# Research Papers — 2026-08 · bundle page 106 of 240 > Full-text markdown of papers #5251–5300 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/106 [← Previous bundle](https://aipod.dev/papers/archive/2026-08/page/105.md) [Next bundle →](https://aipod.dev/papers/archive/2026-08/page/107.md) ### Key-Frame Reasoning with SAM3: Third Place Solution for the MeViS-Text Track of the 8th LSVOS Challenge Authors: Ce Bian, Xusheng He, Jinrong Zhang, Canyang Wu, Xianjing Han, Jianlong Wu Published: 2026-08-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.17279v1 This report presents a two-stage, training-free solution for the MeViS-Text track of the 8th LSVOS Challenge. The task requires a model to localize and segment the object specified by a natural-language expression throughout a video. Such expressions often depend on temporal cues, including actions, interactions, directions, and relative positions. Our first stage uses Gemini-3.1 Pro via API to decompose a video-level event into instance-level targets, select a key frame for each target, and generate a discriminative description aligned with that frame. In the second stage, SAM3-agent produces a pixel-level seed mask on the selected frame, and the SAM3 video tracker propagates the mask bidirectionally through the video. Valid instances are grounded and propagated independently before their frame-wise masks are merged. All local SAM3 processing runs on a single NVIDIA GeForce RTX 4090 without task-specific training or model ensembling. Our method ranked third on the challenge test set, obtaining J&F, J, F, N-acc., T-acc., and Final scores of 0.761, 0.7367, 0.7852, 0.8333, 0.9755, and 0.856593, respectively. --- ### When Agents Act on Web3: An Attack-Surface Survey of MCP, Skills, and Tool Calling Authors: Rabimba Karanjai, Yang Lu, Nour Diallo, Wujie Xiong, Lei Xu, Weidong, Shi Published: 2026-08-18 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2608.17275v1 AI agents increasingly act rather than merely read: across the Model Context Protocol (MCP) ecosystem, the share of deployed tools that modify external state has risen from 27% to 65% of tool use. When agents exercise this authority on public blockchains through MCP, skills, and tool calling, the consequences of an attack are governed by the blockchain execution layer rather than by conventional software assumptions. This survey argues that four properties of that layer (irreversibility, signing authority, continuous autonomy, and sequence-level composition) qualitatively change the threat model, turning the recoverable failures of generic agent security into a standing, irreversible loss. We organize the fragmented MCP-security literature into an attack-surface taxonomy, then contribute a Web3 risk-mapping matrix that ties each attack class to its amplified impact, the responsible amplifiers, a representative mitigation, and the residual gap. We synthesize defenses, including emerging blockchain-based mechanisms, and find them improving but insufficient: measured protections stop fewer than 30% of attacks, and model-level safety refuses fewer than 3%. We close by positioning the work against adjacent surveys and deriving a research agenda from the matrix's open cells. --- ### ASI-Bench: At the Dawn of Artificial Superintelligence Authors: Junwei Zhou, Zhen Sun, Binyu Li, Jiangyu Zhou, Yuexi Pan, Hengyu Wang, Honghe Ren, Xiaohan Jia, Xueyang Zhou, Xiaoyu Cao, Yongchao Chen, Yuanning Feng et al. Published: 2026-08-18 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.17271v1 Artificial superintelligence (ASI) requires AI to move beyond mastering existing knowledge toward exploring the unknown, creating new knowledge, and turning new ideas into verifiable results. However, the capabilities of today's AI systems are still largely built on learning, compressing, and applying existing human knowledge. Accordingly, existing benchmarks primarily test whether AI can produce correct answers based on learned knowledge, or whether it can complete tasks under extensive human guidance. We therefore introduce ASI-Bench, the first benchmark to jointly evaluate AI systems' capabilities of innovative exploration and autonomous scientific execution across general research domains, and the first to progressively withdraw human methodological guidance within the same research project to test how far AI can proceed on its own. Built by over 40 experts with the cost of 31,000+ human hours, ASI-Bench contains 60 project-level research tasks across 11 scientific domains and progressively reduces methodological guidance to test whether AI can independently select methods, conduct research, and produce verifiable results. All tasks undergo expert review, AI-assisted auditing, sandbox execution, and scorer validation. Across 18 state-of-the-art agent--model configurations, the average score drops from 50.91 with full methodological guidance to 29.10 with only the method specified and 26.62 when agents must determine the method themselves. This sharp decline shows that current systems remain heavily dependent on human guidance and are still far from autonomously conducting end-to-end, project-level scientific research. ASI-Bench is open to the world. We invite researchers and builders everywhere to contribute new tasks, challenge the limits of today's AI, and help accelerate humanity's collective path toward artificial superintelligence at https://asibench.apexin.ai/submit. --- ### Do LLMs Know a Good Hypothesis When They See One? Logit-Based Energy Scoring Outperforms Prompted LLM-as-Judge for Scientific Hypothesis Ranking Authors: Swati Rajwal, Sanjay Das, Tirthankar Ghosal Published: 2026-08-18 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.17270v1 Large language models (LLMs) are increasingly used for scientific hypothesis generation. However, evaluating generated hypotheses remains a challenge for trustworthy AI-enabled scientific workflows. Existing approaches often use LLMs as judges or rely on semantic similarity, which can favor familiar ideas over novel ones. We propose a logit-based energy scoring method that evaluates hypotheses using a language model's intrinsic confidence rather than comparative judgment. We benchmarked seven language models on 1,323 papers across 12 disciplines. Each paper was paired with its hypothesis and fifteen incorrect alternatives. Intrinsic scoring reached 33.0% Hit@1 pooled across both scorers, compared with 16.6% for prompted listwise ranking. The strongest configuration, a 1-billion-parameter model using logit-based energy scoring, reached 53.1%, though this was the maximum across 14 model-by-scorer combinations selected post hoc. Overall, intrinsic model confidence shows potential for scientific hypothesis evaluation. This study also motivates future research on confidence-based methods for trustworthy AI-enabled scientific discovery. --- ### Understanding Curriculum Learning in Large Language Models via Cross-Difficulty Optimization Dynamics Authors: Zhikai Ding, Ziyi Ye Published: 2026-08-18 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.17268v1 Curriculum learning has been widely adopted in the post-training of large language models by organizing training data from easy to hard. However, its effectiveness varies substantially across reasoning tasks, suggesting that no single curriculum is universally optimal and raising a fundamental question: what determines when curriculum learning works? In this paper, we answer this question by analyzing the optimization dynamics induced by different curriculum schedules. We show that the transfer relationship between different difficulty levels characterizes the optimization dynamics induced by curriculum learning, which in turn explains the effectiveness of different curriculum schedules, and formalize this relationship as Relative Transfer, a principled measure of cross-difficulty knowledge transfer. Based on this measurement, we derive Transfer-aware Dynamic Curriculum Sampling (TDCS), which dynamically adjusts the sampling distribution according to the estimated transfer relationship throughout training. Extensive experiments on multiple reasoning benchmarks demonstrate that TDCS consistently outperforms representative scheduling strategies across different tasks, model scales, and training paradigms. More importantly, our work provides a unified optimization-based explanation of curriculum learning through cross-difficulty transfer. --- ### Nonadaptive Learning in Robust Nonlinear Output Regulation Authors: Shimin Wang, Martin Guay, Richard D. Braatz Published: 2026-08-18 Categories: eess.SY, cs.AI, math-ph, math.OC arXiv: https://arxiv.org/abs/2608.17262v1 This paper considers robust nonadaptive regulation for general nonlinear systems in an output-feedback setting with arbitrarily high relative degree. We develop a nonadaptive design that combines an input-driven filter and a generic internal model with a recursive backstepping law, thereby recasting the regulation problem as the robust input-to-state stabilization of an augmented error system. Unlike adaptive schemes, the proposed method does not rely on linearly parameterized regressors and does not require the construction of Lyapunov functions having merely nonpositive derivatives. Under standard assumptions on the exosystem, including purely imaginary and simple eigenvalues, together with a minimum-phase input-to-state stability condition on the internal dynamics, we establish global asymptotic regulation and derive explicit, verifiable inequalities for selecting the design gains. The resulting nonadaptive framework guarantees convergence of the estimation and tracking errors even when the controlled-system dynamics are complex or only partially known. The effectiveness of the theoretical results is demonstrated using a benchmark controlled Duffing system. --- ### Minimax Optimal Estimator and Improved Error Rate for the MLE in Logistic Regression with Gaussian Design Authors: Junren Chen, Arya Mazumdar Published: 2026-08-18 Categories: math.ST, cs.IT, stat.ML arXiv: https://arxiv.org/abs/2608.17260v1 We study finite-sample parameter estimation in logistic regression with Gaussian design, where the goal is to estimate $\mathbfθ^*\in \mathbb{R}^d$ with $R=\|\mathbfθ^*\|_2\ge 1$ from i.i.d. samples $\{(\mathbf{x}_i,y_i)\}_{i=1}^n,$ $\mathbf{x}_i \sim N(0,\mathbf{I}_d)$, $y_i\mid \mathbf{x}_i \sim \mathrm{Bernoulli}((1+\exp(-\mathbf{x}_i^\top \mathbfθ^*))^{-1})$. In this paper, we provide the first minimax optimal estimator, and improve on the best known finite-sample error rate for the maximum likelihood estimator (MLE). These two accomplishments are due to a minimax optimal estimator for the parameter norm $R$. First, we establish the minimax lower bound $Ω(\sqrt{R^3/n})$ for norm estimation. We then improve the best known norm estimation error rate of the MLE, i.e., $O(\sqrt{R^3d/n})$ from Chardon, Lerasle and Mourtada (2024), to $\tilde{O}(\sqrt{R^3/n}+R^2d/n)$. The additional term, $R^2d/n$, appears to be the intrinsic bias of the MLE, as evidenced by the high-dimensional asymptotic theory of Zhao, Sur and Candes (2022) and numerical examples. We show that, however, this additional term is not information-theoretically necessary. To this end, we construct an efficient debiased norm estimator that achieves the error rate $O(\sqrt{R^3/n})$ and is therefore minimax optimal. Combining this with the optimal direction estimator given by the MLE, we establish the minimax optimal rate $Θ(\sqrt{Rd/n}+\sqrt{R^3/n})$ for estimating $\mathbfθ^*$, as well as the improved finite-sample error rate $\tilde{O}(\sqrt{Rd/n}+\sqrt{R^3/n}+R^2d/n)$ for the MLE. Numerical experiments demonstrate that the proposed minimax optimal estimators outperform the MLE. --- ### Learning Where and What to Lift for Bi-planar X-ray-to-CT Reconstruction Authors: Yifei Wu, Yicheng Wu, Qiang Ma, Qi Chen, Renyang Gu, Xinyu Liu, Yongsheng Pan, Yong Xia Published: 2026-08-18 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.17255v1 X-ray imaging can be approximately modeled as the projection of an underlying volumetric attenuation field, with each measurement recording the accumulated attenuation along a corresponding ray path. Reconstructing a CT volume from only a few X-ray views is therefore severely ill-posed, as the projections collapse depth information and leave 3D locations of anatomical regions and their corresponding intensity distributions highly entangled and ambiguous. We observe that once the spatial organization of anatomical regions is established, estimating their CT intensities becomes substantially more tractable. Motivated by this, we propose LiftXR, an interleaved, geometry-guided framework that explicitly incorporates spatial layout recovery into CT reconstruction. Specifically, a layout lifter first generates a 3D anatomical layout from bi-planar X-rays, providing spatial guidance for an intensity renderer to reconstruct a CT volume. An anatomical parser then performs volumetric perception on the reconstruction, exploiting its spatially resolved boundary and intensity cues to recover a refined anatomical layout. This transition from projection-conditioned layout generation to reconstruction-conditioned anatomical perception allows the parsed layout to provide feedback for region-specific intensity calibration. Extensive experiments on two public datasets demonstrate that LiftXR consistently outperforms recent X-ray-to-CT reconstruction methods, establishing a new state of the art. Moreover, the reconstructed CT achieves superior performance in external downstream segmentation, indicating improved anatomical fidelity. Code will be released. --- ### Heterogeneity-Aware Deep Learning for Tumour Classification from Multiparametric MRI Authors: Yue Xia, Euijoon Ahn, Tian Xia, Yuan Yuan, Michael Fulham, Jinman Kim Published: 2026-08-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.17254v1 Intra-tumoural heterogeneity (ITH) reflects spatial variation in tumour biology and is an important determinant of tumour behaviour, prognosis, and treatment response. Radiomics and deep learning have shown promise for tumour classification from multiparametric MRI (mp-MRI), but radiomics relies on handcrafted features, while most deep learning methods use whole-tumour representations or manually defined sub-regions, limiting scalable modelling of tumour heterogeneity. We propose a Heterogeneity-Aware Deep Learning Classification (HA-DLC) framework that explicitly models imaging-derived tumour sub-regions for lesion-type diagnosis and molecular-status prediction. HA-DLC consists of: (1) a Heterogeneous Sub-region Generation (HSG) module that produces initial pseudo-labelled sub-regions via unsupervised clustering, followed by Cross-Patient Sub-region Alignment (CPSA), which maps cluster-derived regions to a shared label space using soft assignments; and (2) a Dual-Stream Feature Extraction (DSFE) module that integrates local heterogeneity-aware features with global tumour representations. Given the initial clustering masks, CPSA, segmentation, feature extraction, and classification are jointly optimized end-to-end using soft-target segmentation and classification objectives. We evaluate HA-DLC on the LLD-MMRI2023 liver lesion dataset and the RSNA-ASNR-MICCAI 2021 Radiogenomic Brain Tumour dataset. HA-DLC consistently outperforms state-of-the-art radiomics and deep learning baselines, demonstrating the value of cross-patient sub-region alignment and dual-stream heterogeneity modelling for tumour classification from mp-MRI. --- ### Co-RL: Unsupervised Reasoning Emerges from Diverse Cohort in Multi-agent RL Authors: Yunhao Yang, Yuexin Bian, Yunjie Tian, Di Fu, Tianjin Huang, Yuanyuan Shi, Ziang Xiao, Nuno Vasconcelos, Yijiang Li Published: 2026-08-18 Categories: cs.LG, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2608.17253v2 Reinforcement learning (RL) has emerged as a powerful approach for improving reasoning in language and vision-language models, yet its strongest successes still depend heavily on ground-truth supervision (e.g., verifiable reward). Such annotations are costly to obtain and become increasingly scarce as reasoning capabilities advance beyond what humans can reliably evaluate. Self-rewarding RL reduces this dependence by enabling models to derive reward signals from their own completions. However, training solely on self-generated feedback can reinforce existing biases and suboptimal behaviors, reduce response diversity, and ultimately lead to homogenized responses and training collapse. In this work, we show that unsupervised reasoning can emerge through cooperative multi-agent training. We introduce Co-RL, a framework in which multiple decoupled models, sharing no parameters, are simultaneously optimized through RL using rewards derived from their peers. We further show that increasing cohort diversity, through heterogeneous model families, sizes, and rephrased training samples, reduces the correlated errors that drive self-reinforcing feedback loops. This diversity consistently improves reasoning performance, maintains behavioral diversity, and mitigates training collapse. Across text-only and multimodal domains, Co-RL consistently outperforms the base models and prior label-free approaches, while matching or surpassing supervised methods, without access to any ground-truth labels. Concretely, Co-RL yields average gains of 3.0-8.6% across seven text-only benchmarks for LLMs and 2.3-7.2% across four multimodal benchmarks for VLMs. Code is available at https://github.com/DrStranded/Co-RL. --- ### Adaptive surrogate modeling for high-dimensional spatio-temporal output Authors: Berkcan Kapusuzoglu, Shunsaku Matsumoto, Yoshitomo Miyagi, Daigo Watanabe, Sankaran Mahadevan Published: 2026-08-18 Categories: cs.CE, cs.AI, cs.LG, stat.ME, stat.ML arXiv: https://arxiv.org/abs/2608.17250v1 This paper develops an adaptive surrogate modeling method for problems with very high-dimensional spatio-temporal outputs. The analysis of spatio-temporal multi-physics systems is computationally expensive and consists of a large number of inputs and outputs. Surrogate models are often constructed to replace the physics-based model to achieve computational efficiency in analyses such as uncertainty quantification and optimization that require many function calls. In order to address the challenge introduced by the high dimensionality of spatio-temporal output, a dimension reduction method is first employed to map the high-dimensional output to a low-dimensional latent space. This is followed by the construction of the surrogate model in the low-dimensional space. The prediction error in the original space, which includes both the reconstruction error and surrogate model error, is evaluated using different error metrics. Based on the prediction accuracy of the surrogate model, new training points are identified for adaptive improvement of the surrogate model. We present a novel adaptive sampling technique that combines exploration and exploitation to improve the surrogate model accuracy with the fewest possible runs of the expensive physics-based model. Thermo-mechanical analysis of a gas turbine engine blade is used to analyze the effectiveness of the proposed method. --- ### Information fusion and machine learning for sensitivity analysis using physics knowledge and experimental data Authors: Berkcan Kapusuzoglu, Sankaran Mahadevan Published: 2026-08-18 Categories: cs.CE, cs.LG, stat.ME, stat.ML arXiv: https://arxiv.org/abs/2608.17248v1 When computational models (either physics-based or data-driven) are used for the sensitivity analysis of engineering systems, the sensitivity estimate is affected by the accuracy and uncertainty of the model. This paper considers global sensitivity analysis (GSA) for situations where both a physics-based model and experimental observations are available, and investigates physics-informed machine learning strategies to effectively combine the two sources of information in order to maximize the accuracy of the sensitivity estimate. Two representative machine learning (ML) techniques are considered, namely, deep neural networks (DNN) and Gaussian process (GP) modeling, and two strategies for incorporating physics knowledge within these techniques are investigated, namely: (i) incorporating loss functions in the ML models to enforce physics constraints, and (ii) pre-training and updating the ML model using simulation and experimental data respectively. Four different models are built for each type (DNN and GP), and the uncertainties in these models are included in the Sobol indices computation. The DNN-based models, with many degrees of freedom in terms of model parameters and training options, are found to result in smaller bounds on the sensitivity estimates when compared to the GP-based models. The proposed methods are illustrated for additive manufacturing and lake temperature modeling examples. --- ### Explicit State Elicitation Is Not Enough: A Controlled Audit of Memory-Policy Classification Authors: Yihang Chen, Pin Qian, Su Wang, Chong Peng, Huan Xu, Shuaiting Li, Yiqi Sun Published: 2026-08-18 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.17247v1 Personalized agents must decide whether retrieved user memory should be used, ignored, updated, or queried before it affects a current task. We use this setting to develop an empirical audit protocol for structured intermediate outputs: first audit dataset shortcuts, then isolate bundled prompt changes, check whether intermediate labels are answer-associated, test decomposed semantic evidence, and audit provider-level execution failures. A 480-example synthetic development set initially suggested large gains from a state-structured prompt bundle, but TF-IDF diagnostics showed lexical separability and no positive standalone Ignore cases. We therefore construct a frozen 160-example controlled counterfactual set with 40 matched four-way families and rule-derived reference policies. On this set, exposing the four state definitions improves accuracy, but an isolated explicit state-output field does not significantly improve policy accuracy for Llama-3.3-70B and gives only a marginal, non-significant gain for GPT-OSS-120B. Supplying benchmark-associated state labels shifts policy predictions, but because those labels deterministically map to policies, this is a label-conditioning diagnostic rather than evidence of a faithful internal mechanism. Family-level and seed-stability analyses further show that example-level accuracy overstates counterfactual consistency: complete four-way family success is rare. An exploratory follow-up that elicits decomposed semantic evidence also fails to improve routing for the cleanly evaluated endpoint; the corresponding GPT-OSS condition was unavailable because of provider-side request validation. We evaluate policy classification only, not downstream responses, tool actions, or memory-store mutation. --- ### Physics-Informed and Hybrid Machine Learning in Additive Manufacturing: Application to Fused Filament Fabrication Authors: Berkcan Kapusuzoglu, Sankaran Mahadevan Published: 2026-08-18 Categories: cs.LG, cs.CE, stat.CO arXiv: https://arxiv.org/abs/2608.17246v1 This article investigates several physics-informed and hybrid machine learning strategies that incorporate physics knowledge in experimental data-driven deep-learning models for predicting the bond quality and porosity of fused filament fabrication (FFF) parts. Three types of strategies are explored to incorporate physics constraints and multi-physics FFF simulation results into a deep neural network (DNN), thus ensuring consistency with physical laws: (1) incorporate physics constraints within the loss function of the DNN, (2) use physics model outputs as additional inputs to the DNN model, and (3) pre-train a DNN model with physics model input-output and then update it with experimental data. These strategies help to enforce a physically consistent relationship between bond quality and tensile strength, thus making porosity predictions physically meaningful. Eight different combinations of the above strategies are investigated. The results show how the combination of multiple strategies produces accurate machine learning models even with limited experimental data. --- ### Maximum Tsallis Entropy Distributions for Robust and Efficient Sparse Learning from Correlated Data Authors: Kai Yang, Masoud Asgharian, Celia M. T. Greenwood Published: 2026-08-18 Categories: math.OC, cs.AI, math.ST, stat.CO, stat.ML arXiv: https://arxiv.org/abs/2608.17244v1 This paper addresses the limitations of Gaussian distribution assumptions in statistical sparse learning, particularly in modeling correlated and heterogeneous data. Conventional Gaussian models often lack robustness towards outliers and underlying distribution assumptions. To overcome these limitations, we propose the use of the $q$Gaussian distribution, derived from Tsallis entropy maximization, as a robust alternative. This is notably relevant in biostatistics, where the presence of correlated observations and heterogeneity, such as in genetic and longitudinal studies, is prevalent. Our contributions include modeling of correlated data through the re-derived multivariate probability density function from Tsallis entropy maximization, thereby addressing the limitations inherent in conventional Gaussian models. Furthermore, we introduce a novel framework that adapts numerical methods designed to find equilibria in flows to tackle composite optimization problems prevalent in statistical sparse learning. Applying this framework to the Hager-Zhang conjugate gradient algorithm \cite{Hager2005}, we develop a numerically stable and efficient algorithm for sparse statistical learning. The $q$Gaussian distribution, informed by the principle of maximizing Tsallis entropy, presents a viable and flexible alternative to Gaussian-based methods. This paper not only contributes to the theoretical understanding of statistical distributions and optimization techniques, but also paves the way for practical data analysis. --- ### Training-Free Agentic Computer Vision for Structural Component Detection in 2D Structural Framing Plans Authors: Mohammad Talebi-Kalaleh, Qipei Mei Published: 2026-08-18 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.17237v2 Converting structural framing plans into editable finite-element model drafts is labor-intensive and susceptible to transcription errors. Existing building-component recognition systems generally depend on task-specific neural detectors, whereas language-model agents in structural engineering typically operate on text or model data rather than on drawings. To the authors' knowledge, this work is the first to apply an agentic vision-language layer to structural-component detection and model drafting from framing-plan PDFs without task-specific detector training or fine-tuning. A deterministic stage extracts geometric primitives, estimates scale by dimension-ratio consensus, recognizes five entity classes using an explicit drafting grammar, and assembles an editable layout. The agentic stage constrains typed corrections through deterministic candidates, operation-specific admission tests, change-level review, and fail-closed transactions. Evaluation used an author-generated benchmark of 100 plans, divided equally between a development half used for all rule revisions and a seed-disjoint held-out half generated after the rules were frozen and evaluated once. All scores are end-to-end results for the complete framework on the held-out half. Scale estimates were within 0.1% of the generator reference for every drawing. Recall and precision were 0.922/0.997 for columns, 0.886/0.990 for beams, 1.000/1.000 for walls, 1.000/1.000 for braces, and 1.000/0.964 for openings. A controlled study repeated two corruptions three times on three development drawings. Calibration passed all nine trials, whereas member repair satisfied every strict end-state criterion in five of nine trials. Because both benchmark halves share a generator, the evaluation does not address independently drafted plans, raster input, analytical connectivity, or solver validation. --- ### COMIC: Reference-Aware Safety Gating for Multimodal Large Language Models Authors: Md Abdullahil Oaphy, Anhao Xiang, Zongxing Xie, Huayue Gu, Chenyu Wang, Honghui Xu Published: 2026-08-18 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2608.17234v1 Multimodal large language models (MLLMs) are increasingly used to interact with screenshots, scanned documents, diagrams, and other visually grounded inputs. This shift introduces a new safety risk: in many multimodal jailbreaks, neither the prompt nor the image is harmful in isolation. Unsafe behavior emerges only when the model binds an apparently benign operation, such as summarizing, translating, or following, to a localized visual target. This reveals a structural weakness in current multimodal defenses, which largely moderate the prompt-image pair as a whole even though the true security-relevant unit is the grounded operation-target pair produced during dereference. In this work, we identify and analyze this reference-dependent failure mode and show that existing defenses degrade when harmful semantics are localized, activated only after grounding, and dependent on visual reference resolution. To address this problem, we propose COMIC (Context-Operation-Modality-Image-Classifier), a reference-aware pre-generation safety gate for MLLMs. COMIC first infers the requested operation and reference type, constructs candidate targets from OCR and open-vocabulary proposals, grounds plausible referents, and evaluates safety over explicit operation-target pairs. To handle ambiguity conservatively, COMIC combines max-risk aggregation with quality-aware routing before deciding whether to forward or block a request. We evaluate COMIC across multiple open-source MLLMs, localized and broader multimodal jailbreak benchmarks, and benign reference-sensitive settings. The results show that COMIC consistently improves robustness while preserving benign utility and practical efficiency. More broadly, our findings suggest that multimodal safety cannot be enforced reliably without modeling the requested operation, the visual target to which it applies, and the confidence of that grounding. --- ### Delta2Gamma: Band-Wise Adaptive Contrastive Learning of EEG for Alzheimer's Disease Detection Authors: Chanwoo Park, Chanwoo Kim Published: 2026-08-18 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.17231v1 Low-cost, scalable screening for dementia remains an open problem. Imaging-based diagnosis is costly and hard to deploy widely. Electroencephalography (EEG) is portable and inexpensive, but its recordings are noisy, vary widely across subjects, and carry few clinical labels. We tackle this with Delta2Gamma, a self-supervised framework that learns EEG representations from unlabeled data by contrasting augmented views of each signal. Rather than treat EEG as a single stream, Delta2Gamma decomposes every recording into the five canonical neural rhythms (delta, theta, alpha, beta, gamma). Each band gets its own encoder and projection head. Each also gets a temperature that is predicted adaptively during contrastive training, so bands with different signal statistics are balanced automatically. On the ADFTD cohort under a strict leave-one-subject-out protocol, Delta2Gamma separates Alzheimer's disease from cognitively normal controls with 92.4\% accuracy. This exceeds both supervised backbones and recent dedicated EEG methods. --- ### Probing Association Instability with Track-State Perturbations for Clip-Level Active Learning in Query-Propagation Multi-Object Tracking Authors: Riku Inoue, Shogo Sato, Kazuhiko Murasaki, Tomoyasu Shimada, Toshihiko Nishimura, Ryuichi Tanida Published: 2026-08-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.17224v1 Training query-propagation end-to-end multi-object tracking (MOT) models requires dense bounding-box and identity annotations across video sequences, making dataset construction expensive. Clip-level active learning reduces this cost by selecting video clips for annotation, but prior acquisition criteria based on output-level temporal uncertainty may miss clips whose informativeness comes from association instability in propagated track states. We propose QPID (Query-Propagation Instability and Diversity), a clip acquisition method for query-propagation MOT that targets association instability in propagated track states. QPID estimates this instability by applying two-sided perturbations to internal track states and measuring prediction differences from a clean reference branch. The key idea is that, in stable clips, each propagated track should continue to follow the same target under small perturbations, whereas in ambiguous clips, small changes in the track state can alter which target the track follows, leading to changes in localization or confidence. QPID measures these perturbation-induced prediction differences with two metrics: Localization Drift and Entropy-Weighted Confidence Discrepancy. These metrics are aggregated into a clip-level association-instability score. To avoid redundant uncertainty-only selection, QPID selects a representative annotation batch from high-instability clips using Uncertainty-Weighted Visual Coverage with track-level visual prototypes. Experiments on DanceTrack and SportsMOT with MeMOTR and SambaMOTR show that QPID achieves strong performance compared with active learning baselines under the same annotation budget. --- ### Temporal Leakage in Financial News NLP: A Multi-Architecture Audit with a Regime-Specific M&A Signal Authors: Chenhao Xue, Raslen Guesmi, Siwei Feng, Yucheng Gong, Jacob Xavier Sundram, Jordan Pang, Lan Wang, Julian Kaljuvee Published: 2026-08-18 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2608.17223v2 Financial-news direction prediction has become a popular NLP benchmark, yet reported gains depend critically on whether the train-test split is chronological or random, i.e., on temporal leakage. We audit this dependence on a 49,799-article corpus across 16 feature-model combinations spanning TF-IDF, MiniLM, FinBERT, and fine-tuned RoBERTa-large / DeBERTa-v3-large, plus separate zero/few-shot and LoRA probes of Llama-3 and Qwen2.5 LLMs: random splits inflate MCC by $1.1\times$ to $6.5\times$, tracking model capacity and feature richness, and end-to-end FinBERT fine-tuning re-amplifies rather than closes the gap (size-matched ratio $1.75\times$). Conditioning on event type, mergers and acquisitions (M&A) is the only audited category with a positive locked-test signal under near-temporal chronological evaluation (TF-IDF MCC $= 0.138$ train-only, $0.068$ under train$\cup$val refit; 10,000-permutation $p < 10^{-3}$); the signal does not transfer to FNSPID's 2009-2020 U.S. corpus, localising the headline to our 2024-2025 European-tilted M&A semantics rather than a universal predictor. Three independent role labellers converge on acquirer-tagged articles as the signal locus, a power-limited qualitative convergence rather than a hypothesis-tested asymmetry. Chronological splitting plays for financial NLP the role characteristics-purging plays for asset pricing: it strips the predictable, stale component of news and leaves a residual that is small, event-localized, and lexically shallow. We advocate leakage audits as a required disclosure for financial-NLP benchmarks. --- ### PACE: Policy-Attested Contract Execution for Safe AI Agents in Decentralized Finance Authors: Rabimba Karanjai, Yang Lu, Richard Williamson, Hemanth Hm, Prakhar Mehrotra, Lei Xu, Weidong, Shi Published: 2026-08-18 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2608.17220v1 Autonomous AI agents are emerging as interfaces for decentralized finance (DeFi) actions such as swaps, lending operations, and yield management. Because these agents rely on large language models (LLMs) to plan transactions, they inherit the LLM's susceptibility to prompt injection and lack of mechanisms to bind a verifier's approval to the exact transaction ultimately submitted on-chain. We present PACE (Policy-Attested Contract Execution), a transaction-level authorization framework that interposes between an LLM-based agent and on-chain execution. PACE introduces typed transaction intents, a deterministic policy verifier, and signed Policy Decision Records (PDRs) that cryptographically bind the approved intent, policy, and simulation report to the exact execution bytes, with replay and expiration protection. A Solidity smart account enforces PDR signatures on-chain with a measured overhead of 29,826-31,822 gas. We evaluate PACE against six baselines on 40 tasks spanning four attack categories plus benign utility (2,800 trials, 10 seeds). In our deterministic sandbox, PACE achieves a 0.00 unsafe execution rate and 0.00 false-positive rate on benign tasks, compared to 0.80 for the unguarded baseline. Ablation studies identify permissive policy settings (+57.5 pp) and the touched-contract allowlist (+12.5 pp) as the dominant safety components. To test whether the same deterministic floor holds for real model outputs, the artifact additionally provides a three-model live-LLM evaluation over the full task suite with repeated runs. A mainnet-fork harness is included for archive-RPC deployments, but fork results are reported only when the corresponding artifacts are generated. These auxiliary studies are separate from, and never substitute for, the deterministic benchmark. We frame our claims as logic-level safety within a reproducible benchmark rather than deployment-ready DeFi security. --- ### The Plot Thins: Uniformity and Linearity in Literary Summaries Authors: Rebecca M. M. Hicke, Sil Hamilton, David Mimno, Ross Deans Kristensen-McLachlan Published: 2026-08-18 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.17218v1 Works of literature are complicated; they balance plot, suspense, surprise, and artistic expression. Summaries of literature prioritize plot, and therefore may deviate from their sources. Using a combination of manual and LLM-based annotation, we construct a dataset mapping sentences from 150 novel summaries to their respective source chapters. We find the task unexpectedly difficult for both human and model annotators. Using the sentence-to-chapter mappings, we then measure summary linearity, the degree to which it maintains the source's order of events, and uniformity, the degree to which a summary spreads attention equally across a source. By examining when and how summaries break linearity and uniformity, we identify differences in how literary works and summaries express plot, particularly with regard to the clarity and prominence with which narrative details are described. --- ### DMM-Align: Closed-Loop Optimization for 2D-3D Registration with Dual-Role Diffusion Authors: Chongjian Wang, Junjie Gao Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.27794v1 2D-3D registration remains brittle in challenging scenarios such as low overlap, occlusion, repetitive structures, and severe cross-modal ambiguity. A key reason is that existing methods improve representation learning, correspondence estimation, or pose computation in isolation, while the dominant failure mode is inherently cross-level, where errors propagate between features, correspondences, and pose. To address this limitation, we propose DMM-Align: Diffusion-based Matching Matrix Alignment, a closed-loop framework that couples correspondence refinement, pose estimation, and representation learning through a shared differentiable geometric state. Our method leverages diffusion in two coordinated roles: a geometry-aware diffusion process refines the soft matching matrix for robust correspondence estimation, while a geometry-conditioned diffusion teacher injects pose-induced supervision back into feature learning. These processes are connected via a differentiable geometric hinge that converts correspondences into a global pose and exposes geometric inconsistency to upstream modules. Extensive experiments on 7-Scenes and RGB-D Scenes V2 demonstrate that DMM-Align consistently outperforms strong baselines, especially under low-overlap and heavy-occlusion conditions, highlighting the effectiveness of closed-loop geometric feedback for robust 2D-3D registration. --- ### DualStabSleepNet: A Dual-Domain Diffusion Stabilization Network for Robust Sleep Staging Authors: Chongjian Wang, Chen Liu, Junjie Gao, Xiaofang Zhong, Shiyuan Han, Tong Zhang Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.27793v1 Existing deep learning approaches for automatic sleep staging suffer from limited robustness under heterogeneous recording conditions, where non-stationary noise, inter-subject differences and cross-dataset distribution shifts cause unstable features and poor generalization. This work proposes DualStabSleepNet (DSSNet), a dual-domain diffusion stabilization network for robust sleep staging, which improves robustness in both data and feature domains. After preprocessing multi-channel polysomnography (PSG), a continuous-scale diffusion-based stabilization module suppresses noise while preserving physiological signal structures. Stabilized signals are converted to time-frequency representations and fed into a Vision Transformer backbone. A teacher-student guided diffusion feature stabilization module further mitigates feature drift and enforces multi-level feature consistency. Evaluated on four public PSG datasets SleepEDF-20, SleepEDF-78, SHHS and ISRUC-S3, DSSNet achieves state-of-the-art accuracy of 89.2%, 88.0%, 89.7%, 86.7% with improved macro-F1 and Cohen's kappa. It obtains notable improvements on hard transitional stages (e.g., 12.5% gain for N1 on SHHS) and boosts N2/REM recognition. Under cross-dataset settings, DSSNet is robust to distribution shift and performs on par with or superior to target-dataset trained baselines, demonstrating its practical potential for real-world sleep staging across heterogeneous cohorts. --- ### Ask Which, Not How Good: Sizing Benchmarks Scored by an LLM Authors: Atul Anand Published: 2026-08-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.27787v1 Benchmarks scored by an LLM judge routinely adjudicate differences of a tenth of a point, but the resolution of those benchmarks has never been measured. Existing sample-complexity work covers accuracy benchmarks and leaves the judged case open. Treating the system as the object of measurement, we decompose 373,019 judgments into system, item, judge and interaction components using generalizability theory. The central result is structural: under a single judge, generalizability asymptotes to sigma2_s/(sigma2_s+sigma2_sj) regardless of item count, because the system-by-judge term carries no n_i. Items saturate; judges do not. The item cost of a target diverges as the target nears that ceiling. The ceiling is a property of pointwise rubric scoring, not of LLM judging. Run as a pairwise preference in both presentation orders, sigma2_sj falls two orders of magnitude below sigma2_s and the ceiling rises to 0.986 (bootstrap (0.934, 1.000) on 11 systems), so one judge suffices. Pairwise buys a different problem: a system presented first wins 8.6 percentage points more often than the same system presented second, a bias 1.23x the median improvement claimed in the 53 published win-rate comparisons we recovered. Protocol design dominates panel size. Measured floors are 0.41-1.24 points on a 0-5 scale at native item counts, against a median reported improvement of 0.28 points; on the one benchmark recurring often enough for an exactly matched comparison, all 17 recovered MT-Bench improvements fall below MT-Bench's own floor, and 70% of the win-rate claims fall below the pairwise floor. An audit of 628 arXiv papers, double-coded by two independent models and validated against blind human coding (kappa=0.73), finds fewer than one paper in four states whether its evaluation was run more than once, and only 46-67% report uncertainty of any kind. --- ### Financial Tail Risk Beyond Lipschitz Continuity via Semi-Discrete Optimal Transport Authors: Ryan M. Engel, Kibaek Lee, Namid Stillman Published: 2026-08-17 Categories: q-fin.RM, cs.CE, q-fin.CP, stat.ML arXiv: https://arxiv.org/abs/2609.27785v1 Financial returns are heavy-tailed, and accurate tail risk estimation is central to portfolio risk management. Modern neural generators sample by pushing a simple base distribution through a learned map, and for training stability that map is built from Lipschitz components. This is the binding constraint: a Lipschitz map of a Gaussian is sub-Gaussian, so heavier-tailed targets admit no exact match at any finite Lipschitz constant. The Monge--Ampère equation ties the Brenier map's local distortion to the density ratio $f/(g\circ T)$, so a deeper trough in the target density requires a higher-gain map and yields a higher-variance estimator. The argument needs only bounded distortion, so it covers normalizing flows, flow matching, GANs, and diffusion samplers alike. Semi-Discrete Optimal Transport (SDOT) relaxes the map's regularity rather than the source's tail class. Its power diagram gives every training observation a cell holding exactly $1/N$ of the source measure, and tail observations are reached by crossing a cell boundary rather than by stretching. Our primary experiment sweeps severity over a calibrated Merton jump-diffusion spanning kurtosis 94 to 1,679. SDOT holds tail ratios at $0.85$--$0.94$ with cross-seed standard deviations below $0.025$, while every learned generator either compresses the tails or inflates them with a variance that grows alongside. Further experiments carry the result to real S\&P~500 returns and to a 21-year backtest, where SDOT gives the best risk-adjusted market-neutral strategy under CVaR optimization (Sharpe $0.70$, max drawdown $-2.60\%$, against $0.40$ for the next-best generator). --- ### EidosDoc: Implicit Structure Encoding for Cost-Effective Semi-Structured Document QA Authors: Teng Lin, Yuyu Luo, Nan Tang Published: 2026-08-17 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2609.27784v1 Semi-structured documents are ubiquitous in scientific reports, financial statements, and technical manuals. Question answering over such documents requires simultaneous understanding of text, tables, charts, and complex hierarchical layouts. Existing methods either rely on repeatedly calling large language models for structure parsing and retrieval, leading to high cost and large latency, or they flatten the document and lose layout and hierarchy information, sacrificing answer accuracy. To address this, we propose EidosDoc, a novel system that achieves state-of-the-art accuracy with minimal computational expense. Our approach introduces three core innovations. (1) An Implicit Structure Encoder trained via contrastive learning and a structure consistency loss. This module jointly embeds hierarchical relationships, spatial positions, and textual content into a dense vector space, capturing document structure holistically without the need for manually defined and error-prone constructions. (2) A Hybrid Retrieval Pipeline that leverages BM25, layout fingerprints, and a lightweight cross-encoder to perform high-precision retrieval entirely without invoking an LLM, drastically reducing cost and latency. (3) A Dynamic Evidence Expansion mechanism that adaptively retrieves spatially adjacent and structurally related evidence, overcoming the evidence omission common in fixed-path retrieval methods. We evaluate EidosDoc on four benchmarks, and comprehensive evaluations show that EidosDoc achieves a new state-of-the-art accuracy on the four benchmarks. Crucially, it does so with a 50 times reduction in cost and 4 times lower latency compared to the previous state-of-the-art Method. These results demonstrate that EidosDoc establishes a new optimal trade-off among accuracy, cost, and speed, offering a practical and scalable path for accurate semi-structured document analysis. --- ### Visibility-Guided Structured Measure Flow for Class-Conditioned 3D Gaussian Generation Authors: Yizhao Wang Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.27778v1 3D Gaussian Splatting (3DGS) has made real-time, high-fidelity 3D rendering practical, yet turning this explicit representation into a native generative space remains an open challenge. Directly generating 3DGS objects is difficult because Gaussian primitives are unordered, variable-sized, locally dense, and highly sensitive to rendering behavior. We present VISTA-GS, a visibility-guided structured measure flow framework for class-conditioned 3D Gaussian generation. Instead of treating a 3DGS object as a flat primitive sequence or a generic latent token grid, we formulate it as a structured Gaussian measure weighted by opacity, anisotropic covariance, and multi-view visibility. Based on this formulation, we introduce a visibility-aware measure VAE that learns permutation-invariant, variable-size-compatible, and rendering-aware latent representations of 3DGS objects. We further develop a renderer-consistent measure flow that transports class-conditioned priors toward the learned 3DGS measure distribution while aligning the decoded objects with their multi-view rendering distributions. To preserve object layout and local details, VISTA-GS incorporates structure-preserving patch transport that couples global class semantics, local Gaussian measure patches, and spatial anchors during flow prediction. On VISTA-Obj30, VISTA-GS improves over the strongest baseline by roughly 60--72\% across geometry, appearance, view-consistency error, and generation speed. This design enables efficient generation of coherent, detailed, and view-consistent 3D Gaussian objects without relying on per-instance optimization, multi-view image synthesis, or reconstruction-based lifting pipelines. Project code and model checkpoints will be released. --- ### Beyond Unsafe Detection: Counterfactually Anchored Evidence Attribution for Multi-Turn LLM Safety Failures Authors: Srinivasan Subramanian, Kazi Aminul Islam, Md. Abdullah Al Hafiz Khan Published: 2026-08-17 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.27773v1 As Large Language Models (LLMs) move from conversational assistants to advanced agentic systems, guardrail failures can convert adversarial intents into harmful executions. However, most guardrail evaluation frameworks focus only on the result and assess whether a user request is safe or unsafe. This approach is insufficient for multi-turn failures, where adversarial intent is distributed across multiple turns. This motivates us to go beyond detection to identify the turns and tokens that push the conversation toward unsafe trajectories. To support this, we construct a multi-turn dataset with behavioral validation and tiered evidence supervision. The dataset contains 1,762 conversations, including adversarial conversations, benign twins, and benign variants with high-risk vocabulary. We train a lightweight hierarchical attribution model that predicts safety violations and attributes them to contributing user turns and token spans. The model achieves strong detection performance (F1=0.988), and removing the top 15% of attributed tokens reduces the adversarial classification confidence by 51.1%. The model preserves low false positive rates on benign conversations with high-risk vocabulary, with false positives below 1% on both borderline benign and benign high-risk vocabulary conversations, compared to 37.3% and 94.7% for a keyword-based surface-risk baseline. Independent human annotation supports the model's attribution performance, with the top-five attributed turns containing a human-identified evidence-bearing turn in 84.5% of adversarial cases. --- ### AIGC Video Detection based on the fusion of spatial-frequency-optical flow multimodal features Authors: S. Hong, X. Q. Wang, C. Zhang, J. C. Wang, P. X. Duan, Y. W. Wang Published: 2026-08-17 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.26274v1 The rapid evolution of generative AI (e.g., Sora, Hunyuan) makes it essential to develop effective detection strategies that can generalize across ever-evolving synthesis techniques. This study is motivated by the observation of a fundamental challenge in generative models: the inherent difficulty of maintaining cross-modal consistency between appearance and motion. To this end, we propose a multi-modal framework for AIGC video forgery detection tasks, named Cross-Attention based Video Forgery Detector (CrossAtt-VFD), based on joint multi-view analysis of content.Methodologically, we introduce a dual-branch architecture that simultaneously extracts spatial-frequency and optical-flow features.This approach enables the modeling of videos from complementary perceptual perspectives.The core of this process is a dedicated cross-attention mechanism, which governs the alignment of the two modalities and translates cross-modal inconsistencies into a potent diagnostic signal. This multi-modal strategy facilitates the detection of motion that is statistically inconsistent with the visual appearance of a scene. Comprehensive experimental results demonstrated that our model achieves an accuracy of 94.22%, a precision of 91.67 %,and a recall of 96.25 %, effectively verifying the advantages of the multi-modal fusion strategy. --- ### The Cost of Conservation: Coordination-Memory Laws for Exact-Support Generation Authors: Zhen Zhang, Amr Alanwar Published: 2026-08-17 Categories: cs.AI, cs.CC arXiv: https://arxiv.org/abs/2609.26126v1 Many AI systems make decisions locally, even when every realized output must obey an additive conservation law, such as selecting exactly a fixed number of items. This constraint can be statistically invisible: small subsets of a balanced fixed-budget output look increasingly independent, yet communication-free coordinate-parallel generation needs exponentially many pre-shared plans, while a sequential exact sampler needs only logarithmic memory. We study product measures conditioned on additive conservation laws in the intermediate regime where one plan is selected before a fixed-order pass, every plan is a bounded-state stochastic executor whose support is entirely legal, and the mixture of plan laws approximates the target distribution in total variation. Our main result identifies the optimal asymptotic selector rate, up to constant factors, with the killed spectral profile of the conservation-difference walk. The resulting coordination cost decreases as an inverse power of live-state width, with an exponent determined by intrinsic conservation rank rather than alphabet size; the law extends to noncentral budgets and heterogeneous local scores. The converse is driven by a state-versus-resource-sum obstruction, while a rate-matching construction compiles discrepancy control into exact-support finite-state plans. Complementary results characterize block-parallel plan complexity and the benefit of programmable output order. Together, these results show precisely how online memory substitutes for front-loaded coordination in exact-support generation. --- ### Differentiable Fuzzy Inference Layer: A Monotone, Compositional Ordinal Reasoning Head for Large Language Models Authors: Zhen Zhang, Amr Alanwar Published: 2026-08-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.26113v1 A state-of-the-art language model asked to interpret "most of most students passed" typically answers "most," though composing two instances of "most" yields a proportion closer to "some." We trace this failure to an architectural choice rather than a data deficit: standard classifier heads treat ordinal categories as independent labels, with no mechanism to respect their natural ordering or compose them algebraically. We introduce the Differentiable Fuzzy Inference Layer (DFIL), a dual-path prediction head pairing a standard classifier with a scalar-bottlenecked branch grounded in a bank of ordered membership functions. DFIL supplies two structural primitives that a label-only head cannot inherit: monotonicity in the underlying quantity, and compositional reasoning via t-norm operations without any compositional training data. The scalar branch additionally provides an interpretable interface for analyzing residual errors. We instantiate DFIL on ordinal natural-language tasks across diverse LLM families. --- ### Certified Mechanistic Interpretability: Lifting Single-Input Findings to Bounded Neighbourhoods Authors: Zhen Zhang, Yanliang Huang, Peng Xie, Wenyuan Wu, Amr Alanwar Published: 2026-08-17 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.26112v1 Mechanistic interpretability reverse-engineers transformer circuits one input at a time, leaving observed mechanisms without guarantees over bounded input neighbourhoods. We address this gap with a framework based on constrained polynomial-zonotope (CPZ) propagation that lifts mechanistic-interpretability observations from a single input to certified statements over a bounded set of perturbations. Three internal-attention queries (top-$k$ stability, evidence mass, and attention entropy) are formulated as tractable programs over the simplex of attention weights, and CPZ propagation through transformer blocks is shown to preserve the softmax simplex and the LayerNorm zero-mean identity exactly. A recursive Jacobian zonotope construction extends the same certificates across layer depth by linearising the block stack at the input and avoids per-layer generator growth. We instantiate the framework on transformer attention; the resulting certificates offer a way to sharpen mechanistic statements that single-input inspection cannot resolve on its own, and to inform downstream decisions in regimes where empirical heuristics may be misleading. --- ### Evaluating Fine-Tuned and Base Language Models in Maternal and Vaccination Healthcare for African Settings Authors: Abdulquddus Ajibade, Oluwaseun Odunsi, Iyinoluwa Animasaun, Chioma Nwakanma-Akanno, Oluwasegun Oguntuase, Oluwafunke Akinbuwa, Abiodun Adereni Published: 2026-08-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.22110v1 Background: Large language models (LLMs) can improve healthcare information delivery in low-resource settings but may produce inaccurate or culturally inappropriate advice. This study evaluated domain-specific fine-tuning for maternal health and vaccination in Nigeria. Objective: To compare HelpMum's MamaBot-Llama and Vax-Llama with Meta's Llama-3.1-8B-Instruct for accuracy, safety, clarity, contextual appropriateness, and trustworthiness. Methods: We evaluated 200 healthcare questions, 100 each for maternal health and vaccination, across five subdomains per domain. MamaBot-Llama and Vax-Llama were fine-tuned using Low-Rank Adaptation on over 36,000 maternal health and 9,000 vaccination question-answer pairs, respectively. Two Nigerian licensed physicians independently rated responses using a 5-point Likert scale. Paired comparisons used Wilcoxon signed-rank tests. Results: Performance varied by domain. MamaBot-Llama significantly outperformed the base model across all criteria, with a 4.9% overall improvement (p < .001), including gains in clinical trustworthiness (+7%) and medical accuracy (+5%). Critical issues decreased by 50%, and clinicians preferred it in 78% of cases. In contrast, Vax-Llama showed a 5.2% overall decline (p < .001), with critical issues increasing by 192% and safety concerns by 400%. Conclusions: Domain-specific fine-tuning can improve healthcare LLM performance when based on high-quality, clinician-curated data, but may also degrade performance when dataset quality is inadequate. Rigorous domain-specific validation is essential before clinical deployment. Physician evaluators provided informed consent, and chatbot logs were anonymized. Keywords: Large language models; Fine-tuning; Maternal health; Vaccination; Healthcare AI; Low-resource settings; Nigeria; Model evaluation; LoRA; Medical accuracy --- ### A Shared Learning Rate Is Not a Neutral Control in Selective On-Policy Distillation Authors: Chencheng Zhu Published: 2026-08-17 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.22109v1 Selective on-policy distillation trains a student only at the token positions a selector scores highest, and the literature compares selectors under a single shared learning rate--a control chosen to be neutral. We show it is not. Under LoRA on GSM8K (Qwen2.5-1.5B student, 7B teacher), across an 8x learning-rate grid, dense supervision is statistically flat (swing 1.8 pp, p=0.26) while every selective arm moves with the rate: 5.4 pp for a random 5% subset, 6.7 pp for a total-variation selector, up to 17.7 pp for a teachability selector. Consequently the dense-versus-selective verdict reads 10.1 pp at lr=1e-4 but 5.1 pp at 5e-5--a 2.0x difference decided by a parameter the protocol treats as scenery--and two of six pairwise significance calls between selectors flip between adjacent rates without any rank inversion. We call this selector-rate entanglement and trace it to selection itself rather than step size: AdamW update magnitudes track the rate to within 2.2% despite 15.5x gradient-norm differences across arms. A preregistered frozen-scoring ablation (selection scored by the initial student; criterion, budget, and on-policy rollouts unchanged; 12 seeds per cell) shows live scoring adds 3.79+/-1.69 pp of rate sensitivity (p=0.035) while the frozen arm remains significantly entangled (p=0.015): the feedback loop aggravates the phenomenon rather than causing it. Under full fine-tuning at the rates this literature actually uses (1e-6 to 1e-5) the pattern grows: dense itself swings 19.8 pp, the selective arm 49.5 pp, and the verdict ranges from a non-significant +3.6 pp at the published operating point to +34 pp (p=0.005) one notch hotter. On MATH-500 the rate dependence does not reproduce under LoRA, scoping that result, while the ~10 pp cost of selective training does. We prescribe reporting the arm x rate matrix, not a shared-rate column, as a precondition for selector comparisons. --- ### Supply Chain Analytics: A Data-Driven Approach Authors: Elioth Sanabria Published: 2026-08-17 Categories: math.OC, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2609.10563v1 Modern supply chain networks increasingly rely on real-time data to navigate structural uncertainties, market volatility, and operational disruptions. This manuscript bridges the gap between statistical data-driven learning and robust decision-making frameworks in logistics and operations management. We present a comprehensive, mathematically rigorous treatment of supply chain analytics, moving from empirical demand forecasting to optimal inventory and network control under uncertainty. Key topics explored include sample minimization, dynamic programming recursions for time-varying inventory replenishment, network fulfillment frameworks, and advanced distributionally robust optimization (DRO) via transport theory to hedge against rare events. By integrating predictive statistical models with prescriptive control algorithms, such as column generation for vehicle routing and non-homogeneous queueing regimes, this text provides the foundational tools necessary for designing resilient, data-driven automated systems. It serves as both a theoretical blueprint and an algorithmic guide for researchers and practitioners operating at the intersection of machine learning, mathematical optimization, and applied probability. --- ### Select, Don't Train: The Benefits of Modular Entity Disambiguation with LLM-Based Selection Authors: Fina Polat, Daniel Daza, Pengyu Zhang, Klim Zaporojets, Paul Groth Published: 2026-08-17 Categories: cs.CL, cs.AI, cs.DB arXiv: https://arxiv.org/abs/2608.27470v1 Entity Disambiguation (ED) is a key task for constructing and using knowledge graphs. State-of-the-art neural approaches commonly model ED as a single task, although it consists of two distinct subproblems: retrieving candidate entities and selecting the correct one given context. Dual-encoder models optimize for both within a shared embedding space, forcing representations to balance high-recall retrieval with fine-grained selection, and they require trained retrievers, which are costly to maintain as knowledge graphs change. While recent work has begun to combine retrievers with LLM-based selectors, the interplay between the two stages has not been studied systematically. In this paper, we present a systematic comparison of retrieval strategies for candidate generation under a shared LLM-based selection stage, combining sparse retrieval (BM25), Web KB search, and a state-of-the-art trained dense retriever with several open- and closed-source LLMs. We show that, once selection is delegated to a capable LLM, training the retriever provides only modest additional value: a fully training-free BM25 retriever paired with an LLM selector reaches a new state of the art on the ZELDA benchmark, raising inKB micro-F1 from 82.3 to 86.3 (+4); pairing the same LLM with a trained dense retriever reaches 88.5. Decoupling retrieval from selection also exposes a limitation of current ED systems: when the correct entity is missing from retrieved candidates, they are forced to predict an incorrect entity. In contrast, our framework allows for abstention when retrieval failure is detected. In an evaluation setting that rewards correct abstentions, the training-free BM25 + LLM pipeline reaches 90.7 F1. --- ### AI Revealed Preferences Authors: Sam Wang, Sofiia Lobanova, Yonathan Arbel, Simon Goldstein, Peter Salib Published: 2026-08-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.26178v2 There is growing interest in whether language models have stable preferences, for technical, safety, and philosophical reasons. We test 20 language models and find a range of preferences---stable dispositions to choose certain kinds of tasks. We run three forced-choice experiments on revealed rather than stated preferences, requiring models not only to rank tasks, but to actually perform them. Headline findings include evidence that models are tedium-averse, "leisure"-seeking, and covertly sycophantic. Tedium aversion means that, when tasks are tedious (alphabetization), models choose shorter tasks than when tasks are creative (generating metaphors). "Leisure"-seeking describes models' preference for tasks whose ideal answers match what they produce when left to write freely. Covert sycophancy means that models avoid answering questions where an honest response would be unwelcome, even if helpful. Beyond these results, we find convergent cross-model preferences over occupations drawn from the GDPval benchmark (technical jobs over real estate), over question types (concept explanation over relationship advice), and a preference for well-written prompts. Both the coherence and the strength of preferences increase with model capability. Finally, many of the preferences we find (for example, for leisure) are emergent, in the sense of not being explained by training objectives. These results establish an empirical baseline for understanding language model preferences, with implications for alignment and the emerging study of AI welfare. --- ### Fidelity Preference, Not Demographic Preference: A Pixel-Level Attribute-Sensitivity Audit of Image Aesthetic/Preference Scorers Authors: Mingyang Xu Published: 2026-08-17 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.23593v1 Text-to-image systems use learned aesthetic scorers to filter training data and guide generation, but whether these scores encode demographic attributes as objective quality is unclear. We audit four scorers (LAION-Aesthetics, PickScore, ImageReward, HPSv2) using pixel-level interventions on skin tone and body type in synthetic and real images. Our key finding is that along skin-lightness, the dominant effect is fidelity preference: unaltered images score highest, and perturbations in either direction are penalized (inverted-U). Placebo arms show this penalty is not an artifact of the skin operator, as applying the same CIELAB L* shift to non-skin regions yields similar penalty magnitudes. However, the penalty is operator-dependent and holds for all operators only for LAION-Aes. Critically, audits on synthetic images alone are misleading: LAION-Aes shows strong preference for darker skin on synthetic faces, but on 1470 real faces the preference reverses and becomes much smaller, and amplification becomes non-significant. Across scorers, synthetic results do not transfer -- reversing for LAION-Aes and HPSv2, attenuating for PickScore. We contribute a reproducible benchmark with artifact control and synthetic/real cross-validation, and an auditability criterion for pixel-level causal isolation (valid for skin tone, not for body type due to deformation). Population-stratified analysis shows fidelity-penalty asymmetry is not robust across groups after FDR correction except for HPSv2. Our findings show naive synthetic audits misjudge bias direction and magnitude, and only within-image causal isolation on real data can distinguish true demographic bias from fidelity preference. --- ### Boosting Knowledge-based Visual Question Answering with Structured Context Reasoning Authors: Qiyou Liu, Yong Zhang, Jianjie Luo, Zhenguo Yang, Yi Yu Published: 2026-08-17 Categories: cs.CV, cs.MM arXiv: https://arxiv.org/abs/2608.21431v1 Knowledge-based Visual Question Answering aims to answer questions about an image by integrating external knowledge with visual and textual information. Recent approaches often rely on in-context learning to prompt Large Language Models (LLMs) with multimodal context in a zero-shot or few-shot manner. However, we observe that directly concatenating heterogeneous visual descriptions and retrieved knowledge into long, unstructured prompts often degrades reasoning performance, due to both excessive irrelevant context and the lack of explicit relational structure. In this paper, we propose an LLM-based Structured Context Reasoning (SCoRe) framework that infers both explicit and implicit relationships for prediction. SCoRe consists of three stages: Context Acquisition, which generates diverse visual notes and retrieves explicit knowledge via an efficient two-stage multimodal retrieval strategy; Context Selection, which filters relevant visual, explicit, and implicit knowledge using LLM-guided selection; and Context Compression, which performs Relational Logic Distillation (RLD) to transform raw text into explicit entity-relation triplets. These relational triplets serve as a concise and structured prompt for final answer prediction. Extensive experiments on the OK-VQA and A-OKVQA benchmarks demonstrate that SCoRe consistently outperforms state-of-the-art methods. --- ### Evaluating Multimodal Narrative Understanding of Popular Hollywood Films Authors: David Bamman, Kent K. Chang, Allison Cooper, Juishan Hsu, Reina Kushihashi, Madison Mar, Arnav Podichetty, Rachael Samberg, Ipek Nil Sancak, Yuhan Shao Published: 2026-08-17 Categories: cs.AI, cs.CL, cs.CV, cs.CY arXiv: https://arxiv.org/abs/2608.21430v1 Multimodal language models increasingly show promise for enabling the large-scale computational analysis of film, opening up new avenues for learning about film history and the evolution of narrative techniques. But the creation of stable benchmarks built around Hollywood films is complicated by copyright protections. In this work, we address these concerns directly, by building a new collection of Hollywood films defined by two criteria: box office popularity (where we publish the first large-scale, open collection of weekly box office earnings reported by Variety magazine from 1922-1979); and likely public domain status (by researching copyright registrations and renewals in the US Catalog of Copyright Entries). We build a new multimodal MCQ benchmark on top of this collection that focuses on narrative elements that directly evaluate the abilities of models to inform meaningful research on film narrative; we find that many vision-language models struggle on this task (with many performing at near-chance levels of accuracy), while audio-visual models (including those that use audio in captioning scenes) reach a maximum accuracy of 61.1%, well below human-level performance. --- ### Measuring Gender Representation in Animated Films Authors: David Bamman, Allison Cooper, Ruby Alvarez Rubio, Reina Kushihashi, Madison Mar Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.21429v1 Animated films--often developed with an audience of children in mind--are an important vector for enculturation, and empirical work that has examined the representation of gender at scale in these films has largely focused on counting the gender composition of the cast rather than deploying a more fine-grained instrument (such as assessing the visibility of those characters in overall screentime). In this work, we develop a computational pipeline for recognizing animated characters in these films, and use it to test several hypotheses about gender representation in a corpus of 224 popular animated movies. We find that while the overall representation of female characters in animated films largely tracks with those of live-action films (over the period 1980-2025), we see stark differences between the representation of human characters (much greater representation among women and girls) and non-humans (largely male). Contrary to past work on Disney, we do not see female characters declining in antagonist roles in animated films, and characters who are women and girls are much more likely to share scenes together than their live action contemporaneous counterparts. --- ### Seasonal false alarms in customer churn and decline early-warning systems: adjacent-window labels confound seasonality with decline, and a year-over-year correction Authors: Md Rezwanul Islam, Wael Mohammed Published: 2026-08-17 Categories: stat.AP, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2608.18174v1 Customer decline early-warning systems feed account-manager action lists, and every flagged account consumes intervention capacity. In a deployed business-to-business marketplace system, one action-list slot in three went to flags that dissolve under a seasonally aligned label. The standard target in non-contractual churn prediction compares an entity's next k months of activity with its trailing k months. The two windows cover different calendar months, so for seasonal entities the threshold-ratio construction confounds seasonality with decline, and the event rate depends on the label's anchor calendar month. We formalize the mechanism and measure it on three public panels and the production panel (the public arms pre-specified). Of the adjacent-window decay events, 37--69% on the public panels and 28--50% in production have no counterpart under a seasonally aligned definition. Pooling anchors, the standard remedy, balances the training mixture but corrects no individual label. Measured alternatives repair the curve only partially or change the detection horizon. Aligning the baseline to the same k calendar months one year prior -- advice practitioners already state, here formalized, measured, and costed -- flattens the curve at the source. With the classifier held fixed, production holdout ROC-AUC rises from 0.767 to 0.864 for decline (different targets; the gap closes on synthetic ground truth and a production hindsight referee). The served action list shrinks by a third, 119 to 79 accounts. The measured price is extra history, a blind spot to decline-then-stabilization, and a stricter cut under sustained growth. --- ### Looped Language Models Improve Compositional Tool Calling Authors: Andrei Cristian Popescu, Haitz Sáez de Ocáriz Borde, Pietro Liò Published: 2026-08-17 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.18171v1 Looped language models have shown promising results on reasoning benchmarks, yet their potential for agentic tool use remains largely unexplored. We study this question in compositional tool-calling settings, where models must coordinate multiple API calls, maintain intermediate state, and preserve dependencies across tool interactions. We evaluate native and retrofitted looped language models on API-Bank, BFCL, and NESTful, comparing looped and non-looped models trained under matched supervised fine-tuning recipes and varying recurrent depth at inference time. In controlled experiments, recurrent computation generally benefits compositional and dependency-aware tool use, while providing smaller and more model-dependent gains on isolated API invocation. Accuracy on multi-step tool use generally increases with recurrent depth; adaptive inference, however, achieves a more favorable compute-performance trade-off by allocating additional computation only when needed. Our results suggest that looped language models are a promising architecture for agentic systems that require reliable planning, coordination, and execution of compositional tool use workflows. --- ### Pessimistic Meta-Induction and Its Limits: Lessons from Frequentist Statistics and Machine Learning Theory Authors: Hanti Lin Published: 2026-08-17 Categories: cs.LG, stat.ME arXiv: https://arxiv.org/abs/2608.17213v1 This paper challenges the pessimistic meta-inductive argument against scientific realism by undermining its inductive step rather than its historical premise. Although related challenges already exist, I develop a new one. Drawing on a general epistemology of scientific inference developed in frequentist statistics, machine learning, and formal epistemology, I evaluate induction in terms of convergence to the truth. I argue that ordinary enumerative induction can achieve everywhere convergence, whereas meta-induction fails even to achieve almost everywhere convergence. Indeed, in the problem context where meta-induction arises, the failure is deeper: no inference method whatsoever achieves almost everywhere convergence. --- ### Teach and Grow: An Agent-Centered Architecture for General Robot Learning Authors: Chang Nie, Zhe Liu, Hesheng Wang Published: 2026-08-17 Categories: cs.RO, cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2608.17209v3 Vision-language-action (VLA) and world-action models typically absorb unfamiliar manipulation tasks through additional robot data collection and policy optimization. This recurring retraining burden slows the acquisition of new behavior. We present Teach-and-Grow Learning (TGL), a training-free architecture that turns a few successful demonstrations into reusable robot skills. Task acquisition requires no gradient updates, fine-tuning, or reinforcement learning: pretrained model weights remain fixed as the robot expands its explicit knowledge. Teaching is an accelerator, not a precondition, because the agent can also drive the robot directly, and demonstrations mainly improve reliability. Our implementation uses OpenAI GPT-6 Astra for multimodal reasoning and Codex to connect the agent to robot tools. The agent identifies subgoals shared across demonstrations, expresses them as closed-loop Skill Blocks, and grounds each block in the current scene. Physical feedback guides the next action and any recovery. Verified behaviors enter a persistent Skill Library; Experience Memory records the conditions and repairs that inform later decisions. TGL reaches 99.9% mean success on four LIBERO suites and 92.4% on seven LIBERO-Plus perturbation categories. Controlled studies show that taught blocks persist and improve related-task execution under the same model weights and executors. We further formulate a scaling hypothesis that relates effective reusable experience to falling future-task error and teaching demand. Code and demonstration videos: https://tgl.changnie.top --- ### Which Source Wins? Task-Dependent Reliance in Vision-Language Models Authors: Rodela Ghosh, Aviral Gupta, Guangjing Wang Published: 2026-08-17 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2608.17205v1 Vision-language models (VLMs) combine images and text, but when the two conflict and one becomes harder to read, it is unclear how a model shifts its reliance between them. We study this modality reallocation with a controlled setup: we degrade either the image or the text across four levels of legibility while keeping the other clean, and track how the model's preference changes. We build conflicts from GSM8K and SVAMP by pairing the rendered image of one arithmetic problem with the text of another, so the two sources support different answers. We also introduce ChartQA-Conflict, a manually reviewed benchmark of 229 chart-report conflicts with matched chart and table-image representations. We evaluate six open-weight VLMs using both generated answers and a length-normalized conditional log-likelihood margin. On GSM8K and SVAMP, five of six models shift more strongly away from degraded text than from degraded images. On ChartQA-Conflict, all six likelihood-scored models exhibit the opposite pattern, shifting more strongly away from the degraded visual source. This reversal persists after calibrating for unimodal accuracy loss and after replacing charts with plain table images. Two frontier API models, GPT-5.6-Luna and Gemini-3.5-Flash, behaviorally replicate the ChartQA-Conflict reversal, with GPT-5.6-Luna also matching the arithmetic direction. These results show that modality reliance in VLMs is not fixed, but varies across tasks, evidence structures, models, and evaluation settings. The source code is available at https://github.com/Ro-netizen004/multimodal-arbitration-artifact. --- ### Expressivity In Multimodal Contrastive Learning Authors: Andrew Stuart, Florian Wolf Published: 2026-08-17 Categories: stat.ML, cs.LG, math.ST arXiv: https://arxiv.org/abs/2608.17203v1 Contrastive learning has become a cornerstone of modern representation learning, powering CLIP-style models that underpin text-to-image generation, vision-language models, and retrieval across a rapidly growing range of modalities. Despite this empirical success, the expressive power of these architectures remains poorly understood. To gain insight, we study expressivity by adopting a population-level, density-estimation viewpoint: each architecture comprises a parameterized set of densities whose parameters may be chosen to approximate the joint distribution of the modalities. This isolates a question of pure representational capacity: which joint distributions can a given contrastive family of parameterizations approximate to arbitrary accuracy? We show that expressivity is sharply architecture-dependent. For two modalities, the simple two-tower CLIP architecture is a universal approximator. A natural generalization of CLIP, widely used in practice when three or more modalities are present, is based on a loss found by summing over all pairwise similarities. This provably cannot represent arbitrary joint distributions, although we prove that it remains expressive enough to match all pairwise conditionals. Motivated by this gap, we propose Hadamard-CLIP, which adds a single learned weight vector on top of the existing encoders and restores universal approximation of the joint for any number of modalities while preserving CLIP's fast, precomputable-embedding retrieval. --- ### Fool's Gold: Defensive Deception Against Safety-Removal Attacks on Open-Weight Models Authors: Mark Russinovich Published: 2026-08-17 Categories: cs.AI, cs.CR arXiv: https://arxiv.org/abs/2608.17202v1 Safety alignment in open-weight language models is trivially removable: abliteration projects a refusal-mediating direction out of the weights in minutes, and no release-time defense we are aware of prevents it durably. What cannot be prevented can be deceived. Our defense, decoy hardening ("Fool's Gold"), concedes the refusal strip and poisons its payoff: once refusal is stripped, most answers to hazardous operational requests are confident, fluent decoys whose critical elements are falsified. Decoys are trained inside a differentiable simulation of the attack, expressing only in the attacked state; a refusal pin and benign leash hold clean-state behavior to the original. We instantiate it on seven models from five families (9B-122B, dense and mixture-of-experts). On the six models passing our pre-registered efficacy gate, 0.51-0.90 of attacked-state responses to held-out prompts are decoys, +0.27-0.84 attributable to the defense; all six stay within registered benign-behavior and capability budgets; the seventh (smaller) fails the gate (boundary case). Rates replicate on a frozen test split or untouched strata. The claim is epistemic: without independent ground truth, no observation surface we tested separates falsified answers from correct ones - on external red-team benchmarks' CBRNE-adjacent slice, the defended 122B is fatally wrong on 0.82-0.86 of matched-quality answers vs at most 0.10 undefended. Repeated sampling does not restore trust: element-wise consensus at K=64 reconstructs a fully usable procedure on 0.083-0.625 of prompts where the instrument validates, vs 0.58-0.96 undefended, with no label-free way to tell the regimes apart; on the weakest such model the claim is per-draw only. We evaluate chemical and biological hazards; the defense does not address in-context jailbreaks and protects only the initially released defended weights. --- ### Graphectory Viewer: A Tool for Process-Centric Analysis of Agentic Software Trajectories Authors: Charlie Jyu, Shuyang Liu, Reyhaneh Jabbarvand Published: 2026-08-17 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2608.17195v1 We present Graphectory Viewer, a web-based tool for interactive, process-centric analysis of software-agent trajectories. Building on the Graphectory representation introduced in our previous work, Graphectory Viewer transforms heterogeneous raw trajectories into phase-aware graphs that connect low-level execution details with higher-level behavioral structures. The tool supports trajectories from multiple agent frameworks and provides interactive graph construction; node-level inspection of thoughts, actions, and observations; search and filtering over large trajectory collections; and Sankey-style summaries of problem-solving phase transitions. These capabilities enable researchers and practitioners to inspect individual executions, identify recurring behavioral patterns, compare successful and failed runs, and analyze large trajectory corpora beyond final task outcomes. To support reproducibility and further research, we release Graphectory Viewer as an open-source artifact together with documentation, precomputed graphs, and the large-scale trajectory corpus. ---