# Research Papers — 2026-08 · bundle page 141 of 240 > Full-text markdown of papers #7001–7050 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/141 [← Previous bundle](https://aipod.dev/papers/archive/2026-08/page/140.md) [Next bundle →](https://aipod.dev/papers/archive/2026-08/page/142.md) ### Scaling Automatic Research Agents via World Models Authors: Xiyuan Yang, Sheikh Sarwar, Jingru Cheng, Zhan Shi, Duanshun Li, Huiyuan Chen, Haiyang Zhang, Xing Fan, Chenlei Guo, Jingrui He, Zhenyu Liao Published: 2026-08-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.12564v3 Automating empirical research is a long-standing direction of AI. Recent automatic research (AutoResearch) agents bring this goal within reach, as modern LLMs show the capability to independently implement solutions and learn from the execution outcomes. Behind these gains, post-training (especially RL) plays a central role. In this paper, we identify a fundamental tension when scaling RL for these agents: the two components of every AutoResearch trajectory (agent generation and environment execution) scale in very different manners, since all generation shares compute through batching, while each execution occupies its exclusive sandbox and real machine time. As a result, the environment execution dominates the training cost and becomes the bottleneck as trajectories grow. To resolve this tension, we propose World Model RL (WMRL), which replaces environment execution with a world model to remove this bottleneck. Additionally, the world model can be imperfect, as its rewards are corrupted by bias and noise. Therefore, we further equip WMRL with two mitigations, Online Debiasing and Inverse-Variance Denoising, which offset the bias and suppress the noise respectively. Theoretically, we prove that both mitigations of WMRL strictly improve the convergence guarantee. Empirically, WMRL accelerates training by 3-4x on various tasks at different agent scales, while exceeding the performance of standard RL baselines. Moreover, our post-trained 4B and 9B agents outperform much larger open-weight agents of 48B and 120B on held-out benchmarks. Beyond AutoResearch, WMRL also transfers to post-training embodied VLA policies, which demonstrates the generalizability of our method. --- ### CAS: A Causal Attribution Score for Local and Global Explainable Artificial Intelligence Authors: Michael Georgiades, Charalambia Varnava Published: 2026-08-12 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.12555v1 Predictive explanation methods attribute a model output; they do not, by themselves, attribute an intervention effect on the real-world outcome. We introduce the Causal Attribution Score (CAS), a compact score architecture for causal explanation. CAS starts from an identified interventional coalition game, allocates the joint intervention contrast with causal Shapley contributions, and converts those raw outcome-scale effects into Local CAS, Signed Local CAS, and two complementary Global CAS summaries. The innovation is not a new Shapley formula, but a local-to-global causal reporting layer with an explicit intervention target. In the known-truth benchmark, eight repeated primary-interaction simulations (n = 2,200 each, three actions) gave mean Local CAS MAE of 0.107 for coalition-aware CAS, compared with 0.173 for one-at-a-time normalisation and 0.213 for a global normalised absolute ATE vector. The paired advantage over one-at-a-time normalisation increased from -0.003 under additivity to 0.091 under strong interactions. On both empirical DoubleML datasets, 401(k) eligibility/net financial assets (n = 9,915) and Pennsylvania reemployment bonus/unemployment duration (n = 5,099), predictive SHAP/TreeSHAP rankings differed materially from Feature-CAS rankings of treatment-effect modifiers. In Pennsylvania, dep1 (exactly one dependent) moved from predictive global rank 13 to Feature-CAS rank 2 and was the leading local Feature-CAS modifier. These results isolate the added value of separating what predicts the outcome from what explains heterogeneity in an estimated causal effect. --- ### StrAD: A Streaming Method and Benchmark for Audio Description Generation for Long-form Videos Authors: Julian Spravil, Sebastian Houben, Sven Behnke Published: 2026-08-12 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.12549v1 Visual content is the dominant medium of communication, yet without audio descriptions (ADs), it remains inaccessible to blind and low-vision people. ADs narrate context-relevant visual events during natural audio pauses. Manually creating ADs is expensive, limiting coverage to a small fraction of available content. Most existing automatic AD generation methods frame the task as video clip captioning, requiring ground-truth timestamps and additional context cues such as character databases. Current benchmarks reinforce this framing, consisting of short video segments paired with automatic or task-mismatched annotations. We introduce StrAD, a benchmark for long-form AD generation on full-length videos spanning diverse genres such as movies, documentaries, short films, performances, and video games. We reformulate AD generation as streaming dense video captioning. Our approach processes full-length videos with a sliding window, inserting ADs into existing transcripts without ground-truth timestamps, and supports both fine-tuned models and zero-shot prompting of vision-language models. On the segment-level task with given timestamps, our fine-tuned StrAD-FT sets the state of the art on CMD-AD with 36.3 CIDEr (+10.0 over Shot-by-shot), establishes a reference point on StrAD (51.0 CIDEr), and remains competitive on MAD-Eval at 24.9 CIDEr. On the full-video streaming task, StrAD-FT reaches a SODA score of 2.4 against 1.1 for our zero-shot baseline StrAD-Zero, though both exhibit limitations in temporal localization and narrative coherence. While prior work has tackled full-video AD generation in an offline, multi-stage fashion, ours is the first streaming approach, generating ADs on the fly without ground-truth timestamps. StrAD makes progress on full-video AD generation measurable, a prerequisite for scaling accessibility. --- ### Analysis of Motor Signatures of Social Adaptation in Autism for Efficient Human-Centric Systems Authors: Lara Pereira, Teresa Sousa, Miguel Castelo-Branco, João Ruivo Paulo Published: 2026-08-12 Categories: cs.HC, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2608.12548v1 Dance imitation integrates motor planning, sensorimotor integration, and social cognition, offering a sensitive framework to characterize motor behavior in autism. In this work, we explore a computational analysis framework to identify potential biomarkers that allow the design and development of improved medical and human-machine systems. We analyzed 3D motion capture data from autistic and neurotypical adults performing dance imitation under solo and socially-framed duo conditions. Methodologically, using Dynamic Time Warping, we quantified movement consistency and propose the Social Context Sensitivity Index (SCSI) to measure modulation of variability by social framing. These features were then used on a classifier to discriminate subjects into autistic or neurotypical groups. Results show that neurotypical adults exhibited increased movement variability in socially-framed imitation, especially in upper and lower limbs, whereas autistic adults maintained consistent movement across contexts. Classification achieved 79.2% balanced accuracy in distinguishing groups. These findings suggest that social context sensitivity in motor imitation constitutes a robust biomarker of autism-related motor behavior, highlighting the importance of social modulation in motor assessments and informing the development of inclusive human-centric technologies. --- ### Surface-to-Skeleton 3D Cephalometry: Estimating Hidden Skeletal Landmarks from CT-Derived External Soft-Tissue Surfaces Authors: Tomoki Abe, Taiki Kanaya, Kazuki Saita, Mao Noda, Chie Tachiki, Yasushi Nishii, Hideo Saito Published: 2026-08-12 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.12537v1 Existing 3D facial-landmark methods localize points on visible skin, but whether CT-defined internal skeletal landmarks can be inferred from external soft-tissue geometry remains unclear. We formulate a coordinate-consistent surface-to-skeleton task using same-acquisition CT-derived surfaces, separating estimation from optical-to-CT registration, scanner-domain, and acquisition-state effects, with coverage analyzed separately. From 240 clinical CT scans from two hospitals, we construct a locked retrospective protocol pairing CT-derived external soft-tissue point clouds with 21 skeletal landmarks and three visible soft-tissue landmarks. An integrated hierarchical point-cloud model achieves 2.97 mm mean radial error on skeletal landmarks and 3.03 mm on deep or surface-invisible landmarks in 40 held-out patients. Patient-mismatch controls support patient-specific signal beyond a fixed population configuration or global similarity alone, while coverage ablations indicate dependence on non-anterior geometry. Optical-transfer diagnostics reveal substantial coverage-related and global-configuration components, although deployable optical inference remains unresolved. These results answer the controlled feasibility question affirmatively and provide a basis for hidden skeletal landmark inference. --- ### GENADA: efficient generative time series adversarial attack framework Authors: Michael Baronov, Denis Vorobev, Margarita Rusanova, Petr Sokerin, Alexey Zaytsev Published: 2026-08-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.12535v1 Deep learning models are widely used for time series analysis in domains such as healthcare, finance, energy systems, and environmental monitoring. However, these models remain vulnerable to adversarial attacks, where small input perturbations cause severe degradation in predictive performance. Commonly used gradient-based attacks, iterative first-order methods, are computationally burdensome, as they repeatedly backpropagate through the victim model to compute input gradients during a number of iterative refinement steps. We propose a GENerative ADversarial Attack (GENADA) that learns a generative model to produce deceptive perturbations directly in a single forward pass and a procedure to train it. Variants include single-step and iterative generative attack schemes. The validation considers attacks on several neural models and datasets in the time-series domain, a controlled, low-dimensional setting. Empirically, GENADA achieves comparable attack quality to strong baselines while requiring less time to generate perturbations during inference. --- ### MASCOT: Model-Aware Submodular Coverage for Composite-Attribute Text-to-Image Retrieval Authors: Aaryan Sharma, Vishak Prasad C, Virendra Singh, Ganesh Ramakrishnan Published: 2026-08-12 Categories: cs.MM, cs.CV, cs.IR arXiv: https://arxiv.org/abs/2608.12532v1 Vision-Language Models (VLMs) are highly effective in retrieving semantically relevant images. However, in practice, relevance alone is often insufficient. Systems must also achieve Result Diversification (RD) across composite attributes such as geography and time, a task for which precise control remains challenging. Current re-ranking methods, such as Multi-Source Determinantal Point Processes (MS-DPP), address this using manifold-based repulsion over similarity representations. Although this strategy is effective for broad exploration, it exposes a key limitation in manifold-based models: when subjected to diversity-decrease tasks on discrete metadata, they suffer substantial degradation in early-rank recall. To bridge this gap, we introduce MASCOT (Model-Aware Submodular Coverage for Composite-Attribute Text-to-Image Retrieval). Instead of relying on manifold repulsion, MASCOT formulates multi-attribute diversity as a resource allocation problem, projecting attributes into a soft-binning space weighted by query-driven importance. Averaged across the three PixelProse diversity-decrease tasks, MASCOT preserves an early-rank recall (R@10) of 88.58%, while MS-DPP retains 67.63%. The margin widens under composite constraints: on PP_geo_hour, where temporal and geographic diversity must be suppressed simultaneously, MS-DPP's recall collapses from 0.9737 to 0.4931 and its top-ranked result degrades to R@1 = 0.23, while MASCOT holds R@10 = 0.9410 and R@1 = 0.7202 at a diversity metric above the unconstrained baseline. We do not claim uniform superiority: on aggregate diversity-relevance scores our own simpler ablations attain higher harmonic means on all three decrease tasks, and MASCOT's advantage is specific to recall beyond rank 1 under composite constraints. --- ### SchemaLink: An Intelligent Web Editor for LinkML Schema Curation Authors: Emanuele Cavalleri, Paolo Perlasca, J. Harry Caufield, Justin Reese, Christopher J. Mungall, Marco Mesiti Published: 2026-08-12 Categories: cs.DB, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2608.12529v1 Motivation: LinkML is a suitable language for the representation of the structural and content constraints of different kinds of biomedical data. Even if it is a quite recent proposal, it has been applied in several biomedical contexts. Developing and maintaining LinkML schemas presents several challenges, particularly for novice curators. Non-expert bio-curators may struggle with LinkML syntax and best practices, requiring significant time and effort to develop well-structured schemas. Results: In this paper we propose SchemaLink, a web-based environment for the graphical construction and enhancement of LinkML schemas that address the following requirements: $(i)$ introduce a graphical language for the specification of LinkML schemas, $(ii)$ make uniform the specification of schemas in similar contexts, $(iii)$ simplify the design and curation processes by exploiting a RAG-based approach to assist curators in creating new schemas from scratch and editing already developed ones. Several experimental analyses show the quality of the produced LinkML schemas through the AI-based editing facilities. Availability and Implementation: SchemaLink is available online at: https://SchemaLink.biodata.di.unimi.it. SchemaLink code and testing data are available as open-source on GitHub at: https://github.com/AnacletoLAB/{schemalink-webapp,schemalink-api}. --- ### $\varepsilon$-MemEvo: Adaptive Cross-Task Memory Transfer for LLM Program Evolution Authors: Aofan Liu, Shiyuan Song, Yiyan Qi Published: 2026-08-12 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.12522v1 LLM-based program evolution systems such as FunSearch and AlphaEvolve have shown strong ability to discover novel algorithms, but typically optimize each task in isolation, discarding search experience after completion. We introduce $\varepsilon$-MemEvo, a framework for cross-task knowledge transfer in LLM program evolution. $\varepsilon$-MemEvo stores prior experience as task-agnostic tactic memories: compact natural-language summaries of successful algorithmic strategies rather than raw code, enabling transfer across tasks with different APIs and evaluators. To avoid negative transfer from semantically mismatched memories, $\varepsilon$-MemEvo uses an adaptive injection gate that decides whether retrieved memories should be injected, and at what intensity. We evaluate $\varepsilon$-MemEvo on 8 diverse optimization benchmarks spanning mathematical optimization and systems engineering, using a content-level Leave-One-Out protocol that excludes target-task memory entries. On the primary GPT-5 backbone, $\varepsilon$-MemEvo improves AUCC over AdaEvolve on all 8 tasks, with a mean relative gain of +8.7%, and improves early-stage convergence by +9.4% on average. Ablations show that naive memory injection can fail catastrophically, while adaptive gating remains safe across all five ablation tasks. The data-updated posterior is interpretable in observed states: it favors skip during improving search and shifts from skip to hint across early and late plateaus. These gains incur less than 1% computational overhead. --- ### Can Vision-Language Models Assess Proxemic Risk from Egocentric Robot Images? Authors: Vladyslava Rudas, Dmytro Kuzmenko Published: 2026-08-12 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2608.12515v1 Assessing proxemic danger from a robot's egocentric perspective is critical for safe embodied navigation in human environments and requires both visual and contextual reasoning. We evaluate three opensource vision-language models (VLMs) (\textit{InternVL}, \textit{Qwen-VL}, and \textit{SmolVLM}) on the classification of egocentric robot images into four danger levels, comparing three prompting strategies and two rounds of QLoRA fine-tuning against a stratified random baseline. Without fine-tuning, all models perform near the baseline, while fine-tuning yields only modest overall improvements. However, \textit{Qwen-VL} with an advanced prompt achieves substantially higher recall for high-danger cases than the other models. An analysis of person localization further shows that correct danger classification does not correspond to better spatial grounding, indicating that a model may produce a useful safety label without attending to the relevant region of the scene. These results show that current VLMs remain limited in fine-grained proxemic reasoning and spatial grounding, although targeted prompting and fine-tuning can improve high-danger detection in selected models. --- ### Exploring Oversmoothing with Householder Matrices Authors: Bhaskar Karol Published: 2026-08-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.12514v1 Deep graph neural networks(GNNs) suffer from oversmoothing- a progressive collapse of node representation towards a low information subspace as network depth increases because the normalized graph propagation operator is repeatedly applied directly to the hidden representations. In this work we study Householder Graph Neural Network (HouseGNN). Rather than updating the hidden state like standard GCN, HouseGNN uses the aggregated neighbourhood message solely to estimate a reflection direction; the node embedding is then updated by a Householder reflector followed by GroupSort, yielding a piecewise orthogonal layer that preserves Euclidean norm at every node and at every depth. We prove three core properties: (i) every internal layer preserves the node-wise Euclidean norm; (ii) the Householder reflector is scale scale and sign-invariant in the message; and (iii) pairwise distance between nodes can change through mismatch between node-wise orthogonal operators. --- ### SoK: From Generation to Consumption of Privacy Documents in Software Systems Authors: Shidong Pan, Clark LaChance, Zhen Tao, Sepideh Ghanavati Published: 2026-08-12 Categories: cs.CR, cs.CL arXiv: https://arxiv.org/abs/2608.12511v1 Privacy documents (e.g., privacy policies) are a central mechanism through which digital services disclose data practices and seek user consent. Over the past decades, research on privacy documents has expanded significantly, encompassing not only traditional privacy policies but also short notices (e.g., privacy labels) and interface-level transparency mechanisms. As this research area continues to grow, it has become increasingly difficult to obtain a coherent view of how privacy documents are created, analyzed, evaluated, and maintained across their lifecycle. This SoK provides a unified, lifecycle-oriented view of privacy documents from a software engineering perspective. We systematically review and analyze 290 papers published between 2010 and 2025, organizing them around five research questions that examine how privacy documents are (1) defined and scoped, (2) generated, (3) analyzed and extracted, (4) checked for inconsistencies and noncompliance, and (5) evaluated and improved for usability. Building on our findings, we identify 15 key research trends and 21 open opportunities. We further chart four broader research directions that highlight (i) emerging challenges in AI-centric platforms, (ii) the need for diverse and up-to-date data foundations, (iii) LLM-based unified policy-code analysis, and (iv) dual usability for end-users and developers. We hope this SoK provides a shared foundation for future research on privacy policies and privacy documents. --- ### Fast Length-Squared Sampling for Positive-Semidefinite Matrices Authors: Rajarshi Bhattacharjee, Ethan N. Epperly, Cameron Musco, Aaron Tian Published: 2026-08-12 Categories: cs.DS, cs.LG arXiv: https://arxiv.org/abs/2608.12503v1 We describe a simple rejection-sampling-based algorithm to perform length-squared sampling on an $n \times n$ positive-semidefinite (psd) matrix: that is, to sample a column with probability proportional to its squared $\ell_2$-norm. The algorithm runs in just $O(n)$ expected time, which is significantly sublinear in the input matrix size. The runtime is optimal, even when the input is assumed to be diagonal. Our result has several applications. Length-squared sampling is used by a number of sublinear time algorithms for matrix problems, like low-rank approximation and eigenvalue approximation. Often, it is assumed that the algorithm is given access to the matrix column norms, and thus can perform length-squared sampling efficiently. Our result shows that, at least for psd matrices, we can remove this assumption. We also discuss an application to an asymptotically optimal algorithm for estimating the Frobenius norm of a psd matrix to relative error. Finally, we show that our sampling algorithm yields a very simple sublinear time algorithm for the robust psd low-rank approximation problem introduced by Bakshi et al. (FOCS, 2020), which nearly matches the more complex method developed there. --- ### HIMEC: Directional Change Representation and Fixed-Interface Decoding for Remote Sensing Image Change Captioning Authors: Aysha Ashraf, Shaina Ashraf, Wafaa I. M. Hussin, Ali Haider, Zhi Lu, Zhenming Peng Published: 2026-08-12 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.12502v1 Remote sensing image change captioning (RSICC) converts bitemporal imagery into a sentence describing semantic changes. Most RSICC methods condition caption decoders directly on fused visual features, leaving intermediate change structure and decoder-interface consistency less studied. We present HIMEC, combining Directional Change Representation (DCR) with fixed-interface decoding. DCR separates signed differences into appearance-oriented, disappearance-oriented, and shared-context streams before fusion. A learned-query encoder converts the fused representation into visually conditioned change-query tokens that form the scene decoder's only sample-dependent memory. A training-only auxiliary phrase decoder supplies caption-derived supervision. With a fixed zero input, the scene decoder maintains the same interface during training and inference. Separately, we evaluate a local-to-scene cascade conditioned on teacher-forced local states during training and autoregressive states at inference. On changed LEVIR-CC validation pairs, these states have a mean cosine distance of 0.69. Regime-matched conditioning recovers most of the associated deficit, whereas permuting state correspondence causes no detectable penalty. These findings are limited to the evaluated cascade. In a matched three-seed comparison, HIMEC reaches a Consensus-based Image Description Evaluation (CIDEr) score of $142.81\pm0.60$ on LEVIR-CC, versus $139.51\pm3.40$ for direct fused-feature memory. On SECOND-CC, fixed-zero and regime-matched diagnostic conditioning reach 75.67 and 76.99 CIDEr, respectively, versus 60.77 for the mismatched cascade. The source code will be made publicly available at https://github.com/ayshaashra/HIMEC upon publication. --- ### When Can You Trust Offline Evaluation of Equal-Cost Top-k Allocation? A Controlled, Reproducible Benchmark and Practitioner's Guide Authors: Binshuang Li Published: 2026-08-12 Categories: cs.LG, stat.ME, stat.ML arXiv: https://arxiv.org/abs/2608.12489v1 Organizations decide whom to treat under a budget and want to know what a targeting rule would have earned before deploying it. Off-policy evaluation promises this from logged data, but the deployable rule is a deterministic top-k policy: it removes all averaging over actions, so weak overlap hits the estimate directly. We benchmark six estimators across five datasets and two known-effect sweeps, and validate the mechanisms against a non-simulated paired reference. First, weak overlap is governed by logger-target action alignment, not by logging sharpness alone: what governs support is the logger's probability of the target's actions. Sharpening a logger built from the target's own score barely moves overlap over the tested range; action-level disagreement collapses it. Effective sample size ranks this risk across logging environments, but is weak at ranking candidates within the single log a practitioner holds, and its cut point does not transfer. Second, the optimizer's curse is not fixed by cross-fitting the outcome nuisance. When the rule is fit on the data used to evaluate it, cross-fitting the nuisance alone leaves the reuse bias in place and makes it worse. Honest policy-level splitting avoids the reuse by targeting the learning procedure's value -- a change of estimand, not a de-biasing of the full-sample policy. Third, propensity-estimation error is the largest degradation we measure: an out-of-fold estimate hurts IPS more than any other stress we apply, leaves doubly-robust estimation almost unchanged, and can invert the overlap diagnostic itself. Logging is synthesized and propensities floored at 0.02, so every failure occurs with bounded weights; the floor also reduces the two tuned hybrids to their untuned parents, leaving four practically distinct estimators, and all exact-value surfaces are synthetic or semi-synthetic. We release the benchmark; public data only. --- ### DIVE: Unlocking Self-Improvement in Frozen Language Models Through Diversity-Driven Skill Evolution Authors: Siheng Xiong, Ali Payani, Oguzhan Gungordu, Faramarz Fekri Published: 2026-08-12 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.12486v1 Large language models (LLMs) cannot retain post-deployment experience without parameter updates. We introduce DIVE, a diversity-driven framework that enables frozen LLMs to improve by evolving persistent natural-language skills from task experience and verifier feedback. These skills encode reusable reasoning procedures, verification strategies, common failure modes, and output constraints and are both executed and revised by the same underlying model without access to a teacher model. Since natural-language skill evolution is a stochastic, non-convex search process, optimizing a single skill trajectory can overfit to sampled experience or converge to a suboptimal solution. DIVE mitigates this optimization variance by independently evolving multiple skill populations from bootstrapped experience, adaptively refining them through diverse transformations, and jointly selecting a complementary set of skills. Across six mathematical and logical reasoning tasks and multiple model families, DIVE consistently outperforms existing reasoning methods, prompt-optimization approaches, skill-development frameworks, and memory-based baselines. It achieves rapid self-improvement from accumulated experience, obtaining substantially larger performance gains with fewer rollouts than parameter-based methods such as SFT and GRPO, and prompt optimization with GEPA. Further, the resulting skills transfer across model scales and families, enabling smaller models such as GPT-5-nano to match or outperform larger counterparts, i.e., GPT-5, under conventional prompting. These results establish diversity-driven skill evolution as an effective, interpretable, and parameter-free approach to LLM self-improvement. --- ### Learning Under Treatment-Induced Label Indeterminacy with Expert Annotations of Counterfactual Outcomes: A Case Study in Neurological Prognostication Authors: Xiaobin Shen, Chloe Y. H. Huang, Jonathan Elmer, George H. Chen Published: 2026-08-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.12477v1 Clinical prediction models are often developed as if the outcome of interest were cleanly observed for every patient. This assumption fails when treatment decisions make the clinically relevant outcome permanently unobservable. As a case study of this problem, we consider post-cardiac-arrest neurological prognostication using a cohort of 2,497 patients, including 1,429 patients whose outcomes were rendered indeterminate by treatment decisions. These patients with indeterminate outcomes were reviewed by independent clinical experts, who provided their guesses of counterfactual outcomes about what would have happened to the patients. We refer to these patients as uncertain cases. We also have patients for whom we observe their clinically relevant outcomes; we refer to these patients as certain cases. We propose a framework for evaluating prediction models that explicitly splits the evaluation between certain and uncertain cases. Here, we cannot easily evaluate both types of cases in a uniform manner as the available target labels differ. We then propose a simple prediction model that uses target labels from both certain and uncertain cases in a manner that allows us to trade off between them. Across the proposed neural model and a collection of tabular baselines, models with similar certain-case AUROC can nevertheless differ substantially in both certain-case Brier score and their probability estimates for uncertain cases. Improving alignment with target labels of uncertain cases for our proposed model generally comes at the cost of worse accuracy on certain cases, highlighting an explicit tradeoff that standard evaluation conceals. These results show that when treatment decisions determine whether clinically meaningful outcomes remain observable, conventional evaluation metrics can miss important failure modes in the very patients for whom prognostic support matters most. --- ### Governed Persistent Memory: Source-Bound State Semantics and Fail-Closed Release for Long-Horizon Agents Authors: Guodong Xu Published: 2026-08-12 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.12476v2 Long-term agent memory is usually treated as select--store--retrieve, but retrieval does not decide whether contradictory, superseded, retracted, deleted, or stale records may support an outgoing claim. We introduce Governed Persistent Memory (GPM), an auditable bitemporal state-transition model with source-bound admission, derived lifecycle state, current public barriers, and fail-closed structured release. Five executable clauses cover ledger integrity, source binding, conflict isolation, non-revival after retraction or deletion, and exact claim closure over a fresh view at one verified head. On a prespecified hash-frozen 3,600-case GPM-ReleaseBench, GPM matches all complete outcomes; the strongest of three intentionally simple complete policies matches 1,800/3,600 and makes unmatched releases on 50% of violation cases. A separate sealed end-to-end service evaluation exercises real ingestion and release across eight query families. In its publicly disclosed V3 arm, the governed lane is correct on 2,400/2,400 clusters versus 600/2,400 for ungoverned local Qwen2.5-7B; it repairs all 1,800 baseline failures with no regression (one-sided 95% lower bounds 99.875% and 99.834%). A later V5 reseal over Chinese- and English-command arms, with generation-date pinning and no post-freeze reducer amendment, again obtains 2,400/2,400 per arm. A production-code-independent finite model explores 331,776 semantic and 1,990,656 query states without a full-contract counterexample, and a 100,000-trace three-engine differential yields zero mismatches. These are bounded contract and implementation results, not open-world model accuracy or evidence of world truth. Governed answers in the sealed service evaluation are deterministic service outputs; the 7B result is the ungoverned comparison, not a claim that a language model itself became perfectly accurate. --- ### Don't Cut Corners: How Training Outside the Prior Makes Simulation-Based Inference More Robust Authors: Chaipat Tirapongprasert, Matthew Ho Published: 2026-08-12 Categories: astro-ph.IM, stat.ML arXiv: https://arxiv.org/abs/2608.12470v1 Large astrophysical simulation campaigns often generate training data by sampling parameters across a Uniform prior box. Due to the proposal's sharp edge, neural posterior estimators struggle to learn accurate approximations near the boundaries. We propose Tailed-Uniform, a family of hybrid proposal distributions for sampling training simulations for robust simulation-based inference. By padding the original hard-truncated training box with decaying tails, Tailed-Uniform-trained networks yield more accurate posteriors near and beyond the edges. We demonstrate these improvements on a family of tail shapes, including a widened Uniform box as a control. Our results suggest that additional simulations near the prior boundary better constrain the networks as it approaches the edge of the training box, even for Uniform assumed priors. We show these advantages on a toy problem and cosmological parameter inference from the matter power spectrum. These benefits increase in high dimensions, where boundaries dominate parameter space volume. --- ### Exemplar-based objective classification of gust-induced loads across multiple flight conditions Authors: Paolo Olivucci, Kowshik Srivatsan, David E. Rival Published: 2026-08-12 Categories: cs.LG, physics.data-an arXiv: https://arxiv.org/abs/2608.12448v1 Is it possible to find an objective classification criterion that organizes the complexity of gust-induced loads across many flight conditions? And one that remains as interpretable as a labelling based on coarse parameters, such as the flight attitude? Our approach encodes a large number of experimental observations through a machine-learned representation and applies a summarization procedure to select a minimal subset of highly significant exemplars. The exemplars provide a similarity-based objective classification criterion of all the observations, they can be more conveniently inspected by experts and can become subject of more refined experiments. We demonstrate the approach on a database of 3480 pressure-load measurements induced by random gusts on a flying-wing model across six flight attitudes. We find nine fundamental response types that recur across multiple attitudes; analysis of a type's transient response enables physical intuition into the underlying fluid mechanics. --- ### Geometric and Behavioral Stratification in Transformer Residual Streams Authors: Nelson Guda Published: 2026-08-12 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2608.12447v3 Trained transformer models develop privileged bases: coordinate axes whose statistics differ from the rest of the residual stream. But what kind of direction does such a basis select? We investigate the prediction direction, the unembedding direction of the token a model currently predicts, and find that it functions as a content-defined privileged anchor. Measured with respect to this anchor, residual-stream variation is geometrically and behaviorally stratified by proximity to the prediction. The stratification holds in all eighteen models tested (dense and mixture-of-experts, 7B-120B, base and instruction-tuned). A narrow, scale-invariant prediction interface concentrates readout-relevant structure, while the vast prediction-distal complement expands with model scale. Because the prediction direction sits nearly orthogonal to the principal variance axes, variance-based analyses recover this organization only partly, and the shortfall grows with prompt heterogeneity. Anchoring reveals a steep geometric gradient: prediction-proximal regions are highly structured and cluster related prompts, while the complement is flatter and anti-discriminates among prompt groups. The interface is a narrow slice but functionally decisive. Disrupting the variance directions closest to the prediction causes immediate divergence and frequent task-frame shifts; disrupting the next level down delays divergence and preserves framing. The complement is weakly readout-aligned per direction yet causally and temporally load-bearing, and behavior is driven by direction rather than magnitude. These results establish the prediction direction as a privileged anchor distinct from previously described coordinate axes, and give a geometric account of how high-dimensional computation coexists with linear readout. --- ### Personalized Scorer Modeling: A Learning-Based Framework for Deriving Robust Sleep Stage Labels from Multiple Experts Authors: Seyyed Ali Hoseini, Javad Baseri, Hamid Saadatfar, Edris Hoseini Gol, AmirHossein Eshghi Published: 2026-08-12 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.12446v1 Sleep stage classification is important for the diagnosis and management of sleep disorders, yet most automatic staging studies evaluate models against a single reference hypnogram despite known inter-scorer variability. This study investigates whether multi-scored datasets can be used to construct more reliable reference labels from the collective behavior of multiple experts. We use the publicly available DOD-H and DOD-O datasets. EEG (C3-M2) and chin EMG signals were segmented into 30-s epochs, and 30 features were extracted from each modality, yielding 60 features for EEG+EMG. We propose a learning-based hypnogram (LBH) that models the stage-specific behavior of each scorer using confusion matrices derived from machine-learning models. After column normalization, these matrices estimate the probability of each true sleep stage given each scorer's label; probabilities are aggregated across scorers to assign the final label for each epoch. LBH was evaluated with random forest, support vector machine, and multilayer perceptron classifiers under EEG-only and EEG+EMG settings, and compared with the dataset hypnogram (DH) and best-scorer hypnogram (BSH). LBH consistently improved overall performance. The best results were obtained with random forest and EEG+EMG, reaching 86.07% accuracy, 85.46% precision, and 85.29% F1-score on DOD-H, and 86.04% accuracy, 85.21% precision, and 84.70% F1-score on DOD-O. These findings suggest that personalized scorer modeling can improve reference hypnogram construction without discarding information from individual experts. --- ### Non-Degenerate Risk Certification for Automated Security Decisions: A Decision-Contract Theory with ATT\&CK-Aligned Triage as a Worked Instance Authors: Zhenpeng Li Published: 2026-08-12 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2608.12444v1 An unconditional risk bound on automated decisions can be satisfied without automating anything, since a selector that never acts drives the bound to zero. We show this is structural: any risk certificate is defined over a decision contract, the inputs a system acts on plus the semantic relation under which an output counts correct, and weakening either hides base-classifier error. We develop a decision-contract theory: an error-conservation law showing error is only reassigned among harmful automation, human deferral, and semantic masking; a label-free singleton capacity certifying structural incapacity, with a risk-feasible refinement separating recoverable threshold misalignment from risk-constrained incapacity; and a non-degenerate actionability certificate excluding all-abstain solutions by construction. We instantiate this on ATT\&CK-aligned alert triage for LLM-based intrusion detection, the setting that exposed the vacuity failure. Across 3 IDS datasets, 6 LLMs, and 4 error-rate thresholds, empirical false-attribution risk stays at or below target in 90.3% of configurations, with 83.4% mean correct automation. The capacity diagnostic explains every low-utility configuration; its refinement separates genuine misalignment from risk-constrained incapacity, confirmed by an exhibited alternative threshold; a training-stability re-run finds no confirmed structural-incapacity instance; and real fine-grained attack-subtype labels confirm the coarsening-transfer identity under a genuine many-to-one map, with small but non-zero masking mass. --- ### SSPO: Structure-Aware Similarity-Weighted Preference Optimization for Neural Combinatorial Optimization Authors: Yuanyu Li, Jintao Xu, Zijiang Liu, Yongzhi Qi, Ningxuan Kang, Jianshen Zhang, Wei Qi, Chen Xie, Zuo-Jun Max Shen Published: 2026-08-12 Categories: stat.ML, cs.AI, cs.LG, math.OC arXiv: https://arxiv.org/abs/2608.12443v1 Neural combinatorial optimization (NCO) relies on parallel solution sampling for training, yet existing methods fail to fully exploit the rich information latent in a co-sampled solution group. Preference-optimization methods anchor on the single best solution and discard fine-grained quality and structural signal from all other peers-a failure we term gradient signal polarization. Mean-based baselines instead weight peers uniformly, so structurally near-identical peers flood the baseline with redundant information and keep gradient variance high-a failure we term baseline redundancy. We propose SSPO (Structure-Aware Similarity-Weighted Preference Optimization), which scores all $B$ sampled solutions jointly through a dissimilarity-weighted leave-one-out baseline: structurally distinct peers receive higher weight, resolving both failures in a single mechanism. The baseline uses zero-parameter, problem-adaptive solution embeddings built from the encoder's existing node representations. Experiments on TSP, EFL, and JSP benchmarks show consistent gains over prior best-anchor and uniform-weight baselines. A direct comparison against uniform RLOO on TSP and EFL confirms that structure-aware weighting is the primary driver of improvement. The SSPO-trained EFL policy has been deployed in a production facility-location system at JD$\mathord{.}$com, confirming practical viability at scale. --- ### MV2: Multi-View Multi-Vehicle Driving Dataset for Novel View Synthesis Authors: Sanjay Bhargav Dharavath, Hanvitha Saraswathi Mukkamala, Faizan Farooq Khan, Ioannis Kakogeorgiou, Aditya Arun, C V Jawahar, Zakaria Laskar Published: 2026-08-12 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.12442v2 Differentiable rendering has advanced novel view synthesis (NVS), yet applying it to real-world driving remains difficult due to sparse capture viewpoints, dynamic objects, and limited multi-trajectory data. We introduce the Multi-View Multi-Vehicle (MV2) dataset and benchmark for evaluating NVS models under large viewpoint changes in dynamic urban scenes. MV2 features synchronized captures from a car, scooter, and drone, each following distinct yet synchronized trajectories. Training NVS methods on one vehicle's camera stream and testing on another enables evaluation under substantially larger viewpoint variations than existing single-trajectory datasets. All sequences are registered via Structure-from-Motion and camera poses verified using manual pixel-level correspondence annotations, yielding 50 high-quality scenes with 12000 images. Benchmarking recent NVS and camera pose estimation methods shows that NVS performance degrades with increasing viewpoint disparity, and that feed-forward pose estimators notably lag behind optimization-based approaches, highlighting MV2 as a rigorous testbed for NVS in driving. The dataset, benchmark protocol, and project resources are available at https://mv2-dataset.github.io/. --- ### Dual Spatial-Temporal Attribution: Architecture-Aligned Post-Hoc Explainability for Recurrent Graph Anomaly Detection Authors: Iyad Assaad Nekka, Hamida Seba, Khaled Walid Hidouci, Karima Amrouche Published: 2026-08-12 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.12441v1 Deep learning detectors for anomalies in dynamic graphs have reached strong accuracy, yet they remain opaque: when an edge is flagged, the analyst receives a score but no reason. This opacity is untenable in the cooperative, regulated information systems where such detectors are deployed, where automated decisions must be auditable and trustworthy. We address this gap for AddGraph, the foundational GCN+GRU framework for edge-level anomaly detection in dynamic graphs, which to our knowledge has never been equipped with any form of explainability. We present a strictly post-hoc explainability framework, X-AddGraph, built on a Dual Spatial-Temporal Attribution (DSTA) mechanism whose three components are each aligned with one of AddGraph's architectural modules: a gradient-based relevance attribution over the current adjacency structure (spatial), a direct reading of the contextual attention weights already computed during inference (short-term temporal, at zero additional cost), and a gradient rollback through the recurrent hidden states (long-term temporal). Because the detector is frozen, detection performance is preserved exactly (Delta AUC = 0, verified empirically to ten decimal places). On the UCI Message benchmark, our trained AddGraph baseline reaches an average per-snapshot AUC of 0.8705, exceeding the originally published result; X-AddGraph reproduces every score identically while adding explanations where none existed. Evaluated across four edge populations - confident true positives, low-confidence true positives, false positives, and random samples - the long-term attribution identifies historical snapshots carrying significantly more counterfactual signal than random selection (0.127 vs. 0.074), a capability that no spatially-blind explainer can provide. We release our implementation for full reproducibility. --- ### Specification-first convergence with an AI coding agent: a case study of dismantling a core architectural invariant across 189 files in a 717k-line codebase with no test oracle and no human code review Authors: Joel Abenhaim Published: 2026-08-12 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2608.12440v2 This paper reports a single, fully instrumented case study of a large-scale architectural refactoring by an AI coding agent under a specification-first protocol, with no human review of the generated code and no pre-existing oracle to validate the target behaviour. The task, dismantling a central invariant across a large interdependent codebase, was assessed by the author as effectively infeasible through incremental refactoring, the kind of change that conventionally calls for a rewrite instead. Under the protocol described here, the agent completed it successfully. The system is a 717,725-line production TypeScript application across 3,648 files. The task required dismantling a core lifetime invariant: the guarantee that a UI panel remains open for the duration of an AI request. The target behaviour was that a streaming generation survives the closing of its panel and can be reattached, on reopening, to the same live stream with no loss or duplication. The protocol: formal specification by the agent, 14 refinement cycles auditing that specification against the source code, atomic implementation, a compile/test feedback loop, then 17 verification cycles auditing the code against the frozen specification. Across 31 audit passes, 201 defects were corrected before any human executed the program. The convergence criterion was empirical: two consecutive verification passes returning zero findings. The change touched 189 files (31 new); with the extraction phase, the two commits total 288 files, 34,770 insertions, 16,422 deletions. Across the first and roughly thirty later sessions, the software behaved as specified, no bug observed. Elapsed: three days; cost: USD 2,430. The full specification and raw session logs, 1,500+ pages in French, are published as evidence, allowing inspection of the process and submission to a language model for consistency checking. --- ### Unifying Generative Models with Path Integrals Authors: Ramon Winterhalder Published: 2026-08-12 Categories: cs.LG, hep-ph, stat.ML arXiv: https://arxiv.org/abs/2608.12438v1 We formulate generative modeling as a path integral in which flow-based, diffusion-based, variational, and adversarial models arise as different evaluation principles for a single master action. Its Martin-Siggia-Rose-Janssen-de~Dominicis (MSRJD) form separates free from interacting probability flows and opens them to diagrammatic perturbation theory. The expansion yields a one-loop correction to deterministic samplers at no stochastic-sampling cost, which we validate on solvable and nonlinear drifts, where it reduces a 53 % tree-level error to 1.6 %. Imperfect learned scores enter as insertions and yield a response-weighted score-matching objective, and symmetry-equivariant drift design becomes an operator expansion with EFT power counting. --- ### Multi-AUV Ad-hoc network-based Target Tracking: A Value Gradient Guidance Multi-Agent Diffusion Reinforcement Learning Approach Authors: Jiaao Ma, Chuan Lin, Guangjie Han, Shengchao Zhu, Qian Zhu, Ying Liu, Zhenyu Wang Published: 2026-08-12 Categories: cs.LG, cs.MA, cs.NI arXiv: https://arxiv.org/abs/2608.12436v1 Multi-AUV ad-hoc network-based target tracking requires networked autonomous underwater vehicles (AUVs) to cooperatively track maneuvering targets under constrained acoustic communication, dynamic topology, and uncertain ocean disturbances. Although multi-agent reinforcement learning (MARL) enables decentralized coordination through centralized training, existing methods suffer from high-dimensional joint state-action modeling, noise-sensitive policy generation, leading to unstable training and degraded tracking. To address these issues, we propose VGG-MADiffRL, a value-gradient-guided multi-agent diffusion RL algorithm, and MDCA, a diffusion?based hierarchical control architecture. Leveraging underwater mission characteristics, we model sonar detection mechanisms and ocean current disturbances, formulating cooperative tracking for multi-AUV ad-hoc networks as an MDP. The proposed MDCA constitutes a three-tier closed-loop control framework: a global intelligent control layer, a local online training layer, and a physical action execution layer. This structure enables synergistic optimization across task allocation, local decision processes, and execution feedback. Within MDCA, the local online training layer is the policy learning framework; VGG-MADiffRL builds on diffusion policies and incorporates value gradients to guide action generation in the reverse denoising process, steering the generated actions towards higher expected returns. It employs twin value networks with joint optimization and soft target updates to mitigate overestimation and training oscillations, promoting more stable convergence. Experimental results show that VGG-MADiffRL consistently achieves faster convergence, higher tracking accuracy, and smoother training dynamics in cooperative tracking scenarios, validating its effectiveness and practical engineering value in dynamic underwater settings. --- ### MARCH: Scaling Recurrent Memory with Content-Routed State Anchors Authors: Ming Zhang, Kaisen Yang, Shu Yu, Ermo Hua, Ning Ding, Xia Hu, Bowen Zhou, Chaochao Lu, Youbang Sun Published: 2026-08-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.12435v1 Transformers owe much of their strong long-context retrieval capability to a token-level memory that grows with context length. This flexibility, however, incurs a quadratic computation complexity during training and a key--value cache that grows linearly during autoregressive inference. Recurrent alternatives offer efficient decoding by compressing the entire history into a fixed-size state, but often underperform on recall-intensive tasks since earlier associations usually get overwritten by subsequent updates, and only the most recent contextual information is retained. In this paper, we introduce Memory-Anchor Routing across Context History (MARCH), a network architecture that effectively scales state-space models beyond a fixed-size dimension, while maintaining computational efficiency over long-sequences. MARCH periodically caches cumulative recurrent-state checkpoints as state anchors and associates each anchor with a compact, content-conditioned anchor key. This lets MARCH maintain a memory bank, which can grow as context length increases, providing a controllable trade-off between historical resolution and memory cost. At each token, MARCH produces an anchor query to attend all causally available state anchors, and the output is calculated as an attention-style aggregation over all historical anchors along the current state. We show that after standard pretraining, MARCH consistently outperforms multiple linear attention variants across commonsense reasoning, LongBench, and in-context retrieval. These results demonstrate that content-routed state caching substantially strengthens recurrent long-range memory while preserving its native computation path. --- ### SynWeaver: Website-Prior Task and Trajectory Co-Synthesis for Web Agents Authors: Ruitao Wang, Yuwen Hao, Menglin Yang Published: 2026-08-12 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2608.12429v1 Web agents often struggle to generalize to unseen websites because they lack website-specific supervision. Recent exploration-based data synthesis methods reduce manual annotation, but they still face two key limitations: they often fail to cover the full functionality of a website, and without sufficient website prior knowledge, they tend to propose hallucinated tasks, which in turn limits the diversity and efficiency of downstream trajectory synthesis. We present \textbf{SynWeaver}, a website-prior task-trajectory co-synthesis framework designed to address these challenges. SynWeaver first performs structured website exploration and constructs a website map that covers a broad set of functionally distinct page states and executable interactions on the target website. It then derives page-level and transition-level supervision from this map to train a UI-aware model with website-specific priors, enabling more grounded task proposals. Finally, SynWeaver performs collaborative task-trajectory synthesis, jointly updating the task and execution trajectory when they become inconsistent, and then verifies and repairs the collected results to produce executable, semantically aligned supervision. Experiments on WebArena and WebVoyager demonstrate that SynWeaver consistently outperforms strong synthesis baselines and yields more effective supervision for both in-domain and out-of-domain generalization. --- ### MindMemOS: A Portable and Self-Evolving Memory Operating Layer for AI Agents Authors: Kaichao Liang, Yuqi Cui, Hao Kong, Xinyuan Huang, Guohaotian Hou, Qingcan Kang, Liang Chen, Yiyang Yin, Ke Ye, Jiaquan Guo, Da Chen, Lingan Zeng et al. Published: 2026-08-12 Categories: cs.AI, cs.IR, cs.IT arXiv: https://arxiv.org/abs/2608.12428v1 Memory is a core component of AI agents, enabling them to accumulate experience, maintain personalization, and adapt over long-term interactions. However, existing memory systems often remain fixed after development, limiting their ability to adapt their memory models, organization strategies, and procedural knowledge through continued use. We present MindMemOS, a portable and self-evolving memory operating layer that organizes open-world information using a unified entity property timestructure. MindMemOS supports scenario-adaptive memory modeling, higher-order pattern discovery, autonomous memory refinement, and continuous skill evolution. Its MindMemEvolve algorithm employs validation-driven evolutionary search to optimize memory schemas for target scenarios, whiledreaming consolidates accumulated memories by merging redundant records and resolving conflicts. In addition, implicit corrective feedback serves as a human-in-the-loop signal for identifying and revising potentially inaccurate or misaligned memories. Its MindSkillEvolve algorithm further transforms agent execution trajectories into reusable and progressively refined skills. MindMemOS achieves 94.03% accuracy on LOCOMO and 70.63% on PersonaMem. MindSkillEvolve improves SpreadsheetBench success by 9.2 percentage points over the initial-skill baseline. --- ### Large Language Models Can Follow Instructions, But Not Many at Once: Phase Transitions in Compositional Constraint Satisfaction Authors: Mariya I. Vasileva Published: 2026-08-12 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.12426v1 Large language models are increasingly deployed in settings that require simultaneous adherence to multiple explicit constraints - reasoning structure, safety boundaries, output schemas. Individual constraints are handled proficiently, but the compositional regime, where many must hold jointly, remains poorly characterized: how rapidly does performance degrade, what governs the degradation, and can the collapse be mitigated? We introduce Constraint Saturation Evaluation (CSE), a procedurally generated benchmark that systematically varies the number of simultaneous constraints (k), with every constraint scored by a deterministic, rule-based verifier and zero LLM-judge involvement: 15 models, 36 constraint types, 369,753 checks at k=1-12. Three findings emerge. First, per-constraint pass rate decays gradually and predictably, while the chance of satisfying all k constraints collapses - a model passing individual constraints at ~41% at k=8 succeeds on all eight just 5.7% of the time. Second, constraints do not degrade equally: structural constraints lose 2x more baseline capability per added constraint than lexical ones, ordered by a comprehension-maintenance gap that separates constraints requiring sustained tracking from binary decisions immune to composition. Third, failures are nearly independent, which is what makes the accumulation multiplicative; the residual coupling that does exist tracks shared output features rather than pairwise interference - a wrong sentence count fails every constraint that reads it. Reliable instruction following breaks down beyond 5-6 simultaneous constraints: probe-level success falls below 50% at 7 constraints for the strongest model, and at 3 or fewer for 12 of 15. --- ### AI-Driven Multiscenario Interest Rate Forecasting: A Proof of Concept for Banking Asset Management Authors: Ekkehardt Bauer, Dirk Holländer, David Scholz, Linus Wolff, Christoph Ostermair, Kyrillus Aiad, Joachim Hasebrook Published: 2026-08-12 Categories: q-fin.CP, cs.LG, q-fin.ST arXiv: https://arxiv.org/abs/2608.12424v2 This study focuses on developing an AI-supported prototype for multiperspective interest rate forecasting that combines classical econometric models with modern artificial intel-ligence methods. Tested in a major European bank, the system enables more precise and flexible prediction of interest rate developments, supporting strategic decision-making in Asset-Liability Management (ALM). It integrates topic modeling, sentiment analysis, econometric forecasting, and market-based analyses within an interactive platform. Leveraging AI to analyze large volumes of financial documents and market data enables the identification of monetary policy trends and sentiment signals at an early stage. The core econometric model is a Bayesian vector autoregression (BVAR) that enables simulation-based scenario analyses to evaluate economic developments from multiple perspectives. The system's innovation lies in its integration of several forecasting approaches that consolidate previously separate information sources and present them transparently and interpretably. Financial analysts and risk managers thus gain a better basis for making decisions, allowing them to assess interest rate risks more accurately and manage market movements more proactively. While the prototype demonstrates how AI can transform interest rate management in banking, further development is required to optimize real-time data integration and regulatory compliance. Even at this stage, the study shows that multi-perspective, AI-driven forecasting provides substantial added value for banks by increasing transparency, strengthening evidence-based decision-making, and improving risk management. --- ### Which Site, and When: A Free-Satellite-Data Test of Himalayan Glacial Lake Bursts, Landslides, and Ice Floods Authors: Matthew Kahn, Milan Arjel, Nirmala Adhikari, Mingmar Sherpa, James Pope Published: 2026-08-12 Categories: cs.LG, stat.AP arXiv: https://arxiv.org/abs/2608.12422v1 Two free satellite signals carry real information about glacial-lake outburst risk in the Nepal Himalaya: radar interferometry sees a moraine dam slowly sagging, and satellite weather marks the weeks when a primed lake is under stress. A companion feasibility study found that deformation indicates which lake is destabilizing and weather indicates when it is at risk, but proposed no predictive model. To address this gap, we propose and evaluate models that predict which site is susceptible and when a trigger arrives. We test three related hazards on free data alone: large moraine- and ice-dammed bursts, rainfall-triggered landslides, and smaller floods from ponds on and around a glacier. Each hazard gets two questions, never blended. Using 589 dated outbursts from HMAGLOFDB and several thousand catalogued landslides, we match each event against similar but unfailed sites, and hold every model to a strong simple baseline under spatial cross-validation that withholds whole map tiles, so no model succeeds by recognising a trained-on neighbourhood. Antecedent weather times the trigger at ROC 0.73 for big bursts, 0.83 for landslides, and 0.82 for small floods. Terrain ranks susceptibility only in part: scored naively it appears near 0.9, largely because catalogued failures cluster in wetter ranges; matched against comparable nearby sites the honest figures are 0.76, 0.71, and 0.54 (no better than chance). The burst signal holds within single regions, reaching 0.89 in Nepal alone. Five deep-learning models do not decisively beat a simple gradient-boosted baseline. Three score marginally higher on landslides, a hint too small to confirm. For the lake hazards the baseline wins outright, reproduced by a three-rule decision tree on ruggedness and monsoon rainfall. We close with a ranked Nepal watchlist, a prioritisation aid, not a prediction, and note where free data reaches its limits. --- ### LoKiFormer: Locality-aware Attention with Decoupled Knowledge Memory for Efficient Large Language Model Pretraining Authors: Qiuwu Chen, Zimo Liu, Yuchen Li, Ying Sun, Yifan Zhang, Zhijie Qiu, Zeng You, Ryan Dong, Simeng Ma, Yaofo Chen, Mingkui Tan Published: 2026-08-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.12419v1 Large language models (LLMs) have achieved remarkable breakthroughs across various applications. However, their architectures remain inefficient in pretraining due to two main limitations: (i) self-attention lacks an explicit inductive bias for locality, leading to redundant modeling of sequence-internal local information; (ii) mixture-of-experts (MoE) implicitly couples knowledge storage with computational pathways, hindering flexible access to sequence-external global knowledge. To overcome these limitations, we propose LoKiFormer, a novel LLM architecture that augments the standard decoder with two dedicated modules: 1) Local Fusion Attention (LFA), which incorporates a convolutional fusion to attention, explicitly capturing local patterns and allowing the attention to operate on more informative representations; 2) Knowledge Memory Module (KMM), which introduces a parametric key-value memory that explicitly stores global knowledge in addressable slots, decoupling storage from computation and enabling direct knowledge retrieval. Together, these modules enable LoKiFormer to achieve more efficient and effective integration of information at both levels. Experimental results show that LoKiFormer converges 1.33x faster in pre-training than baseline models, underscoring its superiority over existing LLM architectures. --- ### StateFlow: Building, Evolving, and Accessing 3D World States for Previsualization Authors: Yuyang Yin, Zixiang Li, Longxuan Deng, Hongkai Li, Shifang Zhao, Junnan Liu, Weirong Huang, Mengyu Wang, Tianxiao Fu, Yikai Wang, Peng-Shuai Wang, Xiaojie Jin et al. Published: 2026-08-12 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.12314v1 Previsualization is an intermediate layer between ideas and production in film, games, architecture, and urban design. It lets creators iteratively refine scenes, actions, cameras, and spatial-temporal dynamics. Yet existing generative methods rely on simple prompts to jointly control all of these factors through one-shot image or video synthesis, offering weak controllability and limited support for iterative editing. Fundamentally, a world comprises multiple elements with geometry, appearance, and other attributes, together with cameras. Different frames are produced through local modifications or recombinations of this shared state, which is otherwise largely reused. Therefore, we argue that the missing component is an explicit and persistent working state. To address this, we present StateFlow, a state-centric framework for generative previsualization. Rather than generating videos in one shot, StateFlow uses an editable 3D world to organize scene structure, evolution, and cameras, while off-the-shelf video models enhance visual quality when higher fidelity is desired. This world is maintained as a persistent structured 3D state of scene elements and camera configurations, serving as the core working representation for previsualization. Built on this insight, StateFlow has three stages to construct, evolve, and access the world state. State construction lifts generated 2D content into a coherent 3D world through prior-guided, conflict-aware dual-view initialization, while State evolution translates user intent into structured state transitions while preserving world memory, avoiding full-scene regeneration for each edit. State access uses render-feedback reflection to refine camera plans into visually feasible trajectories, avoiding reliance on VLM semantics alone. Experiments show that StateFlow produces high-quality 3D worlds for video creation and game-like prototyping. --- ### AVA-Encoder: Towards Agent-Native Video Representation Learning Authors: Chuyue Li, Jinpeng Yu, Haozhe Wang, Tian Xueyun, Zhijing Zhang, Bingnan Li, Shuqi Gu, Kan Ren, Jiaming Liu, Ruihua Huang Published: 2026-08-12 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2608.12313v2 Video creative agents still lack an effective way to learn from high-quality human films, limiting their ability to produce cinematic-grade videos. A key challenge is the absence of a structured video representation that is both faithful to film content and directly usable for agentic reasoning and manipulation. To address the challenge, we propose the Agentic Video Auto-Encoder (AVA-Encoder), a novel auto-encoding framework driven by agentic self-evolution to learn agent-native video representations. AVA-Encoder transforms a video into a Film Knowledge Graph (KG) representation and then reconstructs it back into video. This Film KG representation explicitly captures entities, events, assets, and their multimodal relationships in a structured form that can be easily understood, queried, and manipulated by agents. The reconstruction residual drives a dual-loop textual-gradient optimization framework that jointly improves the Film KG representation and the Agentic Video Encoder. Extensive experiments show that AVA-Encoder achieves a 20.7-percentage-point absolute gain, or a 73.1% relative improvement, over the strongest external baseline. In the controlled policy-only setting, its pseudo-trained Agentic Video Encoder policy also outperforms a carefully human-tuned policy while using 74.3% fewer shot-level and 70.1% fewer keyframe-level system-prompt tokens. We release the complete AVA-Encoder framework, a reliable agentic video reconstruction benchmark, and the first dataset of high-quality Film KG representations. --- ### AVA-Encoder: Towards Agent-Native Video Representation Learning Authors: Chuyue Li, Jinpeng Yu, Haozhe Wang, Tian Xueyun, Zhijing Zhang, Bingnan Li, Shuqi Gu, Kan Ren, Jiaming Liu, Ruihua Hua Published: 2026-08-12 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2608.12313v1 Creative agents still lack an effective way to learn from high-quality human films, limiting their ability to produce cinematic-grade videos. A key challenge is the absence of a structured video representation that is both faithful to film content and directly usable for agentic reasoning and manipulation. To address the challenge, we propose the Agentic Video Auto-Encoder (AVA-Encoder), a framework for learning agent-native video representations via agentic auto-encoding. AVA-Encoder transforms a video into a knowledge graph (KG) representation and then reconstructs it back into video. Its hierarchy and state nodes store structured text, while a linked asset layer holds generated images, audio, and video. Typed edges preserve the relations between these text descriptions and assets in a form that agents can easily understand, query, and edit. The video reconstruction differences drive a textual-gradient optimization framework, which expresses evaluation feedback as natural-language update directions for Data-Independent Encoding Policy Pseudo-Training in the outer loop and optional Data-Dependent KG Representation Refinement in the test-time inner loop. Extensive experiments show that AVA-Encoder improves by 20.7 percentage points over the strongest external baseline. In the controlled policy-only setting, its pseudo-trained shot-level Agentic Video Encoder policy also outperforms a carefully human-tuned policy while using 74.3% fewer system-prompt tokens. We release the complete AVA-Encoder framework, a reliable agentic video reconstruction benchmark, and the first dataset of high-quality film KG representations. --- ### DreamFly: Causal Memory and Receding-Horizon Diffusion Planning for Aerial Vision-Language Navigation Authors: Yan Deng, Fei Xu Published: 2026-08-12 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.12308v1 Aerial vision-language navigation (VLN) requires an embodied agent to integrate visual evidence over time, plan future actions, and determine when it has reached a navigation goal under partial observability. Although recent VLA models offer a promising perception-to-action paradigm, adapting them to aerial navigation remains challenging due to limited historical context, short planning horizons, and unreliable implicit termination. To address these challenges, we propose DreamFly, a diffusion-based aerial VLN framework built on Dream-VLA. DreamFly introduces a causally aligned historical memory that augments the current visual representation using only observations preceding the current decision step, enabling temporal reasoning without future information leakage. We further formulate navigation as receding-horizon diffusion planning, where the policy predicts a $K$-step action chunk but executes only the first action before replanning. This plan-$K$, execute-one strategy uses future actions as auxiliary planning targets while preserving closed-loop visual feedback. Finally, LiteStop estimates the stop probability directly from action logits at the initial all-mask state, decoupling explicit termination from action generation. Experiments on the OpenFly benchmark demonstrate consistent improvements in seen and unseen environments. DreamFly achieves 32.04%/29.46% SR and 28.22%/23.54% SPL on the test-seen/test-unseen splits, respectively, outperforming all compared methods on both metrics while attaining the lowest navigation error. These results demonstrate the effectiveness of jointly modeling historical context, future action structure, and explicit termination for aerial VLN. --- ### AI4AI at Test-Time: Strong-to-Weak Capability Transfer via Harnesses Authors: Cheng Qian, Wenting Zhao, Liangwei Yang, Heng Wang, Jielin Qiu, Heng Ji, Silvio Savarese, Huan Wang, Shelby Heinecke Published: 2026-08-12 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.12307v1 Recent work on distillation transfers the capabilities of large models to smaller ones often by updating the latter's parameters, through teacher forcing, on-policy distillation, and related training-time methods. In this paper, we ask whether such transfer can instead occur at test time. We study strong-to-weak scaffolding: whether a stronger builder model can construct inference-time harnesses that help a weaker target model solve tasks more reliably without any parameter updates. Using four representative Theory-of-Mind benchmarks, each builder model uses 5% of the data as a validation set to iteratively refine its harness over multiple rounds, after which the finalized harness is evaluated on the full test set. Empirically, this form of test-time capability transfer is highly effective, nearly doubling average target-model performance from 0.49 to 0.91. Our analysis shows that the gains come primarily from offloading unstable model reasoning into deterministic code, benchmark-specific routing, and strict answer-format enforcement, rather than from encouraging the target model to reason more extensively or sample more broadly. We further find that builder-model reasoning effort improves harness quality monotonically, platform effects are modest relative to the builder model's own capability, and weaker target models receive the largest gains. These results suggest that inference-time harness design is an important complement to conventional training-time distillation, enabling strong models to transfer cognitive structure to weaker models without retraining. --- ### Redistribution-based Cost Inference Improves Sparse Safe Offline RL Authors: Ebenezer Gelo, Geraud Nangue Tasse, Steven James, Benjamin Rosman Published: 2026-08-12 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.12306v1 Safe offline RL typically assumes access to dense per-step cost annotations, but in practice supervisors provide only trajectory-level stop-feedback: a binary signal at the first unsafe transition, with no per-step attribution. We frame this as a temporal credit assignment problem and propose the Redistribution-based Cost Inference (RCI) framework, which converts sparse stop-feedback into dense per-step costs via return decomposition, then trains a constrained offline policy on the augmented dataset. We show that return-equivalent redistribution preserves the feasible policy set and the optimal Lagrangian in a CMDP, establishing that the transformation is lossless in theory while yielding better-conditioned cost critic learning in practice. Experiments on highway driving and robotic manipulation demonstrate substantially lower violation rates than sparse and classifier-based baselines, with robustness to heterogeneous dataset compositions and label noise. --- ### Constructing Dynamic Master Logic Models as Knowledge Graphs for Complex System Diagnostics Using Retrieval-Augmented Large Language Models Authors: Saman Marandi, Yu-Shu Hu, Mohammad Modarres Published: 2026-08-12 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.12304v1 Dynamic Master Logic (DML) provides a hierarchical framework for representing system behavior by linking functional objectives to underlying structural elements. However, DML construction typically relies on expert interpretation of technical documentation, limiting scalability for complex systems. This study presents a framework for automated construction of DML models from system descriptions and their representation as Knowledge Graphs (KG-DML), using Retrieval-Augmented Generation and Large Language Models as enabling tools. Building on prior work with small-scale systems, the framework extends automated KG-DML construction and evaluation to substantially larger and more complex systems. Model construction proceeds across the DML hierarchy using targeted retrieval while preserving functional dependencies and explicit logical relationships. The resulting KG-DML supports diagnostic reasoning, safety assessment, upward failure propagation, and downward dependency tracing. A multi-level validation methodology evaluates layer-specific precision and recall, logical gate consistency, and overall structural integrity. Application to the Low-Pressure Coolant Injection system of a decommissioned Boiling Water Reactor demonstrates consistent reconstruction across repeated runs. The results show that automated KG-DML construction can transform technical documentation into executable functional models for diagnostic and reliability analysis. --- ### A Framework for Designing Reward Functions: From Objectives to Features to Human-Aligned Reward Functions Authors: Di Yang Shi, W. Bradley Knox Published: 2026-08-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.12302v1 We present a formal process to enable non-experts to instantiate and iterate on human-aligned reward functions, i.e. reward functions that adhere to a given preference ordering over trajectories. Given a task described in natural language, our process produces a linear reward function in three steps: distill the task's objectives into a set of fundamental objectives and derive measurable outcome variables that capture those fundamental objectives, select a causally representative subset of outcome variables as the reward terms, and fit weights to those reward terms via preference elicitation. Our contributions describe the first step and formalize the latter two steps. The first is a guided workflow for deriving outcome variables. The second is a reduction of reward term selection to minimum-cost partial cover on a causal DAG, solved in polynomial time via max-flow. The third is a geometric framing of weight fitting as a convex feasibility problem iteratively narrowed by preference queries, solved by existing separation oracle methods. To the best of our knowledge, this is the first reward-design method that maintains a deterministically conflict-free feasible weight region, narrowed to a desired tolerance via a separation oracle with O(n log κ) preference queries. --- ### Class Activation Mapping in Explainable Computer Vision: A Method-Centered Review of CNN, Transformer, and Foundation-Model-Era Visual Explanations Authors: AmirHossein Eshghi, Hamid Saadatfar, Seyyed Ali Hoseini, AmirMohsen Eshghi, Siavash Arjomand Bigdeli Published: 2026-08-12 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.12299v2 Class activation mapping (CAM) is one of the most widely used visual explanation families in explainable artificial intelligence. Its purpose is intuitive: it converts internal model evidence into a heatmap that highlights the image regions, convolutional channels, tokens, or patches that support a target class or concept. Since the first CAM formulation in 2016, the field has moved far beyond global-average-pooled CNN classifiers. CAM-style methods now include gradient-based post-hoc explanations, gradient-free score and ablation methods, high-resolution upscaling, weakly supervised localization and segmentation, transformer token attribution, causal and debiasing methods, and foundation-model-era approaches that use CLIP, DINO, SAM, or feature-distribution comparisons. This review synthesizes a strict corpus of 57 method-centered papers published from 2016 onward. The paper develops a taxonomy that separates methods by attribution mechanism, architectural dependence, and evaluation objective. It then reviews gradient-based CAMs, recent and hybrid CAM-style methods, and model-based or architecture-aware methods. Across the corpus, the main trend is clear: the field is shifting from explaining one class score in one low-resolution CNN layer toward comparative, multi-layer, probabilistic, token-aware, and foundation-model-aware explanations. At the same time, evaluation remains fragmented. Faithfulness, localization, robustness, computational cost, and human trust are often measured with different protocols. The review therefore emphasizes not only what each method contributes, but also which gap it leaves open and which later methods attempt to close that gap. --- ### Class Activation Mapping in Explainable Computer Vision: A Method-Centered Review of CNN, Transformer, and Foundation-Model-Era Visual Explanations Authors: AmirHossein Eshghi, Hamid Saadatfar, Seyyed Ali Hoseini, AmirMohsen Eshghi, Siavash Arjomand Bigdel Published: 2026-08-12 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.12299v1 Class activation mapping (CAM) is one of the most widely used visual explanation families in explainable artificial intelligence. Its purpose is intuitive: it converts internal model evidence into a heatmap that highlights the image regions, convolutional channels, tokens, or patches that support a target class or concept. Since the first CAM formulation in 2016, the field has moved far beyond global-average-pooled CNN classifiers. CAM-style methods now include gradient-based post-hoc explanations, gradient-free score and ablation methods, high-resolution upscaling, weakly supervised localization and segmentation, transformer token attribution, causal and debiasing methods, and foundation-model-era approaches that use CLIP, DINO, SAM, or feature-distribution comparisons. This review synthesizes a strict corpus of 57 method-centered papers published from 2016 onward. The paper develops a taxonomy that separates methods by attribution mechanism, architectural dependence, and evaluation objective. It then reviews gradient-based CAMs, recent and hybrid CAM-style methods, and model-based or architecture-aware methods. Across the corpus, the main trend is clear: the field is shifting from explaining one class score in one low-resolution CNN layer toward comparative, multi-layer, probabilistic, token-aware, and foundation-model-aware explanations. At the same time, evaluation remains fragmented. Faithfulness, localization, robustness, computational cost, and human trust are often measured with different protocols. The review therefore emphasizes not only what each method contributes, but also which gap it leaves open and which later methods attempt to close that gap. --- ### Beyond Trial-and-Error: Agentic Optimization for Image-to-Video Adherence Authors: Aman Tyagi, Hemanth Boinpally, Jonathan Chen, Douglas Gebert, Steven Hickson Published: 2026-08-12 Categories: cs.CV, cs.AI, cs.MM arXiv: https://arxiv.org/abs/2608.12290v1 Modern black-box Image-to-Video (I2V) models offer powerful capabilities in automated content creation, yet their lack of fine-grained control and reliability presents significant challenges in professional workflows. Their inherent stochasticity causes minor variations in textual prompts or hyperparameters to yield drastically different outputs often necessitating inefficient, brute-force trial-and-error processes. To address these limitations, we introduce the ``Agentic Self-Improvement" framework, which reframes video synthesis into a closed-loop, goal-directed optimization. Our framework systematically navigates the generation parameter space using a novel two-stage approach. In the first stage, an iterative prompt optimization loop uses a multimodal Large Language Model (mLLM) to refine the input prompt. This refinement implements two automated evaluations: Davidsonian Scene Graph (DSG) queries ensure semantic adherence, and Common Mistake Questions (CMQ) for artifact detection. At the second stage, we use Bayesian optimization to efficiently co-optimize stochastic seeds and CFG scales. This search is guided by a suite of quality metrics, including the novel Video-Text Adherence (VTA) score derived from the DSG and CMQ evaluations. Our framework significantly outperforms unguided search methods: in human preference studies, videos generated via our agentic approach were strongly preferred over baseline outputs, achieving win rates up to 69\%. This work provides a practical and extensible methodology for enhancing the predictability and control of state-of-the-art video generation models, moving the field beyond speculative curiosities toward reliable, production-ready tools. --- ### Large Language Model-Driven Small-Capitalization Trading: Integrating Financial News Sentiment, Macroeconomic Indicators, and Technical Signals Authors: Alireza Kargarzadeh, Nariman Khaledian, Navid Parvini, Arman Khaledian Published: 2026-08-12 Categories: q-fin.PM, cs.CL arXiv: https://arxiv.org/abs/2608.12283v1 Large language models can extract richer signals from financial news than fixed sentiment lexicons, and recent work has explored feeding such signals into portfolio construction. We study an uncertainty-aware construction that feeds model-predicted risk -- decomposed into aleatoric and epistemic components -- directly into the covariance matrix of portfolio allocators, rather than treating portfolio risk as fixed or adjusting only expected returns. We evaluate the pipeline on Russell 2000 equities under three stock-selection regimes: a pure-alpha trigger that isolates abnormal stock moves not explained by macro indicators, a pure-beta trigger that captures macro-indicator moves before the stock itself fires, and a beta trigger in which both channels agree. Across the full holding-period grid, the separated pure-alpha and pure-beta legs usually dominate the beta intersection on Sharpe and return. Two horizons are especially informative. At one day, pure beta can work under low and moderate transaction costs because it captures immediate lead-lag spillovers from liquid macro and sector indicators into exposed small-cap stocks, but this advantage disappears at 100 bps when turnover and microstructure noise dominate. At 40 days, pure beta works for a different reason: slower macro repricing overtakes the firm-specific pure-alpha channel. The strongest conservative row is pure beta with GPT-4o mini sentiment, a Student-t target, a 40-day holding period, and risk parity allocation, reaching Sharpe 2.33 at 100 bps. The results suggest that stock-selection regime and allocator choice matter at least as much as the sentiment model, and that separating firm-specific and macro-exposure triggers is more informative than requiring both to fire simultaneously. --- ### VAKRA: Evaluating Multi-Hop Reasoning Across APIs and Retrieval Under Tool-Use Policies Authors: Ankita Rajaram Naik, Anupama Murthi, Benjamin Elder, Siyu Huo, Raavi Gupta, Abhinav Jain, Praveen Venkateswaran, Abdulhamid Adebayo, Danish Contractor Published: 2026-08-12 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.12282v1 Agents deployed in enterprise settings must reason across structured APIs and document collections, yet existing benchmarks evaluate these capabilities in isolation. We introduce VAKRA (e\textbf{V}aluating \textbf{A}PI and \textbf{K}nowledge \textbf{R}etrieval \textbf{A}gents), a benchmark of over $8{,}000$ executable APIs across $62$ domains with tasks spanning three settings of increasing difficulty: diverse API interaction styles, multi-hop reasoning over structured APIs, and multi-source reasoning with natural-language tool-use policy constraints. Correctness is verified by re-executing predicted tool calls against live APIs, accommodating multiple valid paths. Using a fixed ReAct harness to isolate model capabilities from agent architecture, we evaluate frontier and open-weight models and find that even the best model achieves only 70.4\% on single-hop endpoint-style tasks and drops to 50--51\% on compositional APIs; performance degrades by over 50\% as reasoning depth increases, and policy-constrained questions expose severe failures (as low as 2.4\% on unanswerable queries). Trace analysis shows failures concentrate at language-mediated reasoning - entity disambiguation, cross-source grounding, rather than tool invocation mechanics. Code is available https://github.com/IBM/VAKRA. Dataset is available https://huggingface.co/datasets/ibm-research/VAKRA --- ### Curvature-Aware Zeroth-Order Optimization for Memory-Efficient Test-Time Adaptation Authors: Junming Zhang, Shuyu Yin, Peilin Liu, Rendong Ying, Fei Wen Published: 2026-08-12 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.12279v1 Test-time adaptation (TTA) aims to enhance the cross-domain performance of pre-trained models by adapting to unlabeled test data. While most existing TTA methods rely on backpropagation (BP) for finetuning, BP-free methods such as zeroth-order (ZO) methods are more desired in practical on-device scenarios. ZO methods rely only on forward computation, which can largely reduce the complexity and memory overhead of on-device deployment. However, ZO methods suffer from much higher variance compared with first-order methods in estimating the gradient. To address this, we propose an improved ZO method to substantially boost the performance of ZO optimization based TTA. First, we provide an observation to reveal the persistent low-rank Hessian structure of the loss during the adaptation process. Based on this insight, we then propose a loss-landscape curvature-aware zeroth-order (CAZO) method, which leverages a sliding-average estimation of the diagonal Hessian to construct a covariance matrix for anisotropic perturbation sampling. CAZO operates by freezing pretrained weights and optimizing minimal adapter parameters via forward-only passes based gradient estimation, which can substantially reduce the memory overhead compared to BP-based methods. Extensive experiments demonstrate that CAZO significantly outperforms existing TTA methods, achieving state-of-the-art performance while maintaining an excellent balance between accuracy and memory efficiency. Code is available at https://github.com/Hollyming/CAZO. ---