal reward mapping transforms the continuous unbounded MSE gap into bounded signals with amplified near-oracle sensitivity, addressing the uniform sensitivity and outlier-dominated advantage compression inherent in naive reward designs for regression-based GRPO. Experiments on eight benchmarks demonstrate that REATS outperforms competitive ensemble baselines while providing natural language explanations and demonstrating strong transfer learning and out-of-domain generalization to unseen candidate models. --- ### The Evaluation Protocol Determines the Result: An Independent Reproduction of LeWorldModel on TwoRoom Authors: Joyjeet Singh Published: 2026-08-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.10145v1 LeWorldModel trains a latent world model with a prediction loss and a single anti-collapse regulariser, and reports approximately 87% of goals reached on TwoRoom, its simplest diagnostic environment. We reproduce that result by independent reimplementation on roughly $25 of rented compute, with all evaluation on one laptop CPU. We reach 94.0% at the repository's evaluation goal offset, against 84.0% for the authors' own released checkpoint measured under our protocol on identical episodes, and we reproduce the reported representation result directly (position probe Pearson r = 0.9988 against a reported 0.996). Reaching that point required four conventions that determine the outcome and appear in no released configuration file: dense action gathering across a frameskip block, a programmatically-set action-encoder width, ImageNet pixel normalisation, and action z-scoring. A reproducer following the released configurations alone obtains a model whose predictor cannot converge. The evaluation protocol is itself contested by the released material. The paper's appendix and the repository's configuration specify different goal offsets and step budgets; on the authors' own weights these yield 14.0% and 84.0%, and only the configuration's values reproduce the reported figure. On fifty identical episodes, changing nothing but how the goal is constructed moves that checkpoint from 84.0% to 8.0%. Two findings generalise. One-step prediction accuracy does not predict long-horizon planning success: across three checkpoints spanning a sevenfold range in prediction error, including the authors' own, it orders short-horizon success monotonically and fails to order long-horizon success at all. And a batch normalisation layer inflated our reported validation loss by up to a factor of 300, concealing a training loss that was flat throughout. --- ### SeFoRA: Sketch-Aggregated Federated Low-Rank Adaptation with Heterogeneous Client Ranks Authors: Yue Xia, Tayyebeh Jahani-Nezhad, Mayank Bakshi, Rawad Bitar Published: 2026-08-10 Categories: cs.LG, cs.DC arXiv: https://arxiv.org/abs/2608.10144v1 We consider federated parameter efficient fine-tuning of large neural networks with low-rank adaptation (LoRA,~Hu et al.\ 2022). Combining LoRA with federated PEFT introduces challenges absent from either setting alone: clients may use different LoRA ranks, making their factor matrices dimension-incompatible, and factor-wise averaging suffers from a bilinear mismatch. We propose SeFoRA, a sketch-aggregated federated LoRA algorithm in which each client transmits a linear sketch of its local updates, enabling direct aggregation at the federator. As a result, SeFoRA alleviates the bilinear mismatch, and allows for aggregation in a small subspace of the full model. We introduce a rank-homogeneous version called SeFoRA-Ho which allows for direct adapter aggregation in this setting. We prove convergence to a neighborhood of the first-order stationary point at rate $\cO(1/T)$ for the rank-homogeneous setting. Numerical experiments on fine-tuning RoBERTa-Large on GLUE datasets show how our algorithms outperform the state-of-the-art. --- ### The Parser Already Knows: Lightweight Bias Correction in Constrained Decoding Authors: Işıl Özgü, Yaoxuan Wu, Guy Van den Broeck, Miryung Kim Published: 2026-08-10 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2608.10137v1 Grammar Constrained Decoding (GCD) forces Language Models (LMs) to produce syntactically valid outputs by masking out non-conforming tokens at each step. However, rigid masking distorts the model's underlying probability distribution, often biasing generation toward valid but suboptimal outputs. While online sampling restores this distribution, it requires computationally expensive iterative resampling. As a result, existing methods force a compromise between output quality and inference latency. Our key insight is that the internal parser and lexer states inherently maintained during incremental parsing already encode future grammatical validity -- exactly the information required to restore the LM's true distribution. We propose a lightweight, offline-trained logit correction conditioned on this syntactic and lexical state together with candidate next tokens. Because these states are already computed as a necessary part of incremental parsing for masking, extracting them adds negligible overhead while leaving the base LM's weights completely untouched. Across several grammars, this correction substantially closes the gap between the masked distribution and the LM's true distribution, consistently outperforming both masking and online sampling. Even its lightest variant, which relies on the candidate next token alone, still matches or exceeds both baselines: the next token itself carries an implicit lookahead, much like how parsers commonly use a lookahead token to resolve ambiguous decisions. By restoring the probability mass that masking removes, it reconciles the LM's probabilistic integrity with grammar conformance. --- ### P3CA: Encoder-Agnostic Interpretation of Vision Foundation Model Embeddings via Spatial Probing Authors: Amoon Jamzad, Dilakshan Srikanthan, Faranak Akbarifar, Nooshin Maghsoodi, Parvin Mousavi Published: 2026-08-10 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2608.10131v1 Vision foundation models are increasingly used as reusable encoders in medical image computing, yet their high-dimensional spatial embeddings are difficult to inspect beyond downstream task performance or global dimensionality reduction. We propose position-prompted PCA (P3CA), an encoder-agnostic method for local probing of channel-rich spatial tensors. Given a user-selected spatial prompt, P3CA estimates the feature normalization and dominant covariance directions within that region, then applies the resulting projection to the full tensor to visualize where locally informative directions are expressed. This produces a region-conditioned representation lens without modifying the encoder, retraining, or requiring task-specific labels. We implement P3CA in EmbedVision, an interactive 3D Slicer-based workflow, and evaluate it across natural images, colorectal pathology foundation-model embeddings, and spatial transcriptomic tensors. Across these settings, prompted projections reveal local structure suppressed by global PCA, improve prompt-matched pathology discrimination from frozen three-dimensional projections, and support comparison between learned and measured spatial representations. --- ### Procedural Fairness Failures in RLHF from Preference Averaging Authors: M P V S Gopinadh, Karthik Kamuju, Kummari Avinash, John Joshua, Srinivasa Raju Rudraraju Published: 2026-08-10 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.10126v1 Reinforcement Learning from Human Feedback (RLHF) aggregates heterogeneous preferences into a single reward model, assuming preference homogeneity. When preferences are heterogeneous, this aggregation induces a procedural fairness failure where majority preference groups dominate reward learning while minority preferences are systematically under-represented. This work defines procedural fairness in alignment as preserving distinct preference signals during reward modeling and shows that standard RLHF violates this via preference averaging. Preference-Aware RLHF (PA-RLHF) is introduced, separating optimization across preference modes at the reward learning stage. In a controlled setting, PA-RLHF improves overall alignment accuracy from 46.9% to 67.9% and reduces the fairness gap between best and worst aligned groups from 15.9 to 9.6 percentage points. These results show that procedural fairness failures in alignment can arise from structural design choices in reward learning, even in controlled, noise-free settings, with direct implications for large language models and agentic systems, where biased reward models can compound inequities across sequential decisions. --- ### ChronoSSM: Training for Temporally Aware Representations in Autoregressive State Space Models Authors: Adrien Schoen, Nachiketa Ratnakar Patil, Arjun Bhagoji, Francesco Bronzino Published: 2026-08-10 Categories: cs.LG, cs.NI arXiv: https://arxiv.org/abs/2608.10120v1 Modern sequence models, from Transformers to State Space Models, have enabled powerful generative modeling across diverse domains, yet they are typically trained to predict what happens while treating when it happens as a secondary concern. In data-mining settings where events are associated with explicit timing information, this separation can limit temporal reasoning, anomaly detection, and faithful reconstruction of event chronology. A common strategy is to treat timing as an auxiliary signal, training a separate timing model using representations learned solely for event prediction. However, this two-stage approach implicitly assumes that representations optimized for event prediction already contain sufficient temporal structure. We introduce ChronoSSM, an autoregressive State Space Model (SSM) that jointly models events and timestamps with a shared backbone trained using combined token and temporal generation objectives. We compare the joint regime, where temporal supervision updates the backbone, with the two-stage regime, where timing is learned only using the frozen event representations. Across four domains spanning dense and partial timestamp supervision, joint training consistently makes inter-arrival information more recoverable from frozen representations without any systematic degradation in content-generation quality overall. Our results show that temporal supervision can produce more temporally informative representations without materially degrading autoregressive event modeling. --- ### PERCEPT: A Corpus for POS Tagging and Analysis of Persian-English Code-Mixing Authors: Ghazal Kalhor, Zahra Jafari, Amirarsalan Shahbazi, Behnam Bahrak Published: 2026-08-10 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.10109v1 Social media has become a major venue for multilingual communication, where users frequently mix multiple languages within a single utterance. Although code-mixed corpora have been developed for several language pairs, Persian-English code-mixing remains relatively underexplored. Existing Persian resources lack Universal Dependencies (UD) part-of-speech (POS) annotations for code-mixed words, limiting both linguistic analyses and the development of syntax-aware NLP models. To address this gap, we introduce PERCEPT, the first publicly available large-scale Persian-English code-mixed corpus annotated with Universal Dependencies POS tags for code-mixed words. The dataset comprises 6,800 posts collected from X, Instagram, and Digikala. We further present an LLM-assisted annotation framework that automatically assigns POS tags and document-level topics. Human evaluation demonstrates high agreement between the automatically generated annotations and gold annotations, confirming the reliability of the annotations. Using PERCEPT, we conduct the first comprehensive linguistic analysis of Persian-English code-mixing across multiple social media platforms. Our analyses reveal that nouns are the predominant category for code-mixed words, while the distributions of other POS categories vary across platforms. We further find that the positional distribution of code-mixed words is remarkably consistent across platforms, whereas the triggering effect is substantially more pronounced in Digikala. PERCEPT is publicly available at https://github.com/kalhorghazal/PERCEPT. --- ### MESA:Task-Adaptive Multi-Structure Evidence Selection for Long-Horizon Agent Memory Authors: Beidi Zhao, Yaoqi Chen, Yuru Feng, Menghao Li, Qianxi Zhang, Baotong Lu, Jianan Lu, Zhirui Wang, Xinjiang Wang, Shusen Xu, Zengzhong Li, Xiaoxiao Li et al. Published: 2026-08-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.10108v1 Long-horizon agents accumulate trajectories spanning hundreds of interleaved reasoning, action, and observation steps, where answering a query may depend on evidence buried far back in the history. External memory stores such trajectories as structured representations, yet each structure provides a distinct and incomplete view. Existing multi-memory systems either read a fixed set of structures for every query, inflating context and introducing noise, or route each query to a single structure, preventing the composition of complementary evidence. A controlled analysis on AMA-Bench shows that the optimal memory configuration is typically neither a single structure nor the full union, but a tailored composition of multiple structural memories that varies with query and task demands. Motivated by these findings, we formulate structure-level dynamic selection: selecting and fusing a query-adaptive subset from a library of specialized memory structures. We propose MESA (a Multi-structure Evidence Selection framework for long-horizon Agent), which builds five complementary structure views of each trajectory and learns from end-to-end answer-level feedback to select and fuse a query-specific subset for a frozen answer model. To learn under this weak supervision, MESA employs harness optimization with prior-guided search and UCB-guided scheduling to balance exploration and exploitation. On AMA-Bench, MESA outperforms the strongest baseline by 8.5% while using 41% fewer evidence tokens than the all-structure alternative. --- ### 4D-WAM: 4D Consistent World Modeling for Autonomous Driving Authors: Jiacheng Fu, Yibo Yuan, Meng Tian, Yue Li, Jiangtong Zhu, Jianhua Han, Yueyi Zhang, Jianwu Fang, Jianru Xue, Hang Xu, Zhiwei Xiong Published: 2026-08-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.10107v1 Emerging World-Action Models (WAMs) have demonstrated promising performance in autonomous driving by jointly modeling future driving scene evolution and trajectory planning. However, existing WAMs are typically trained with video data, which is only 2D projections of the underlying 4D driving scene. Consequently, WAMs fail to understand and capture the structure of 4D scenes and thus generate visually plausible yet 4D inconsistent future predictions that mislead downstream planning. To alleviate this issue, we present 4D-WAM, a model that leverages geometric foundation models for training-time supervision to enable 4D consistent world modeling. Specifically, we feed WAM-predicted future frames into a geometric foundation model, and use 4D-aware responses to define a 4D consistency loss. This loss encourages the model to understand, represent, and predict physically consistent 4D scenes during training, without additional inference cost. Moreover, we identify an early-decision phenomenon in WAMs and propose a decision-oriented timestep sampling strategy that emphasizes supervision at early, high-noise stages, where driving decisions are primarily formed. By propagating 4D supervision to this critical decision-formation phase, the proposed strategy further improves trajectory planning. Extensive experiments demonstrate that 4D-WAM effectively models 4D consistent scene evolution and achieves state-of-the-art performance on challenging NAVSIM-v1 and NAVSIM-v2 benchmarks. --- ### Hand-Written PTX Tensor-Core GEMM Kernels: A Multi-Precision Study on NVIDIA L4 Authors: Matt J. Borowski, Blazej Osinski Published: 2026-08-10 Categories: cs.DC, cs.AI arXiv: https://arxiv.org/abs/2608.10103v1 High-performance Tensor Core kernels rely on a low-level PTX pipeline built from asynchronous data movement with cp.async, warp-level matrix loads with ldmatrix, and matrix multiply-accumulate operations with mma.sync. However, most application code accesses Tensor Cores indirectly through the WMMA C++ API. This paper asks a focused, practical question: when does replacing WMMA with hand-written PTX actually pay off? To answer this question, we conduct a controlled, single-GPU study on an NVIDIA L4 GPU (Ada, SM89), comparing double-buffered WMMA baselines with a family of hand-written PTX GEMM kernels across FP16, INT8, and INT4 arithmetic and square problem sizes from $N=512$ to $N=8192$. Every kernel is profiled with Nsight Compute across the full metric set, and PTX speedups are reported relative to the corresponding same-precision WMMA baseline. Hand-written PTX provides no end-to-end speedup for FP16, because its instruction-level gains are offset by operand-packing overhead. In contrast, the PTX kernels achieve consistent speedups of 1.4x-1.8x for INT8, driven primarily by lower instruction counts and better global-memory coalescing, and 2.9x-4.3x for INT4, where native mma.sync.m16n8k64.s4 execution avoids the software-emulated sequence used by the WMMA path. Relative to the FP16 WMMA baseline, the best quantized kernels reach 34.4x (INT8) and 98.7x (INT4) at $N=8192$. Across these experiments, occupancy is a poor predictor of throughput. For large matrices, performance instead tracks memory-system behavior -- particularly global-load coalescing and DRAM-active cycles -- more closely than Tensor Core utilization. These results identify the precisions and operating regimes in which the additional complexity of hand-written PTX is justified. --- ### Exploring Semantic Stability Across Reviews in the Linux Kernel Authors: Lucas Ciziks, Paulo Meirelles, Marco Aurélio Gerosa Published: 2026-08-10 Categories: cs.SE, cs.AI, eess.SY arXiv: https://arxiv.org/abs/2608.10101v1 Code review is credited with substantially changing a patch's code between its first submission and the version that eventually lands. However, prior work typically studied only the final merged patch without comparing it to the first submission. We present a function-level measurement that tracks 10,117 trajectories (each function followed across the numbered revisions of one patch series) through the patch history of the Linux IIO subsystem, comparing similarity scores against unrelated function pairs as a baseline. A naive reading yields near-total similarity, but this is largely an artifact of composition: 75.3% of tracked trajectories are never textually modified between versions, contributing a trivial 100% similarity that inflates the headline. Restricting to the trajectories with a real edit, semantic purpose is still largely preserved (mean similarity 0.990 vs. a 0.909 baseline), but drift appears to concentrate in the first review round mainly because later rounds contain more functions that nobody touched, not because edits become more conservative over time. After controlling for it, a statistically detectable but small residual effect remains. This points to an open question: whether near-ceiling similarity reflects preserved purpose or a measurement tool that cannot detect the significance of small, localized edits. We present this work as a first look and outline next steps. --- ### Deciding When to Switch: E-Processes for Adaptive Minimax Training for Generative Adversarial Nets Authors: Hyunjoo Kim, Sicheng Wu, Agastya Venkatraman, Guang Lin, Sehwan Kim Published: 2026-08-10 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2608.10096v1 Modern data science increasingly gives rise to hypothesis-testing problems that are not naturally formulated in terms of parameters within prespecified statistical models. One important example is the dynamic evaluation of optimization algorithms, where decisions must be made during training about whether further updates remain beneficial or the algorithm should switch to a different phase. This issue is particularly relevant in stochastic min-max optimization. Generative adversarial networks (GANs) provide a canonical example, as their training requires repeated decisions about when to switch between discriminator and generator updates, yet existing methods typically rely on fixed update ratios or heuristic criteria. We formulate this switching problem as sequential hypothesis testing and develop an e-process-based adaptive training procedure. During discriminator updates, one e-process tests the null that the discriminator-induced separation between the empirical data distribution and the generator law remains below a target level. During generator updates, with the discriminator fixed, a second e-process tests the reverse null that this separation remains above a refresh level. Conditional on the observed training sample, we prove that fresh empirical indices and latent draws yield conditional e-values that can be accumulated into e-processes, providing anytime-valid Type I error control under adaptive model updates and data-dependent switching. Across multimodal synthetic distributions and image benchmark datasets, the proposed method matches or outperforms the best fixed-ratio baselines under several widely used GAN objectives. --- ### Signpost Watermarking: Joint Optimization for Visual Watermark Coexistence Authors: Shruti Agarwal, Vishal Asnani, John Collomosse Published: 2026-08-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.10091v1 We present a method for training imperceptible visual watermarks to coexist with other such watermarks. Recent work has shown that independently trained image watermarking models can coexist with surprisingly limited interference, enabling watermark ensembling. However, this coexistence is a serendipitous property rather than an explicit optimization objective, leaving interference uncontrolled and potentially reducing decoding robustness or visual quality. We first show empirically that the same coexistence property extends to video watermarking. We then show that both image and video watermarks can be trained with a decoder-aware objective to improve coexistence. Our results suggest a practical path to signpost watermarks that indicate the presence of independently deployed provenance watermarking systems, supporting layered provenance signaling for content authenticity and rights. --- ### CHORUS: Complementary Experts for High-Coverage Testbench Stimulus Generation Authors: Hejia Zhang, Sheng Lu, Zhongming Yu, Chia-Tung Ho, Brucek Khailany, Jishen Zhao Published: 2026-08-10 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.10090v1 Large language models (LLMs) have advanced code generation, where executable feedback provides a more reliable learning signal than textual imitation alone. Hardware verification is an important application of code generation and accounts for a substantial fraction of modern chip design effort, with high-coverage testbench stimulus generation as a key task. We present CHORUS, a post-training framework that pushes performance beyond what a conventional supervised fine-tuning (SFT)-to-reinforcement learning (RL) pipeline achieves. CHORUS builds on two observations. First, staged SFT produces behaviorally diverse checkpoints, and dense-reward RL turns them into strong experts with comparable aggregate performance but distinct task-level strengths. Second, these complementary strengths can be exploited through either training-free model merging or further post-training to outperform the best individual expert. By consolidating the resulting specialists into a single 4B model, CHORUS achieves 88.0% Pass@1 on CVDP-ECov, outperforming DeepSeek-R1 (671B) by 13.5 percentage points. --- ### Status Association Does Not Reliably Predict Decision Leakage Authors: Abdullah X Published: 2026-08-10 Categories: stat.AP, cs.AI, cs.CY, cs.LG arXiv: https://arxiv.org/abs/2608.10089v1 Bias evaluations often move too quickly from evidence that a model encodes a social association to claims that the same association will alter consequential decisions. We test whether that inference is warranted using Chilean surnames as controlled socioeconomic probes. We evaluate eight frozen model-provider cells on 1,032 prompts each, yielding 8,256 verified primary responses. The design separates forced latent association from matched consequential decisions across academic selection, professional hiring, research fellowship selection, and legal-aid intake. Elite-coded surnames received higher forced high-status probability mass than common surnames in seven of eight models and higher mass than rare-frequency controls in all eight. Yet elite-minus-common decision effects were close to zero for most systems. Five models were statistically equivalent within a predeclared (Plus-Minus)0.10 standard-deviation margin, while the remaining three were imprecise or borderline, with no consistent elite advantage. Association strength did not reliably predict decision leakage across models (r = 0.201, p = 0.633) or across frozen surname-pair-by-model cells (r = 0.065, p = 0.565). The central result is a measurement dissociation: latent social association and consequential treatment are empirically distinct constructs. Evaluations should measure the transition from association to action directly. --- ### When Repository Labels Are Not Image-Level Truth: A Supervision Auditing Framework for Chest Radiograph AI Authors: Yesika Alexandra Agudelo-Londoño, Jhon Wilmer Pino-Román, Brahian Carrera Rodríguez, José Miguel Castañeda-Bedoya, Juan Pablo Gómez-López, Aura C. Puche-Sarmiento, Niharika S. D'Souza, Juan Sebastian Osorio-Valencia, Jon E. Duque-Grajales, Jazmín Ximena Suárez-Revelo, Jorge Mario Vélez-Arango, Gabriel Castrillón Published: 2026-08-10 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2608.10084v1 Public chest X-ray repositories are widely used to train medical AI systems, yet their labels are typically extracted from radiology reports rather than verified directly on images. As a result, repository labels are often treated as image-level ground truth without validating whether they reflect what is actually visible in the radiograph. We introduce Repository Supervision Auditing (RSA), a framework that evaluates repository-derived labels against expert image-level annotations before model development. Using cardiomegaly in MIMIC-CXR as a case study, RSA compares repository labels with radiologist-reviewed image annotations, characterizes disagreement sources, and builds a curated cohort for deployment-oriented evaluation. Repository-derived cardiomegaly labels showed near-zero agreement with expert image-level assessment, identifying only 1% of expert-confirmed cases. Most discrepancies resulted from non-mention rather than explicit report negation, with expert-confirmed cardiomegaly identified in nearly half of studies assigned a repository-derived No Finding label. Using the resulting expert-curated cohort, a DenseNet121 model achieved a test ROC-AUC of 0.853. These findings show that repository labels may not reliably represent image-level truth and highlight supervision auditing as a critical step for developing trustworthy medical imaging AI. --- ### LEGO: Leveled Language Gaussian Splatting Authors: Yuning Peng, Haiping Wang, Yuan Liu, Yipeng Lu, Zhen Dong, Bisheng Yang Published: 2026-08-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.10057v1 We introduce LEGO for advanced open-vocabulary scene understanding. Beyond basic concept recognition, its core innovation lies in capturing the intrinsic semantic hierarchies within the scene, such as the "flowerpot -> bouquet -> bud -> petal" lineage. While foundation models like SAM can identify multi-granular structures in 2D, their partitions are strictly perspective-bound and lack cross-view consensus. LEGO self-adaptively re-grades volatile multi-view SAM granularities into a unified, 3D-consistent hierarchy. This provides precise supervision for the structurally coherent, multi-level segmentation of 3D scenes. By grounding these segments with CLIP embeddings, LEGO recovers open-vocabulary semantic logic across hierarchical levels. Furthermore, by incorporating spatial relationships, we elevate these segments into level-wise language scene graphs, effectively empowering Large Language Models to perform complex, context-aware spatial reasoning and precise visual grounding. Experimental results demonstrate that LEGO establishes new state-of-the-art performance across both promptable and open-vocabulary 3D segmentation benchmarks, exhibiting advanced hierarchical scene decomposition and context-aware spatial reasoning. --- ### Navigating the Proximity-Safety Balance: Constraint Decomposition for Human Following in Pedestrian Crowds Authors: Shiting Gong, Jianpeng Yao, Jinfeng Wang, Marco Pavone, Jiachen Li Published: 2026-08-10 Categories: cs.RO, cs.AI, cs.LG, eess.SY arXiv: https://arxiv.org/abs/2608.10056v1 Following a target human in crowded environments involves an inherent conflict between staying close to the target and navigating safely among surrounding pedestrians and obstacles. This conflict becomes more severe in dense scenarios, where aggressive following risks collisions and conservative margins lead to target loss, especially when pedestrian behaviors are unfamiliar or unpredictable. Existing reinforcement learning (RL) methods typically encode these competing objectives into a single dense reward, but the resulting proximity-safety balance is implicit and difficult to adjust across conditions. To address this, we decompose the human-following task into a sparse task reward and independent cost constraints within a multi-constraint RL formulation, where each constraint is managed through cost thresholds with direct behavioral meaning rather than implicit reward weight ratios, allowing explicit and tunable control over the trade-off. We further quantify the prediction uncertainty of human motions and integrate these estimates into the RL costs to enhance safety under unpredictable conditions. Extensive experiments across both in-distribution and out-of-distribution settings demonstrate that our method achieves an effective proximity-safety balance compared to baselines. Real-robot deployment further validates the feasibility of our method in real-world scenarios. More details are available on our project page: https://nav-ps-balance.github.io/. --- ### Observational Policy Ranking for SMB Financial Guidance from Multi-Action Accounting Logs Authors: Shrutendra Harsola, Vignesh Subrahmaniam, Vikas Raturi, Kamalika Das, Xiang Gao, Kratika Gupta, Ruocheng Guo, Padmaja Jonnalagedda, Ananya Pramod, Sricharan Kumar Published: 2026-08-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.10050v1 Small and medium-sized businesses need timely financial guidance, yet historical accounting logs record self-selected and often co-occurring business changes rather than randomized recommendations. We formulate this setting as observational policy ranking: from pre-decision financial information, a policy selects one of 34 ledger-derived business-change categories for a target financial KPI. Using 85,078 company-month observations from 7,505 firms, we introduce Covariate-Adjusted Residual Policy Learning (CAR-PL), an action-wise R-learner that operates directly on multi-hot logs and regularizes selection by observational support. We compare CAR-PL with an uplift T-Learner, a conservative contextual value model, a zero-shot LLM, and non-personalized references on company-disjoint held-out firms under a shared model-assisted scoring rule. CAR-PL has the highest Gross Profit point estimate (0.084), the T-Learner has the highest Revenue point estimate (0.085), and the contextual value model has the highest Quick Ratio point estimate (0.062). CAR-PL and the T-Learner are not statistically separated on either growth KPI in matched company-clustered comparisons, while CAR-PL selects 33-34 categories and produces less concentrated selections across the catalog. Outcome-model-only scoring retains the same KPI-level point-estimate leader or top pair, and category rankings remain similar when the all-zero treatment reference is replaced by the most common training co-action pattern. These findings support objective-specific ranking of SMB financial guidance from multi-action accounting logs. --- ### Physics-Informed Machine Learning in Prognostics and Health Management: A Systematic Literature Review Authors: Christopher Braun, Julian Raible, Marco F. Huber Published: 2026-08-10 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.10047v1 In modern industry, keeping complex systems reliable, safe, and efficient hinges on Prognostics and Health Management (PHM). Machine Learning (ML) has largely driven advancements in diagnostics and prognostics, yet purely data-driven models face inherent limitations, such as poor generalization, an inability to infer causal relationships, and a lack of interpretability. Physics-Informed Machine Learning (PIML) helps mitigate these limitations by incorporating prior physical knowledge directly into the ML pipeline, thereby fostering growing interest in its application to PHM. This work investigates how PIML is being leveraged in the context of PHM through a systematic literature review of 212 studies. The review introduces a four-class classification scheme, consisting of observational bias, inductive bias, learning bias, and hybrid approaches, and further categorizes studies by PHM task. Across all four classes, the reviewed studies consistently demonstrate improved predictive performance over conventional baselines across a broad range of assets, although the literature is heavily skewed toward lithium-ion batteries and bearings, and dominated by problem-specific solutions. Overall, the review indicates that physics-informed approaches already provide tangible benefits, whereas claims of improvements concerning some of the aforementioned limitations lack sufficient supporting evidence. Future research should prioritize transferable design patterns, benchmarks comparing integration strategies, and uncertainty-aware models that are lightweight and robust enough for online deployment in real-world settings. --- ### Detecting Soft Skills in ML Engineering Roles CVs Authors: Aidin Azamnouri, Nouran Ayad, Justus Bogner, Stefan Wagner Published: 2026-08-10 Categories: cs.LG, cs.CY arXiv: https://arxiv.org/abs/2608.10046v1 Soft skills shape collaboration among ML engineers, data scientists, and software engineers building ML-enabled systems, yet what we know about them comes almost entirely from the demand side. Job advertisements, surveys, and hiring manager interviews capture what employers ask for. How candidates themselves articulate these competencies has not been studied, and existing CV-mining work is both keyword-based, so it cannot see skills conveyed through narrative, and descriptive, reporting frequency rankings without testing whether group differences exceed sampling variation. We close both gaps. Using a balanced corpus of 300 curated CVs spanning the three roles, we extract explicitly listed and implicitly narrated soft skills with an LLM-based pipeline validated against a human-annotated ground truth, a distinction that existing extractors were not designed to make. We then convert the demand-side literature's claims into 13 falsifiable hypotheses about role signatures, seniority progression, and disclosure style, and test them with effect sizes under family-wise error control, so that candidate-side data can corroborate or contradict the demand-side account rather than merely illustrate it. Eleven hypotheses are supported, one partially, and one refuted. Candidates disclose soft skills through narrative rather than keyword lists by roughly three to one, and most so for the competencies employers value most: leadership, coordination, and mentoring (88-96% narrative). Seniority nearly triples the odds of articulating leadership. That competency, assumed universal in prior work, is articulated by software engineers at half the rate of their peers. Technical candidates do articulate soft skills, but a keyword-based screening systematically misses them. --- ### Finding the Signal in the Spam: Jointly Learning Rewards and Worker Reliability from Pairwise Comparisons Authors: Kaustubh Shivshankar Shejole, Tanish Agarwal, Arpit Agarwal, Avishek Ghosh Published: 2026-08-10 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.10045v1 The problem of learning from pairwise comparisons has been widely studied across many domains such as recommendation systems, social choice, and more recently, fine-tuning large language models. In this problem, the goal is to learn item rewards based on pairwise comparisons between them. In many scenarios, these comparisons are elicited from crowdworkers using platforms such as Amazon Mechanical Turk, Scale AI, etc. However, crowdworkers are often unreliable due to limited domain knowledge or revenue-maximizing (spamming) behavior. In this work, our goal is to understand whether worker reliability (competency) can be learned jointly with item rewards. To this end, we adopt the Boltzmann-rational model for pairwise comparisons, which extends the Bradley-Terry-Luce model by incorporating worker competencies. We derive an EM-based algorithm for learning under this model by introducing Polya-Gamma latent variables to transform the logistic likelihood into a conditionally Gaussian form, enabling tractable optimization and leading to a simplified $Q$ function in the E-step of the algorithm. This technique allows us to reduce our formulation to a matrix sensing problem, using which we establish theoretical convergence guarantees for our algorithm. We conduct extensive experiments on real-world and synthetic datasets. These experiments demonstrate the advantages of using our algorithm over several baselines and confirm its strong robustness to both spammers and adversarial workers, highlighting its practical effectiveness in realistic crowdsourcing and reward learning settings. The code and data is publicly available at https://github.com/KaustubhShejole/BoRa_EM. --- ### UserToolBench: A User-Profile-Hidden Benchmark for Personalized Decision Making in Tool-Use LLMs Authors: Xuexiong Yin, Zechuan Chen, Yongsen Zheng, Yuxiang Zhang, Jingyuan Yang, Bin Wang, Yubin Wang, Keze Wang Published: 2026-08-10 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.10042v1 Tool-use LLMs are increasingly asked to act on users' behalf, but existing benchmarks usually focus on profile recall, style imitation, generic tool use, or response-level personalization. We introduce UserToolBench , a benchmark for personalized decision making in tool-use LLMs. UserToolBench tests whether a model can infer latent user preferences from interaction history, recognize when clarification is needed, and produce user-aligned tool-call trajectories under incomplete information. The benchmark is built from privacy-sanitized real interaction traces and combines structured persona profiles, public API-style tool ecosystems, and long-horizon multi-turn trajectories. It includes 10 user profiles, 36 tool sets, 1,065 turns, 170 unique tools, and evaluation-focused task types covering lack-of-information, single-tool, and multi-tool settings. Experiments with strong tool-use LLMs show that current models still have difficulty with personalized delegation. Multi-tool coordination, missing-constraint inference, and long-horizon behavioral consistency remain major bottlenecks. These results suggest that personalization evaluation should move beyond asking whether outputs sound user-specific and instead ask whether LLMs make correct decisions for the users they represent. --- ### FlowScout: From Execution Feedback to Reliable Tool-Using Agent Workflows Authors: Shuo Hao, You Lu, Bihuan Chen, Xin Peng Published: 2026-08-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.10039v1 Agentic workflows have become an important abstraction for building reliable LLM-based automation systems by organizing large language models (LLMs), tools, and control logic into explicit execution structures. However, constructing high-quality agentic workflows remains largely manual and requires substantial domain expertise. Recent studies have explored automatic agentic workflow generation from historical task-solving records, but they mainly produce LLM-centric workflows, where real tool executions are abstracted and simulated by LLM nodes, limiting the usability and stability of generated workflows. To address these limitations, we propose FlowScout, an execution-guided framework for generating tool-integrated agentic workflows from historical task-solving records. Specifically, FlowScout represents an agentic workflow as a directed graph composed of LLM nodes, tool-calling nodes, and dependency edges. It first mines a common tool coordination skeleton from historical records to construct an initial workflow, and then refines the workflow topology through Monte Carlo tree search guided by execution feedback. We evaluate FlowScout on four representative task domains and compare it with three baselines, i.e., PM4Py, ReAct and AFlow. Experimental results show that agentic workflows generated by FlowScout improve tool invocation correctness by at least 92.69% and execution quality by at least 17.66% over the baselines, while achieving lower performance variation across repeated runs. --- ### DOCSCHISEL: Adaptive Tool Documentation Optimization Framework for LLM Agents Authors: You Lu, Kun Zhang, Bihuan Chen, Xin Peng Published: 2026-08-10 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.10037v1 Large language models (LLMs) increasingly rely on external tools to accomplish complex real-world tasks, making tool documentation a critical grounding resource for LLM agents. Existing studies mainly focus on improving the tool-use capabilities of LLM agents, while largely treating tool documentation as a fixed input. Although several recent works attempt to optimize tool documentation through rewriting or compression, little is known about how the information contained in tool documentation affects agent performance across different settings. To bridge this gap, we conduct a large-scale empirical study on tool documentation for LLM agents. Our study reveals substantial heterogeneity in the information fields provided by existing tool documentation. Moreover, the effectiveness of different information fields is highly dependent on the task domain, LLM backbone, and agent paradigm, indicating that no fixed tool documentation can consistently generalize across diverse agent settings. Motivated by these findings, we propose DocsChisel, an adaptive tool documentation optimization framework for LLM agents. DocsChisel analyzes failed execution traces of a target LLM agent to identify documentation-related issues, and iteratively optimizes tool documentation by adding, removing, and refining information fields for each tool. We evaluate DocsChisel against two state-of-the-art baselines, i.e., EasyTool and DRAFT. Experimental results show that DocsChisel improves the task success rate of LLM agents by 95.89% over the original tool documentation and by 75.15%, on average, over existing baselines, while incurring limited optimization time and token overhead --- ### Causal Behavioral Evaluation of AI Agents at Scale via Automated Behavioral Science Authors: Soo Yong Lee, Jongha Lee, Jaewan Chun, Hyunjin Hwang, Fanchen Bu, Dongyeong Hwang, Ziv Ben-Zion, Taekwan Kim, Denny Borsboom, Jaemin Yoo, Kijung Shin Published: 2026-08-10 Categories: cs.AI, cs.MA arXiv: https://arxiv.org/abs/2608.10030v3 As AI agents are increasingly deployed in complex and new environments, knowing the conditions that influence their behavior becomes an indispensable step for their reliable and safe deployment. Yet causal behavioral evaluation of AI agents remains manual and labor-intensive. We introduce Abs2Sim and AEROBAT, a system of methods that support causal behavioral evaluation of AI agents via automated behavioral science. Given a user-specified target behavior, the methods automatically execute a full pipeline of behavioral science research---generating hypotheses about the behavior, designing and executing controlled experiments, making behavioral assessments, analyzing the results, and writing reports. For 12 target behaviors, the methods generated and tested 73 hypotheses: designing 1,160 controlled experiments and executing 22,954 simulation rounds in total. Moderate-to-strong statistical evidence emerged for 30 hypotheses, revealing potential modulators of AI behavior. In sum, our results demonstrate that automated behavioral science can extend the reach of behavioral evaluation of AI agents. --- ### Perception Before Supervision: Self-Contained Visual Distillation from Counterfactual Blind Spots Authors: Shravan Venkatraman, Omkar Thawakar, Ritesh Thawkar, Abdelrahman Shaker, Rao Muhammad Anwer Published: 2026-08-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.09931v1 Self-improvement for multimodal large language models (MLLMs) is typically driven by reward-based methods that provide only coarse scalar feedback. Distillation offers a richer alternative through dense token-level supervision, but in the visual domain it usually depends on privileged context constructed using external annotations and tools, or stronger models. We introduce \textbf{CVPD} (Contrastive Counterfactual Visual Process Distillation), which, to the best of our knowledge, is the first fully self-contained framework for dense, on-policy, token-level visual self-distillation for MLLMs. CVPD identifies visual blind spots where zooming into a region changes and sharpens the model's answer distribution, while removing the same region leaves the full-image behavior largely unchanged. Such regions reveal perceptual information that the model can encode but fails to consistently utilize under full-image conditioning. We propose a three-gate Counterfactual Criterion that identifies these regions directly from the model's own responses and converts them into dense contrastive supervision for self-distillation. On Qwen3-VL-8B-Instruct, CVPD outperforms six self-evolving baselines across twelve benchmarks, including methods that rely on external GPT-4o supervision, without a single regression. It achieves gains of $+3.60$ on OCRBench, $+3.38$ on MMStar Fine-Grained Perception, and $+3.08$ on MMStar Logical Reasoning, while maintaining or improving performance on broader multimodal benchmarks. --- ### Beyond Naturalness: Probing Automated Text-To-Speech Evaluators on Linguistically Grounded Dimensions Authors: Oluwanifemi Bamgbose, Simon Rosen, Jash Shah, Lindsay Devon Brin, Hoang H Nguyen, Anke Koelzer, Rachel Hansen, Tara Bogavelli, Fanny Riols Published: 2026-08-10 Categories: cs.SD, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.09930v1 Automated Text-to-Speech (TTS) evaluation methods (Mean Opinion Score (MOS) predictors and Audio Large Language Models (Audio-LLM) judges) are expected to reflect human perception, yet it is unclear how well they capture the distinct aspects of speech that listeners actually perceive. We deconstruct "naturalness" into a linguistically grounded annotation schema spanning 10 distinct perceptual dimensions, and use it to construct the first dimension-level meta-evaluation benchmark for TTS, comprising 860 utterances annotated by trained linguist raters. Results from benchmarking four MOS predictors and four Audio-LLM judges reveal that MOS predictors collapse onto acoustic signal quality, while Audio-LLM judges show selective, prompt-dependent detection that does not generalise across all dimensions. Neither class reliably captures a breadth of linguistically structured speech errors. Our dataset, annotation schema, and evaluation code are publicly released to support more targeted and interpretable TTS evaluation. --- ### Multimodal Model Diffing for Feature Discovery and Control Authors: Hunar Batra, Lachin Naghashyar, Ashkan Khakzar, Philip Torr, Christian Schroeder de Witt, Constantin Venhoff, Ronald Clark Published: 2026-08-10 Categories: cs.CV, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2608.09928v1 Multimodal Large Language Models (MLLMs) exhibit strong visual understanding, yet the internal features that cause these behaviors remain difficult to identify, audit, or control. While applicable to post-hoc inspection, hidden states that are decomposed into interpretable feature directions using sparse autoencoders (SAEs) neither readily isolate which features are changed by multimodal training, nor are they directly useful for targeted control. We introduce MMDiff, a multimodal model-diffing framework that trains multimodal SAEs and turns them into feature-level interfaces for discovering and controlling multimodal behavior. MMDiff supports three uses: (i) feature isolation, by diffing a base-LM SAE against its multimodal-adapted counterpart to identify features altered by multimodal training; (ii) task-specific feature detection, via per-token contrastive firing analysis that isolates causal features; and (iii) feature-level control, by causally removing or steering the discovered feature directions. We train multimodal SAEs for three MLLM families, LLaVA-MORE, PaliGemma 2, and InternVL3.5, and evaluate on visual-spatial understanding, multimodal safety, and OCR. MMDiff discovers sparse, causally specific features whose removal selectively degrades target behaviors by an average of 12% on spatial tasks and 17% on OCR, and reduces attack success rate by 24% on multimodal safety attacks, with no impact on VQA performance. Steering these features improves spatial and OCR accuracy by +3.6% and +1.8% on average over a standard single-layer steering baseline. These results show that multimodal SAEs can serve not only as interpretability tools, but as mechanisms for auditing, steering, and controlling MLLMs behavior toward safer and more capable generations. --- ### Learning How the World Evolves: Extrapolative Video World Models via Latent Dynamics Reasoning Authors: Haodong Li, Shaoteng Liu, Tianyu Wang, Chongjian Ge, Sihui Ji, Jiahan Zhang, Xin Lin, Haolin Lu, Zhe Lin, Manmohan Chandraker Published: 2026-08-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.09926v1 The world evolves following its dynamics, i.e., its laws of motion. However, leading video diffusion models largely fit the pixels without modeling how the pixels transit over time. Thus, they render visually plausible frames but may not accurately obey the laws. To capture the dynamics purely from pixels, we introduce Latent Dynamics Reasoning (LDR). LDR casts the latent transition as an explicit kinematic integration, where the lower-order dynamics are integrated numerically and the model regresses only the third- and higher-order residual that drives the rollout. For this integration to extrapolate better, LDR runs it on a structured latent rather than dense convolutional features. Following PhyWorld, we validate LDR on a controlled white-box physics benchmark spanning five tasks (uniform motion, parabola, collision, bouncing, looming), focusing on out-of-distribution scenarios that reveal whether a model has truly learned the underlying dynamics. LDR extrapolates the learned dynamics far better: the gap between its in- and out-of-distribution error is over 20$\times$ smaller than the video diffusion baseline's, under both single- and joint-task training at 256$^2$ resolution, while using 26$\times$ fewer parameters and running 143$\times$ faster. LDR can even generalize under severe shift: for example, trained only on red balls moving left-to-right, it correctly predicts the motion of a blue square moving right-to-left. To our knowledge, this is the first video world model that extrapolates learned dynamics beyond its training distribution. Project page: https://lat-dyn-reason.github.io/ ---