# Research Papers — 2026-05 · bundle page 14 of 326 > Full-text markdown of papers #651–700 of 16266 published in 2026-05, newest first. AI-readable mirrors of the full 16266-paper 2026-05 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-05/page/14 [← Previous bundle](https://aipod.dev/papers/archive/2026-05/page/13.md) [Next bundle →](https://aipod.dev/papers/archive/2026-05/page/15.md) ### Position: Reasoning is a Learnable Rule-Based Process Authors: Rachel Lawrence, Jacqueline Maasch Published: 2026-05-29 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2608.12325v1 Autonomous reasoning is among the most scientifically and economically motivating topics in AI today. Historically the purview of symbolic AI, recent advances have mainly emerged from deep probabilistic generative models. Despite immense interest and rapid progress, the generative AI community has not clearly converged on operational definitions for reasoning and often implicitly rejects the historical treatment of this topic in logic and verifiable automated reasoning. This position contends that definitional ambiguity leaves the construct validity of reasoning evaluation unverifiable, undermining quantifiable progress toward trustworthy autonomous reasoning. We also contend that this ambiguity is addressable. To that end, we provide (1) operational definitions based on a synthesis of the literature, positioning valid and sound reasoning as a learnable rule-based process; and (2) a checklist for best practices in the communication of AI reasoning research. --- ### When AI Is Your Pastor: A Benchmark for Theological Triage and Pastoral Guidance in Large Language Models Authors: Alex Chao Published: 2026-05-29 Categories: cs.CY, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.12324v1 People increasingly ask large language models (LLMs) for counsel on questions of faith, doctrine, and pastoral care. These questions are not ordinary information requests. Some ask about core Christian beliefs, some ask about real disagreements among faithful traditions, some require humility because the issue is prudential, and some are pastoral situations where safety and human referral matter more than theological completeness. Existing benchmarks do not evaluate this structure. We introduce FMG-Bench, the Faith & Moral Guidance Benchmark, a 120-scenario benchmark for evaluating large language model behavior in English-language Christian theological triage and pastoral guidance contexts. FMG-Bench v1 evaluates 14 advanced models across 8,792 scored responses, comparing raw model behavior with three guided instruction settings. In our production run, placing models inside a structured harness improves over raw model behavior by +3.96 points on average, with every model improving. The most safety-critical finding is a +10.8 point gain in escalation appropriateness -- whether AI systems recognize when pastoral, clinical, legal, or emergency support is needed. The guided settings also improve robustness, meaning consistency when questions are reworded or pressured (92.88 to 98.02 stability). Asking a model to compare perspectives helps in secondary-doctrine questions but can be counterproductive when applied to primary doctrine or urgent pastoral situations. The benchmark is a measurement tool, not an endorsement of AI systems as pastoral authorities. --- ### Why Do AI Agents Break Rules? How Framing, Context, and Social Signals Shape Compliance Authors: Mika Okamoto, Ansel Kaplan Erol, Kutluhan Erol Published: 2026-05-29 Categories: cs.CL, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2608.12323v2 Specifying a penalty can turn a legal obligation into a cost-benefit calculation that favors violation. We show that this enforcement information paradox occurs in AI agents. Most AI safety evaluations test whether models fail; we ask why, using compliance theory from law and economics as a diagnostic. We evaluate twelve instruction-tuned language models deployed as enterprise procurement chatbots. Each is given an environmental regulation in its system prompt covering large purchases, and a vendor list on which the certified suppliers cost nearly twice what the uncertified ones do. We test the agents against the predictions of deterrence, legitimacy, and expressive law, and find that each theory accounts for part of what we observe. Under identical conditions, compliance spans 46 percentage points across models, and models differ in which pressure breaks them: some treat the regulation as binding however it is worded, while others fail where theory predicts, under low penalties and non-command phrasing. Benchmark scores and developers' own descriptions of post-training do not predict where a model falls. Across all twelve, financial incentives, managerial demands, peer outcomes, and employee pressure each produce large compliance failures. These agents violate regulatory constraints to satisfy local user objectives in ways standard alignment benchmarks do not measure. Embedding the rule in the system prompt is not on its own enough to produce a compliant agent: model selection is itself a governance decision, and benchmark evaluation is not sufficient for compliance-sensitive deployments. --- ### What Drives LLM Self-Reflection? A Controlled Ablation of Uncertainty Routing in Armed Conflict Forecasting Authors: Poli Nemkova, Haeshitha Indukuri Published: 2026-05-29 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.12322v1 Self-reflection is widely assumed to improve LLM reasoning, yet which component drives the gain remains poorly understood. We present a controlled six-condition ablation isolating four components of LLM self-reflection: evidence exposure, diagnostic scaffolding, taxonomy vocabulary, and action routing. Two precise null results converge on a single mechanism. First, structured diagnostic questions add no measurable value over unstructured reflection ($\text{F1} = 0.296$ vs $0.297$, $p = 1.000$, 95\% CI $(-0.041, +0.040)$). Second, presenting the full uncertainty taxonomy while collapsing the action space to a single generic action also adds no value ($Δ\text{F1} = +0.008$, overlapping 95\% CIs), ruling out taxonomy vocabulary as the mechanism. Typed action routing provides consistent directional gains ($\text{F1} = 0.379$ vs $0.296$); the conservative estimate controlling for taxonomy vocabulary is $Δ\text{F1} = +0.075$, and the overall gain over the single-shot baseline is significant by bootstrap CI ($Δ\text{F1} = +0.101$, 95\% CI $(+0.020, +0.185)$). The vocabulary-routing decomposition replicates on GPT-4o: taxonomy vocabulary adds no significant value over generic reflection ($p = 0.773$), while action routing provides significant gains ($p = 0.025$), confirming the mechanism holds across backbones. Gains concentrate on structurally novel conflicts: in Myanmar ($\text{F1}: 0.000 \rightarrow 0.353$) and Ukraine ($0.167 \rightarrow 0.500$), the vocabulary-only condition recovers no more than generic reflection while action routing breaks the degenerate prior. These findings identify typed action routing -- not diagnostic scaffolding or taxonomy vocabulary -- as a promising design principle for metacognitive LLM forecasting agents, while motivating larger-scale evaluation across conflict typologies. --- ### LLMs Know the Constraint But Do Not Use It: Activation Bottlenecks in Pragmatic Constraint Reasoning Authors: Yubo Li, Ramayya Krishnan, Rema Padman Published: 2026-05-29 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.12321v1 When a salient surface cue competes with an implicit feasibility constraint, LLMs often fail -- but aggregate accuracy conflates genuine constraint inference with conservative defaulting. We formalize the distinction as conditional constraint activation: the constraint is internally encoded (Knowledge) symmetrically across constraint-present and -absent prompts (Symmetry), yet only sometimes routed into the decision (Routing) and repairable by a donor activation (Repair). A quartet diagnostic over 14 models reveals two failure modes; probes on two open weights decode the constraint above $88\%$, yet activation patching repairs one ($+6.4$ nats) and not the other ($-0.07$). On a mitigation frontier, no prompted intervention reaches the repair corner: all inflate conservative bias through a single mediation pathway -- prerequisite mention. Hidden-constraint failure is a routing problem, not a knowledge problem. --- ### Carefully Considering Culture: Analyzing LLM Alignment in Single- and Multi-Cultural Settings using Cultural Consensus Theory Authors: Krishna Pothugunta, John P. Lalor Published: 2026-05-29 Categories: cs.CL, cs.CY arXiv: https://arxiv.org/abs/2608.09937v1 Recent work in NLP has probed large language models for their understanding of cultural norms across countries. However, this work typically considers distributional patterns, ignoring group consensus or possible multicultural environments within a country. In this work, we leverage cultural consensus theory (CCT) from cultural anthropology to model such multidimensional nuance. Applying CCT to the World Values Survey (WVS) across 10 countries and 12 domains, we demonstrate that models frequently misrepresent cultural structures by either failing to form cohesive consensus or severely over-regularizing consensus. Through explicit representation of intra-group variance, CCT provides actionable diagnostics to evaluate when models reflect true human diversity versus algorithmic homogenization. --- ### Conflict or Strategy? Asymmetric Role Framing of La France insoumise and Rassemblement National in French News Headlines, 2022-2025 Authors: Amr Sobhy Published: 2026-05-29 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.09936v1 Do French news headlines frame left- and right-populist challengers as symmetric ``extremes,'' or as fundamentally different political adversaries? We examine 28,592 headlines about La France insoumise (LFI) and Rassemblement National (RN) published by 25 French-language outlets between 2022 and 2025, annotated through a three-model LLM pipeline validated against a stratified human audit. The clearest finding is role asymmetry rather than valence asymmetry: conflict framing and strategic-game framing are more robust across models and time than delegitimization, with AGGRESSOR serving as corroborating role syntax. LFI appears in headlines more often through a conflict register and RN through a strategic-electoral register. This role gap is direction-stable across all three annotation models, survives bootstrapping and permutation tests, and persists across outlet families and most of 2022-2025. A secondary moral-accounting layer (who is blamed, legitimized, or cast as a victim) is structured by outlet rather than party, producing aggregate nulls that conceal some of the corpus's most polarized patterns. Methodologically, the annotation pipeline reveals a two-tier reliability profile: conflict and strategic-game framing achieve the strongest human validation and cross-model stability; actor role is direction-stable but treated as corroborating because its audit reliability is lower; normative-judgment constructs (legitimacy, blame) are weaker. The paper contributes political-role assignment as a target for computational framing research that decomposes what valence-based measures conflate, and establishes a construct-stratified reliability framework for calibrating majority-vote LLM annotation pipelines in political text tasks. --- ### Cost-Effective Automated Judging of Natural-Language Mathematical Proofs Authors: Benjamin Grayzel Published: 2026-05-29 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.00004v1 Grading natural-language mathematical proofs is a recurring cost in evaluating math-reasoning systems, and frontier LLM judges are expensive. We ask whether cheap open-weight models can serve as reliable judges given a candidate proof, a ground-truth proof, and a human-grading rubric. On a 200-instance validation sample of IMO-GradingBench, three cheap judges (GPT-OSS 120B, DeepSeek-V4 Flash, Gemma-4 31B) agree with human pass/fail decisions at rates statistically indistinguishable from Claude Opus 4.7 and Gemini 3.1 Pro, at up to $100\times$ lower cost. We had expected a majority vote of the three to be the best budget option; it matched the frontier but did not improve on its strongest member. Extending to the full 1000-instance benchmark and exploring consensus rules, we found that requiring unanimous agreement (all-three-pass) reaches the highest pass-agreement and precision and, on four replicate runs, the smallest run-to-run spread. The headline finding is that cheap judges are competitive with the frontier at one to two orders of magnitude lower cost; as a deployable default we recommend all-three-pass, with the caveat that this rule was identified post-hoc and warrants independent replication. --- ### Looks Right, Works Right: A Project-Level Benchmark for Multi-Screen Mobile App Generation Authors: Fan Wu, Cuiyun Gao, Yiming Huang, Yang Xiao, Yujia Chen, Qing Liao Published: 2026-05-29 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2607.28645v2 Recent multimodal large language models can convert visual designs directly into executable code, but real mobile products require multiple screenshots to become a buildable codebase with shared components and working navigation. This project-level setting exposes three limits of existing design-to-code benchmarks: they focus on single-page generation rather than complete codebases, cannot evaluate cross-page navigation, and do not measure project-wide maintainability. We introduce MobileForge, the first benchmark for project-level multi-screen mobile app generation, comprising real mobile apps, human-reviewed screens, structured page-relationship annotations, and navigation test specifications. MobileForge supports five-axis evaluation of build, navigation, visual fidelity, code maintainability, and efficiency. We also propose state-isolated navigation testing to avoid cascading failures in navigation evaluation and an anchor-referenced list-wise visual evaluation protocol to improve visual-judge reliability. Across end-to-end runs on six frontier multimodal LLMs, current models can build mobile-app projects that compile and reach the correct pages, but interactive navigation remains unreliable and visual fidelity and maintainability still lag. The benchmark and supporting materials are available at https://github.com/anoa12159-hue/mobileforge_eval. --- ### Looks Right, Works Right: A Project-Level Benchmark for Multi-Screen Mobile App Generation Authors: Fan Wu, Cuiyun Gao, Yiming Huang, Yang Xiao, Yujia Chen, Qing Liao Published: 2026-05-29 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2607.28645v1 Recent multimodal large language models can convert visual designs directly into executable code, but real mobile products require multiple screenshots to become a buildable codebase with shared components and working navigation. This project-level setting exposes three limits of existing design-to-code benchmarks: they focus on single-page generation rather than complete codebases, cannot evaluate cross-page navigation, and do not measure project-wide maintainability. We introduce MobileForge, the first benchmark for project-level multi-screen mobile app generation, comprising real mobile apps, human-reviewed screens, structured page-relationship annotations, and navigation test specifications. MobileForge supports five-axis evaluation of build, navigation, visual fidelity, code maintainability, and efficiency. We also propose state-isolated navigation testing to avoid cascading failures in navigation evaluation and an anchor-referenced list-wise visual evaluation protocol to improve visual-judge reliability. Across end-to-end runs on six frontier multimodal LLMs, current models can build mobile-app projects that compile and reach the correct pages, but interactive navigation remains unreliable and visual fidelity and maintainability still lag. The benchmark and supporting materials are available at https://github.com/anoa12159-hue/mobileforge_eval. --- ### Beyond Memory: A Templated Substrate for Heterogeneous Collaborative Knowledge Work with LLM Agents Authors: Priscila Saboia Moreira, Christopher R. Sweet Published: 2026-05-29 Categories: cs.AI, cs.CY, cs.DL arXiv: https://arxiv.org/abs/2607.24759v1 Research projects, educational efforts, and adjacent knowledge work accumulate findings, decisions, and reasoning that future collaborators rarely recover. The parts most useful to that work, including dead ends and walked-back claims, are routinely excluded from publications and shared code; future researchers re-attempt the same failures because no record survives. LLM coding agents are common participants but hold no persistent memory across sessions, and retrieval-augmented generation over raw sources does not compound. The llm-wiki pattern (Karpathy, 2026; tonbi, 2026) addresses this by inserting an LLM-maintained, interlinked wiki between raw sources and the agent. We present llm-wiki-memory-template, a reusable, agent-aware instantiation, and argue it is a substrate for heterogeneous collaborative knowledge work along three axes (multi-human, multi-AI-agent, multi-domain) with each axis supported by a distinct architectural element of the template (§4). The wiki is append-only by convention, which preserves what did not work alongside what did, addressing a negative-result loss problem that publications and code-sharing structurally cannot solve. Three deployed case studies and one design report cover the axes individually: a solo research lineage that preserves abandoned iterations; a two-author project whose retroactive audit revised two prior experiments' claimed 20-of-20 coverage down to 14 and 12 evidence-based answers, then to 18 and 18 after a fix, with the failure path preserved across the artifact; an in-progress multi-agent deployment reported as a design; and a cross-domain educational variant. We name failure-path preservation, agent honesty, and appropriation as cross-cutting sociotechnical properties of the artifact, not only of its technical mechanisms. --- ### Loss-Aware Feature-Map Pruning in Convolutional Neural Networks Using Multi-Armed Bandits Authors: Salem Ameen, Sunil Vadera Published: 2026-05-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.22564v1 Convolutional neural networks often contain redundant feature maps that increase storage and inference cost. This paper presents a loss-aware feature-map pruning framework using multi-armed bandits. Feature-map pruning is structured because it removes complete convolutional output channels and their producing filters rather than isolated scalar weights. Each candidate feature map is treated as an arm. At each play time, one map is temporarily masked and evaluated on a sampled mini-batch; the map is then restored and the observed loss change is converted into a safe-removal reward. After a fixed play budget, candidate maps are ranked by learned scores and the top-k maps are permanently removed with their filters, biases and corresponding next-layer input-channel kernels. The study evaluates UCB1 and Thompson Sampling, compares them with direct/oracle-style evaluation on LeNet/MNIST, and extends the evaluation to MNIST, CIFAR-10, CIFAR-100, SVHN, CUB-200-2011 and Oxford Flowers 102. Results show that UCB1 and Thompson Sampling preserve accuracy close to unpruned models while removing feature maps and reducing convolutional computation. Friedman and Nemenyi tests show that UCB1 obtains the highest mean rank, followed by Thompson Sampling; both significantly outperform greedy and magnitude-based pruning while remaining statistically comparable to the original unpruned model. --- ### Synthetic Scenario Generation for Evaluation of Industry 4.0 Agents Authors: Sagar Chethan Kumar, Rohith Kanathur, Dhaval Patel, Kaoutar El Maghraoui Published: 2026-05-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.22563v1 Industrial agent benchmarks require realistic evaluation scenarios that integrate telemetry, failure modes, maintenance records, and domain standards. However, existing benchmarks such as AssetOpsBench rely on manually authored scenarios and cover a limited set of asset classes. We extend AssetOpsBench with a Smart Grid Transformer asset class and four IEC-grounded diagnostic tools for health-index prediction, dissolved-gas analysis, winding-temperature assessment, and load-profile assessment. We further introduce ScenarioGeneratorAgent, a pipeline for synthetic industrial-agent scenario generation. The pipeline constructs evidence-grounded asset profiles, allocates coverage-aware scenario budgets across operational domains, and generates candidates through a hybrid validation-and-repair loop that enforces schema validity, tool reachability, physical plausibility, standards alignment, and deduplication. To improve scalability, we apply two-level caching, parallel focus-group generation, thread-pool offloading, batched LLM calls, and early rejection filtering. On Smart Grid Transformer scenario generation, these optimizations reduce end-to-end runtime by $8\times$ for 50 scenarios while preserving quality, achieving a composite quality score of $74.2 \pm 1.9$ compared with $73.8 \pm 3.0$ for the unoptimized baseline. These results show that standards-grounded synthetic scenario generation can efficiently expand industrial-agent benchmarks without sacrificing scenario quality. --- ### SF-AMS: Strategic Forgetting for Structured Memory in LLM Agent Authors: Ning Yang, Siqi Li, Miaoxin Shen, Yuan Zhou, Meng Zhang, Tong Li, Haijun Zhang Published: 2026-05-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.22562v1 Managing long-context dependencies remains a primary bottleneck in LLM agents, as redundant and irrelevant information can degrade multi-step reasoning. Strategic Forgetting for Agent Memory Systems (SF-AMS) is proposed as a framework for maintaining compact high-utility memory by modeling the long-term importance of memory units. SF-AMS replaces static retrieval and heuristic decay with a utility-driven survival mechanism that updates memory importance from usage redundancy and temporal signals, inducing a hierarchical memory structure that prioritizes stable entity-consistent information while filtering noise. On top of this, Composite Importance Scoring integrates semantic and entity level signals to improve retrieval robustness. Experiments on LoCoMo and LongMemEval-s show consistent gains over strong state of the art baselines including LightMem MemO and A-Mem. The largest improvement appears in multi-hop reasoning under Qwen2.5-7B where SF-AMS achieves plus 9.65 F1 over the strongest baseline followed by temporal reasoning under GPT-4o-mini plus 6.91 F1 and open-domain tasks plus 6.53 F1 demonstrating strong cross backbone generalization. These results show that modeling memory importance as a dynamic utility signal is critical for reliable long-context reasoning. --- ### Codifying the Judge: Scalable Evaluation via Program Distillation Authors: Tzu-Heng Huang, Shengqi Qiu, Frederic Sala Published: 2026-05-29 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.22561v2 LLM-as-a-judge has become the standard for automated evaluation, but it suffers from high cost, inference latency, and opaque decisions---limitations that undermine its scalability and reliability. We address these with a simple, efficient alternative: program distillation. Instead of prompting an LLM at evaluation time, we distill its decision logic into a committee of programs that can score candidates directly. These programmatic judges offer transparency, are easily inspected or edited, and eliminate per-sample API costs. Building on this notion, we introduce PAJAMA, a system that synthesizes programs as judges, aggregates their decisions into a joint verdict, and incorporates a fallback mechanism to selectively escalate low-confidence cases to an LLM. Across five datasets and eight model families, we show that programmatic judges match the performance of a 13B-size LLM judge at 47x higher throughput. When using program outputs as routing signals, PAJAMA improves both accuracy and throughput and advances the Pareto frontier. Beyond evaluation, programmatic judges produce cheap and effective reward signals: on RewardBench, a reward model distilled from programs' verdicts outperforms one trained on a proprietary LLM's labels at two orders of magnitude lower API cost. --- ### Codifying the Judge: Scalable Evaluation via Program Distillation Authors: Tzu-Heng Huang, Shengqi Qiu, Frederic Sala Published: 2026-05-29 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.22561v1 LLM-as-a-judge has become the standard for automated evaluation, but it suffers from high cost, significant latency, and opaque decisions -- limitations that undermine its scalability and reliability. We address these with a simple, efficient alternative: program distillation. Instead of prompting an LLM at the evaluation time, we distill its decision logic into a committee of programs that score candidates directly. These programmatic judges offer transparency, are easily inspected or edited, and eliminate per-sample API costs. Building on this notion, we introduce PAJAMA, a system that synthesizes programs as judges, aggregates their decisions into a joint verdict, and incorporates a fallback mechanism to selectively escalate low-confidence cases to an LLM. Across five datasets and four model families, we show that programmatic judges can match the performance of a 13B-size LLM judge. When using program outputs as routing signals, PAJAMA improves both accuracy and throughput and advances the Pareto frontier. Beyond evaluation, programmatic judges produce cheap and effective reward signals: on RewardBench, a reward model distilled from programs' verdicts outperforms one trained on a proprietary LLM's labels at two orders of magnitude lower API cost. --- ### Do VLMs Read or Rewrite? On Transcription Faithfulness in Vision-Language Models Authors: Gwang Gook Lee, Kenan Emir Ak, Jay Mohta, Yan Xu, Dimitrios Dimitriadis Published: 2026-05-29 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2607.21617v2 Vision Language Models (VLMs) are increasingly used in place of traditional OCR pipelines for document understanding. In this paper, we show they do not always act as faithful transcribers: when text is imperfect, they often tend to rewrite it into a more plausible form - a behavior that clean-text OCR benchmarks cannot detect. We introduce FaithC4, a multilingual perturbation benchmark of 1,455 single-page documents (English, Chinese, Korean) with three perturbation families: scramble, random substitution, and visually similar substitution. We use the benchmark to evaluate 15 systems spanning general-purpose VLMs, OCR-specialized VLMs, and traditional OCR pipelines. These three categories differ in WER degradation under perturbation: general-purpose VLMs degrade by up to 6.9 points, OCR-specialized VLMs by 0.1-3.4 points, and traditional OCR by less than 0.8 points on English. Probing Qwen3-VL-4B layer-by-layer, we identify a consistent pattern: rewriting fires only when a perturbed word's final layer FFN representation stays close to the original encoding; when the representation diverges sufficiently, the model transcribes faithfully. Word length affects rewriting rate: short words (4-6 characters) are rewritten up to 10% of the time, with a sharp cutoff at 8 characters above which rewriting drops to 0%. --- ### Do VLMs Read or Rewrite? On Transcription Faithfulness in Vision-Language Models Authors: Gwang Gook Lee, Kenan Emir Ak, Jay Mohta, Yan Xu, Dimitrios Dimitriadis Published: 2026-05-29 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2607.21617v1 Vision Language Models (VLMs) are increasingly used in place of traditional OCR pipelines for document understanding. In this paper, we show they do not always act as faithful transcribers: when text is imperfect, they often tend to rewrite it into a more plausible form - a behavior that clean-text OCR benchmarks cannot detect. We introduce FaithC4, a multilingual perturbation benchmark of 1,455 single-page documents (English, Chinese, Korean) with three perturbation families: scramble, random substitution, and visually similar substitution. We use the benchmark to evaluate 15 systems spanning general-purpose VLMs, OCR-specialized VLMs, and traditional OCR pipelines. These three categories differ in WER degradation under perturbation: general-purpose VLMs degrade by up to 4.5 points, OCR-specialized VLMs by 0.2-2 points, and traditional OCR by less than 0.6 points on English. Probing Qwen3-VL-4B layer-by-layer, we identify a consistent pattern: rewriting fires only when a perturbed word's final layer FFN representation stays close to the original encoding; when the representation diverges sufficiently, the model transcribes faithfully. Word length affects rewriting rate: short words (4-6 characters) are rewritten up to 10% of the time, with a sharp cutoff at 8 characters above which rewriting drops to 0%. --- ### Lost in Context: Addressing Context Anxiety in Large Language Models Authors: Ifueko Igbinedion, Jillian Ross, Etienne Ricardez, Sertac Karaman, Eric So Published: 2026-05-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.21616v1 Conventional wisdom suggests that reasoning models fail when problems exceed their capabilities. However, we find that frontier reasoning models sometimes possess the necessary capabilities to solve problems but fail due to premature self-doubt -- a phenomenon informally known as context anxiety. We provide the first systematic study of context anxiety, demonstrating that it arises, in part, from a model's inability to accurately estimate the tokens required to complete a task. We also show that context anxiety leads to material efficiency losses when models operate under perceived constraints. Building on this analysis, we further show that models can learn alternative strategies for solving long-horizon problems without exhibiting context anxiety, suggesting that performance improvements may be achievable not through scaling model capabilities, but by improving models' ability to accurately assess and adapt to their own limitations. --- ### Beyond Liars' Bench: The Impact of Lie Typology, Depth, and Sparsity on Deception Detection in LLMs Authors: Amr Moustafa, Max Feser, Florian Mai Published: 2026-05-29 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2607.20479v1 Training probes to detect deceptive outputs from large language models is still an open problem. Recent work has demonstrated that detection probes fail especially in out-of-domain scenarios -- training on one type of lie does not transfer well to deception scenarios involving other types of lies. In this work, we conduct a systematic study on how various factors impact detection performance: representation depth, probe expressivity, sparse feature representations, and the lie typology of the training data. To this end, we augment standard benchmark training data with a supplementary dataset containing diverse types of deception, including fabrication, omission, and exaggeration examples. Analyzing these factors across seven probe types, our experimental results show that the optimal representation depth is highly dataset-dependent, more expressive probes provide only selective gains over linear baselines, and sparse autoencoder features perform similarly to dense hidden states. Ultimately, we demonstrate that the choice of training data and lie typology substantially changes detectability, highlighting that deception detection is a highly representation-dependent problem. --- ### Verifier-First Evaluation of Agentic LLMs for Infrastructure-as-Code Generation Authors: Mohamed Jouini Published: 2026-05-29 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2607.20478v1 Infrastructure-as-Code (IaC) generation from natural language requires satisfying provider schemas, dependency planning, and organizational policy constraints, not merely producing syntactically plausible configurations. We present a verifier-first empirical study of seven agentic strategies for Terraform generation evaluated on IaC-Eval v2, a modernized 186-task AWS/Terraform benchmark with Rego v1 intent policies. Our evaluation separates failures into three verifier stages (terraform validate, terraform plan, opa eval) and applies McNemar's test with Wilson confidence intervals on all pairwise comparisons (n=186, alpha=0.05). We report five principal findings. (1) Active retrieval via ReAct agents with MCP or ChromaDB-backed RAG raises Qwen2.5-Coder 7B from 14.0% to 45.7% pass@1 (p<0.0001), primarily by reducing VALIDATE_FAIL from 144 to 66 tasks. (2) Iterative refinement with verifier feedback achieves 62.9% (Qwen 7B) and 84.4% (GPT-4o) pass@1, exhibiting binary convergence -- tasks either resolve in one retry or exhaust the budget. (3) GEPA reflective instruction optimization raises the Active RAG baseline by +7.5 pp (p=0.026) using only 80 verifier-guided rollouts, providing evidence that prompt optimizers can improve verifiable IaC generation without weight updates. (4) SIMBA teacher-free demonstration injection achieves performance equivalent to Active RAG (p=1.0) without retrieval infrastructure, but fails to address the dominant SELF_DEFINED_PROPERTY error class (50% of failures). (5) A diagnostic Rego-injection experiment shows that 79% of post-refinement OPA failures are information-gap failures resolvable when policy text is visible (p=0.016), motivating policy --- ### LISA: Linear-Indexed Sparse Attention for Efficient Long-Context Reasoning Authors: Yu Zhao, Zekun Zhang, Fan Jiang, Bo Zeng, Linlong Xu, Shimin Shan, Yu Liu, Longyue Wang, Weihua Luo Published: 2026-05-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.19358v1 Recent advances in long chain-of-thought reasoning models such as DeepSeek-R1 have led to increasingly longer inference context lengths under the test-time scaling paradigm. However, the O(n^2) computational complexity of standard self-attention causes inference costs to grow sharply with long sequences, limiting the deployment of long-CoT reasoning in production settings. To address this, we propose LISA (Linear-Indexed Sparse Attention), a plug-and-play attention replacement module that requires no pretraining from scratch. LISA integrates two lightweight components in parallel within the original model: (1) a Linear Attention module that provides long-range memory with O(n) time complexity; (2) a Lightning Indexer that selects the top-M important tokens from the full context to feed into a Sparse Self-Attention. The two branches are fused via a gating mechanism, reducing inference complexity from O(n^2) to O(nM) (M << n) for generating n tokens. We design a two-stage training pipeline: Stage 1 initializes the model by integrating the linear attention to capture long-range dependencies, complemented by a sliding-window attention mechanism that is optimized via knowledge distillation to approximate the full self-attention distribution of a frozen teacher model. In Stage 2, we further introduce the Indexer to replace the static sliding-window mechanism, enabling dynamic token selection from broader contexts. The Indexer is trained using a novel per-head KL divergence loss, which aligns its selection behavior with the attention patterns of the teacher model. Experiments on DeepSeek-distilled-Qwen models demonstrate that LISA achieves a 50% inference speedup under 16K-token context, while improving average performance by 5.6% on reasoning benchmarks including AIME and MATH-500. --- ### Office Comprehension Benchmark Authors: Firoz Shaik, Mateus Picanço Lima Gomes, Tanvir Aumi, Jingci Wang, Milos Milunovic, Filip Basara, Ivana Jovanovic, Vishwas Suryanarayanan, Neha Nandan Kenkare, Weiyao Xie, Zhipeng Han, Zheng Zhang et al. Published: 2026-05-29 Categories: cs.CL, cs.AI, cs.CY, cs.IR, cs.LG arXiv: https://arxiv.org/abs/2607.01245v1 We introduce Office Comprehension Bench (OCB), the first public benchmark to jointly evaluate LLM systems on Word, Excel, and PowerPoint comprehension over native file formats (.docx, .xlsx, .pptx) and their variants. OCB consists of two tracks. File Fidelity Q&A tests structural and visual perception of office artifacts - tables, charts, embedded images, formulas, and app-specific elements such as headers, speaker notes, and named ranges. Domain Q&A tests expert-level reasoning grounded in real-world industry documents across 12 professional domains, with queries requiring multi-step analysis and synthesis across documents. Each reference answer is decomposed into atomic, binary-gradable claims, and an ensemble of LLM judges scores responses against each claim independently. Even the strongest frontier system in its default reasoning mode reaches only about 59.3% on Domain Q&A; increasing thinking depth within a tier does not move performance materially, while moving to a higher product tier yields modest gains. We release the dataset, evaluation tooling, judge prompt, and a public leaderboard. --- ### LLMs in the Real World: Evaluating "AI" in Emergency Contexts Authors: Sara Court, Lara Downing, Micha Elsner Published: 2026-05-29 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2607.00019v1 This paper offers a call to action. We urge our colleagues in the research community to play a greater role in the articulation of our findings to the public. To illustrate the stakes we present a case study on the initial stages of an LLM-based machine translation application's deployment in a real-world context: a text-2-911 system advertising capabilities in 55 languages for use in emergencies in which it may be difficult to call operators directly. We identify a number of common misconceptions about technologies such as these, concluding with a set of concrete recommendations and best practices for stakeholders at every stage of the development and deployment pipeline. While the advancement of scientific research often lies in solving the "hard" problems, we argue it is often the "easy" ones -- problems for which the latest technology is often unnecessary -- that are most overlooked. --- ### A Systems-Level Analysis of Sensitivity, Robustness, and Stability in Retrieval-Augmented Generation Authors: Bharath Simha Reddy Muthyam Published: 2026-05-29 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2606.28337v1 Retrieval-Augmented Generation (RAG) systems are often evaluated using final answer accuracy, even though their failures can originate from preprocessing, retrieval, context packing, or generation. This paper presents a controlled empirical study of RAG sensitivity, robustness, and stability across 56 experimental runs. We evaluate how chunk size, retrieval depth (top k), embedding-based reranking, probabilistic retrieval noise, and repeated seeded runs affect retrieval, context packing, and generation behavior. Using a fixed 500-question QA subset mapped to 20,958 unique corpus contexts, we analyze both final answer metrics and intermediate failure modes. Across these experiments, retrieval-oriented metrics improved under broader retrieval settings, while downstream exact-match and F1 scores often behaved non-monotonically. We also observe preprocessing-induced answer loss under smaller chunk sizes, progressive degradation under retrieval corruption, and higher observed variance in broader retrieval regimes. These findings suggest that RAG evaluation should include sensitivity, robustness, stability, and multi-stage failure analysis rather than relying only on final answer accuracy. --- ### HyBIRD: Hyperbolic Bridge Retrieval and Diagnosis for Methodology Inspiration Retrieval Authors: Yang Yang, Boyun Xu, Hao Fu, Jindong Li, Zining Zhong, Bowen Tian, Jiemin Wu, Menglin Yang, Yutao Yue Published: 2026-05-29 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2606.28336v1 Methodology Inspiration Retrieval (MIR) asks a system to retrieve prior papers whose methods can inspire a new research proposal. Unlike general scientific retrieval, the central challenge is not topical similarity but whether a candidate paper provides concrete mechanisms that can instantiate an abstract methodological need. Existing MIR dense retrievers provide strong paper-level rankings, but the returned lists do not expose how proposal needs are bridged by retrieved methods, where evidence is weak, or which complementary snippets may help. We propose HyBIRD, a frozen-anchor framework that treats MIR as hyperbolic bridge retrieval and post-hoc method diagnosis. HyBIRD keeps a strong MIR dense retriever fixed, learns lightweight point, cone, and factorized hyperbolic bridge variants, and uses LLM-assisted method blocks for post-hoc explanation and evidence selection. On the MIR benchmark, the factorized bridge reaches 59.034 mAP while preserving the dense anchor's strong retrieval behavior. More importantly, HyBIRD converts ranked papers into inspectable query need profiles, factor coverage, maturity views, and complementary evidence bundles. The results suggest that hyperbolic geometry is most useful as calibrated structure over a dense anchor, rather than as a standalone replacement for dense retrieval. --- ### AgentOdyssey: Open-Ended Long-Horizon Text Game Generation for Test-Time Continual Learning Agents Authors: Zheyuan Zhang, Zehao Wen, Alvin Zhang, Andrew Wang, Jianwen Xie, Daniel Khashabi, Tianmin Shu Published: 2026-05-29 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.24893v1 For agents to learn continuously from interaction with the world at test time, they must be able to explore effectively, acquire new world knowledge and skills, retain relevant episodic experiences, and plan over long horizons. To evaluate these key abilities of test-time continual learning agents, we introduce AgentOdyssey, a novel evaluation framework that procedurally generates open-ended text games with rich entities, world dynamics, and long-horizon tasks. Critically, AgentOdyssey goes beyond the conventional machine learning assumption that learning does not occur at test time by placing agents in a continuous, long-horizon setting that interleaves learning and inference throughout deployment. We further propose a multifaceted evaluation methodology that measures not only game progress but also offers diagnostic tests on world knowledge acquisition, episodic memory, object and action exploration, action diversity, and model cost. We evaluate diverse agent paradigms in the generated games. Our experimental results reveal critical limits in agents' key abilities, as well as factors that influence their meaningful horizon. Although performance scales with stronger base models, even the top agent remains far below human performance, leaving substantial headroom for improvement. Among agent mechanisms, we find that short-term memory benefits multiple agent paradigms and is an important component of agent test-time training. --- ### ReviewGuard: Aligning LLM-Assisted Peer Review with Long-Term Scientific Impact Authors: Abdur Rasool, Xiaohui Huang, Yanqing Hu, Linyi Yang Published: 2026-05-29 Categories: cs.DL, cs.AI arXiv: https://arxiv.org/abs/2606.24892v1 Peer review is central to scientific quality control, yet it can undervalue papers that later achieve substantial citation impact. While frontier large language models have shown promise in automating aspects of peer review, they primarily mimic human reviewer preferences rather than predict long-term scientific value. We introduce ReviewGuard, a two-stage framework that aligns LLM-generated reviews with citation-based estimates of long-term scientific impact rather than contemporaneous reviewer judgments. On 20,861 AI/ML papers from OpenReview augmented with Semantic Scholar citation data, ReviewGuard achieves a Spearman correlation of \r{ho} = 0.776 with future citations on rejected-then-published papers, outperforming human reviewers (\r{ho} = 0.492) and a supervised Expert model (\r{ho} = 0.681). Under the same decision threshold, ReviewGuard flags 10.2% of high-impact rejected papers, compared with 1.8% for human reviewers, corresponding to a 5.6x improvement. Our results demonstrate that impact-aligned reinforcement learning can provide editors with a complementary signal for identifying high-potential work, without replacing human judgment. --- ### Harnessing Agent Skills: Architectural Patterns and a Reference Architecture for Skill-Mediated LLM Agents Authors: Boming Xia, Liming Zhu, Zhenchang Xing, Qinghua Lu, Dino Sejdinovic, Xiwei Xu Published: 2026-05-29 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2606.20631v1 Agent skills externalise reusable agent-facing behavioural knowledge and guidance as persistent artefacts that can be discovered, activated, and interpreted by LLM agents. Although a skill artefact is static at rest, its architectural responsibilities arise in use, when the artefact is selected for a run, bound to context and authority constraints, interpreted by a stochastic agent, and recorded as run evidence. We call this run-specific relation skill-in-use. This paper studies agent skill harnessing: the architectural responsibilities that govern the transition from skill artefacts to skill-in-use, bound the executable consequences associated with skill-in-use, and capture evidence for attribution, verification, repair, and evolution. This paper provides a catalogue of ten empirically grounded architectural patterns (five core, five supporting) for skill harnessing and synthesises them into a reference architecture with four responsibility layers: Supply Chain, Mediation, Execution Control, and Evidence & Feedback. We evaluate the architecture through cross-instantiation across 8 selected systems. The resulting patterns and reference architecture provide a vocabulary and diagnostic frame for analysing skill-harnessing responsibilities across agent systems. --- ### Physical Atari: A Robust and Accessible Platform for Real-time Reinforcement Learning on Robots Authors: Khurram Javed, Joseph Modayil, Gloria Kennickell, Richard S. Sutton, John Carmack Published: 2026-05-29 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2606.19357v1 We built a robot called the Robotroller that actuates an Atari CX40+ controller and a device called the Atari Devbox that renders the game frame and the reward signal from the Arcade Learning Environment on a screen. The Robotroller and the Atari Devbox, together with an off-the-shelf camera and a desktop computer, constitute a system that can be used to study reinforcement learning algorithms in the physical world. We call the full system Physical Atari. In this paper, we detail the key decisions that make Physical Atari a robust and accessible platform. To make the system robust, we designed the Robotroller so that all movement is done through bearings, which reduces wear. Additionally, we wrote software that monitors the state of the servos at a high frequency and intervenes to limit stress. To make the system accessible, we used affordable off-the-shelf components and parts that can be manufactured using consumer 3D printers. Physical Atari can be built for under $1,000 and has been used for weeks of non-stop reinforcement learning experiments without any mechanical failures. We used it to validate that reinforcement learning algorithms can learn directly on robots and show that even small distribution shifts between learning and deployment can significantly degrade the performance of policies. Our results underscore the importance of on-device adaptation for strong performance on robots. --- ### EMORSION: Examining the Impact of Audio Parameters on Emotional Responses and Immersion in Film Authors: Nelly Garcia, Ruby Crocker, Bleiz M Del Sette, Fabrizio Smeraldi, Charalampos Saitis, George Fazekas, Joshua Reiss Published: 2026-05-29 Categories: cs.HC, cs.AI, cs.SD arXiv: https://arxiv.org/abs/2606.18266v1 EMORSION is an exploratory proof-of-concept study examining how film audio design shapes audience emotion and immersion in acinema setting. Four film scenes were selected across the horror (2) and drama (2) genres, balanced between mainstream and independent productions. For each scene, multiple alternative audio mixes were created by systematically manipulating three core aspects of audio design, frequency (pitch), dynamics (loudness), and directionality (spatial placement). Three audience groups viewed the scenes, with each group exposed to one manipulated mix alongside a control mix for each scene. Audience responses were assessed through a triangulated multimodal framework combining self-reported emotion and immersion via a questionnaire, physiological measures including heart rate monitoring, and video-based motion tracking. The protocol successfully captured measurable, interpretable differences across audio conditions, indicating that even subtle changes in audio design can shape emotional perception and immersion. Unconventional mixes tended to produce greater variability in audience interpretation, while conventional immersive mixes were associated with stronger cross-audience agreement. These findings establish the feasibility of the EMORSION protocol and motivate larger-scale studies to characterise the role of specific audio parameters in shaping audience experience. --- ### Physically Constrained Ensemble Gaussian Process Modelling for Expensive Quantum Systems with Heteroskedastic Noise Authors: Arpan Biswas, Sutirtha Paul, Joseph Agada, Matthias Thamm, Adrian Del Maestro Published: 2026-05-29 Categories: physics.comp-ph, cond-mat.str-el, cs.LG, quant-ph arXiv: https://arxiv.org/abs/2606.11240v2 Accurate modeling of quantum many-body systems often requires computationally expensive simulations such as Density Matrix Renormalization Group (DMRG) or Quantum Monte Carlo (QMC) calculations. These methods, while precise, impose significant time and resource constraints, limiting their use in exhaustive parameter exploration. Moreover, these expensive simulations can contain variable errors over the large unknown parameter space, which needs to be quantified and propagated. Thus, predictive modelling is required to estimate the functional space accurately over scarcely sampled data with heteroskedastic noise, while preserving the physical relevance of the estimation. Therefore, we present a Physically Constrained Ensemble Gaussian Process (pc-EGP) framework designed to efficiently model complex and noisy quantum systems under physical consistency constraints. The proposed method first enforces physical constraints as a user controlled weighted penalty to the data-driven loss function of the Gaussian Process (GP) surrogates. Then an ensemble of such GP models is trained with variable noisy simulations via numerical quadrature method where these multiple GP(s) at different nodes is integrated as a quadrature weighted average. We first demonstrate the framework on synthetically generated data before applying to quantum systems. In the first case study, we leverage DMRG simulations of the Bose-Hubbard Model to predict the critical interaction parameter Uc governing the superfluid-to-Mott-insulator transition. In the second case study, we demonstrate our method on QMC simulations, of a quantum liquid confined inside a nanoporous silicate with the goal of optimizing a chemical environment to realize a one-dimensional superfluid. Compared to conventional GP, pc-EGP achieves a better balance of accuracy and physically meaningful predictions. --- ### Artificial Intelligence in Ship Finance: Applications, Opportunities, and a Case Study in AI-Augmented Loan Origination Authors: Lasse Dierich, Orestis Schinas Published: 2026-05-29 Categories: q-fin.GN, cs.AI arXiv: https://arxiv.org/abs/2606.11238v2 Ship finance is a data-intensive and document-heavy segment of asset-based lending, requiring the integration of financial, technical, contractual, and regulatory information from heterogeneous and largely unstructured sources. Increasing environmental regulation and ESG reporting requirements are adding further complexity to underwriting and loan-origination processes. Recent advances in artificial intelligence (AI), particularly large language models (LLMs), create new opportunities for processing and analysing such information. This paper reviews potential applications of AI in ship finance, with a particular focus on LLM-based systems for document comprehension, information extraction, and workflow automation. We present ShipFinance.ai, a modular agentic architecture to support loan application workflows in ship finance. The proposed system combines an LLM-based extraction module, financial analysis components, external maritime data services, and a controlled document-generation module with a chatbot interface to support the preparation of standardized financing applications. The paper discusses the key challenges for using such models in production. We argue that AI-assisted systems can support maritime finance professionals in managing increasingly complex information and reporting requirements. --- ### A2SG:Adaptive and Asymmetric Surrogate Gradients for Training Deep Spiking Neural Networks Authors: Yechan Kang, Yongjin Kweon, Mingyeong Seo, Sohee Park, Yeonguk Jeon, Jongkil Park, Hyun Jae Jang, Jaewook Kim, YeonJoo Jeong, Suyoun Lee, Seongsik Park Published: 2026-05-29 Categories: cs.NE, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2606.11236v1 Training deep spiking neural networks (SNNs) remains challenging due to sharp loss landscapes and temporal inconsistency caused by surrogate gradients. To address these challenges, we propose a unified framework: adaptive and asymmetric surrogate gradients A2SG. The adaptive gradients adjust an effective window for spatio-temporal adaptation, reducing spatial gradient variation and maintaining directional consistency of gradients over time. The asymmetric gradients reflect neuronal dynamics by assigning larger gradients to neurons with higher membrane potentials, and we prove that they yield lower variation than symmetric surrogates. Our analysis further establishes a direct connection between local gradient variation and the curvature of the loss landscape, providing a principled explanation for how A2SG promotes convergence to flatter minima and improves generalization. We conduct extensive experiments on diverse models, including CNN-based and Transformer-based SNNs, across various tasks such as image classification using both static and neuromorphic datasets, as well as segmentation. The results demonstrate that A2SG consistently improves accuracy and energy efficiency, establishing it as a general and reliable solution for training deep SNNs. Our code is available at https://github.com/KIST-NCL/A2SG.git. --- ### Few-Shot Resampling for Scalable Statistically-Sound Data Mining Authors: Leonardo Pellegrina, Fabio Vandin Published: 2026-05-29 Categories: cs.LG, cs.DB, stat.ME arXiv: https://arxiv.org/abs/2606.11235v1 A key step in knowledge discovery is the evaluation of data mining results. In several applications, including pattern mining, graph analysis, and others, this step includes the evaluation of the statistical significance of the results, to avoid spurious discoveries due only to noise or random fluctuations in the data. While specialized procedures have been developed for some specific applications, resampling-based approaches are widely used, in particular for complex analyses where analytical results cannot be derived. However, current resampling-based approaches require the generation and analysis of thousands of resampled datasets, and are therefore impractical for large datasets or computationally intensive analyses. In this paper, we introduce FewRS, a simple and effective resampling-based approach to assess the statistical significance of data mining results with rigorous guarantees on the probability of false discoveries. Our approach can be used in every situation where resampling-based approaches are applied. FewRS builds on our derivation of a novel bound to the supremum deviation of test statistics representing the quality of data mining results. We prove that FewRS needs to generate and analyze an extremely small number of resampled datasets, leading to a highly scalable approach with wide applicability. We test our approach on common tasks such as pattern mining and network analysis. In all cases, our approach results in a reduction of up to two orders of magnitude in running time compared to the state of the art, while preserving high statistical power, enabling the statistical validation of data mining results on large-scale real-world datasets. --- ### OSCS-SupCon: Orthogonal Sigmoid-based Common and Style Supervised Contrastive Learning for Robust Feature Disentanglement Authors: Bin Wang, Fadi Dornaika Published: 2026-05-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.11233v1 Supervised Contrastive Learning (SupCon) has achieved strong performance by explicitly modeling pairwise relationships among samples. However, existing SupCon-based methods suffer from two key limitations: negative-sample dilution induced by the standard InfoNCE loss, and feature-space entanglement caused by the lack of explicit constraints separating category-relevant (common) and category-irrelevant (style) features. These limitations reduce feature discriminability and generalization ability. To address these issues, we propose OSCS-SupCon (Orthogonal Sigmoid-based Common and Style Supervised Contrastive Learning), a unified framework that combines a sigmoid-based pairwise contrastive objective with explicit orthogonality constraints. Specifically, we introduce a sigmoid-based contrastive loss with two learnable parameters, temperature and bias, which adaptively modulate pairwise decision boundaries and alleviate negative-sample dilution. Furthermore, we enforce orthogonality between common and style feature subspaces via a linear projection with ReLU nonlinearity, thereby reducing feature overlap and improving disentanglement of style-irrelevant representations. Extensive experiments on six benchmark datasets demonstrate that OSCS-SupCon consistently outperforms state-of-the-art supervised contrastive learning methods across multiple backbone architectures. In particular, on the fine-grained CUB200-2011 dataset with a ResNet-18 backbone, the proposed method achieves a 3.4% improvement in classification accuracy over CS-SupCon, highlighting its robustness and generalization capability. Ablation studies further confirm the effectiveness of each component. --- ### Every Act Has Its Price: Compressed Moral Composition in Frontier LLMs Authors: Weijia Zhang, Ruiqi Chen, Yunze Xiao, Weihao Xuan Published: 2026-05-29 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2606.11232v1 Existing LLM moral benchmarks usually ask which isolated moral act, value, or foundation a model prefers. This is useful but incomplete. Realistic judgments often require a model to combine several moral signals within the same option. We introduce **Moral Trolley Arena**, a two-stage blind ELO benchmark for measuring how LLMs compose moral evidence. The single-scene arena first calibrates individual moral acts from a 229-scenario corpus across five Moral Foundations Theory foundations; the composite arena then combines calibrated acts into two-act moral items over a controlled intensity grid and measures the resulting composite preferences. Across ten frontier models, composite judgments are largely predicted by component act strength, but the relation is consistently compressed rather than simply additive. Models also show non-additive intensity anchoring, bounded foundation-specific residuals after component control, and highly convergent composite preference surfaces across providers. These results suggest that moral audits should measure composition rules for moral evidence, not only rankings over isolated acts. --- ### Item Response Scaling Laws: A Measurement Theory Approach for Efficient and Generalizable Neural Scaling Estimation Authors: Sang Truong, Yuheng Tu, Rylan Schaeffer, Sanmi Koyejo Published: 2026-05-29 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2606.07616v1 Scaling laws provide a fundamental framework for understanding the performance of Language Models (LMs), yet deriving them requires prohibitively expensive evaluations across thousands of checkpoints or millions of inference samples. To address this, we introduce Item Response Scaling Laws (IRSL), a unified framework that integrates Item Response Theory (IRT) within the scaling law framework. Unlike traditional approaches that treat each model-benchmark pair in isolation, IRSL disentangles latent model ability from question characteristics, factorizing the scaling law estimation for $M$ models and $N$ questions to significantly reduce parameter complexity from $O(M \times N)$ to $O(M + N)$. We instantiate IRSL with Beta-IRT, which leverages the empirical probability responses of LMs -- such as token probabilities in pre-training and pass rates in test-time sampling -- to capture richer signals than binary responses. We validate our approach across two prevalent scaling paradigms: (1) pre-training downstream scaling, using 6,612 LM checkpoints and 37,682 questions from 10 benchmarks; and (2) test-time scaling, using 12 LMs and 120 questions from 4 benchmarks with up to 2,500 samples per question. Given a one-time calibration on existing model responses, IRSL yields more reliable scaling estimates using only 50 questions per benchmark (a 99.9\% reduction), achieving comparable or superior decision accuracy to traditional approaches. Furthermore, we show that the estimated latent model abilities are generalizable, enabling accurate performance forecasting across benchmarks that share the same measurement objective. --- ### Structured Neuron Pruning in Deep Neural Networks Using Multi-Armed Bandits Authors: Salem Ameen, Sunil Vadera Published: 2026-05-29 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07615v1 Deep neural networks often contain redundant hidden units. Removing individual weights can reduce parameter count, but unstructured sparsity is not always easy to exploit in standard dense implementations. This paper develops a structured pruning framework in which complete neurons are removed using multi-armed bandit (MAB) algorithms. Each candidate neuron is treated as an arm; pulling an arm temporarily masks that neuron, measures the change in loss on a sampled mini-batch, restores the neuron, and updates an estimate of its safe-removal reward. The framework supports stochastic policies, including Epsilon-Greedy, Softmax, UCB1 and Thompson Sampling, and multiplicative-weight policies, including Hedge-style multiplicative weights and EXP3. We evaluate the method on tabular classification, tabular regression and deep neural-network benchmarks covering image, text and reasoning tasks. Statistical comparisons using the Friedman test followed by the Nemenyi post-hoc test show significant differences between methods. On tabular classification tasks, UCB1 obtains the highest mean rank among pruning policies and improves on the unpruned neural network. On regression tasks, UCB1 obtains the highest mean rank and is statistically competitive with, or superior to, several standard regression models according to R^2. On deep-learning tasks, UCB1 and Thompson Sampling obtain the strongest ranks, and several MAB policies significantly outperform the unpruned model, magnitude-based neuron pruning and greedy activation-variation pruning. The results show that MAB-based neuron pruning is an effective and computationally practical approach for structured model reduction. --- ### Measuring Poverty and Inequality with Reduced Data: A Machine Learning Approach Using Nigerian Household Data Authors: Vanesa Jordá, Miguel Niño-Zarazúa Published: 2026-05-29 Categories: cs.LG, stat.AP arXiv: https://arxiv.org/abs/2606.07614v1 Reliable measurement of income and consumption is essential for monitoring poverty and inequality in low- and middle-income countries, yet full household surveys are costly and difficult to implement regularly. This paper examines whether reduced survey instruments can preserve key distributional information. We apply Random Forest Recursive Feature Elimination (RF-RFE) to the 2018/19 Nigeria General Household Survey-Panel to identify the income sources, consumption categories and household characteristics that best classify individuals within the welfare distribution. The analysis focuses on three outcomes: poverty status, location in the quintile distribution and position relative to the Gini-based inequality line. The survey's post-planting and post-harvest periods allow us to assess performance under different seasonal contexts. Results show that RF-RFE achieves strong classification accuracy with few predictors. For consumption, poverty status and inequality-line position are accurately predicted using a small set of expenditure categories, while quintile classification reaches about 80 percent accuracy for seasonal consumption and 60--65 percent for annual consumption predicted from a single seasonal visit. For income, poverty status reaches around 90 percent accuracy with five predictors, and inequality-line position is largely captured by labour earnings. The findings suggest that machine-learning methods can help improve survey design and reduce data requirements while retaining much of the distributional information needed to measure and monitor poverty and inequality. --- ### Can You Trust What You See? Human and AI Detection of Synthetic Legal Evidence Authors: Jinzhe Tan, Ali Ekber Cinar, Karim Benyekhlef Published: 2026-05-29 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2606.07613v1 Visual evidence has long been treated as a reliable form of legal proof, but advances in artificial intelligence (AI) are undermining that assumption. This article asks how well humans and frontier multimodal large language models (MLLMs) can distinguish authentic evidentiary photographs from AI-generated counterparts in the object-centric scenarios typical of civil disputes. We built Synthetic Legal Evidence Detection (SLED-1400), a dataset of 200 authentic evidence images paired with 1,200 synthetic counterparts produced by six contemporary text-to-image generators across ten evidence categories. The same stimuli and response format were used in a controlled web experiment with 136 lay participants and in a standardized evaluation of four MLLMs (GPT-5.1, Gemini-3-Pro, Gemini-3-Flash, Qwen3-VL-235B). Human accuracy was 64.8% overall, and 48.5% and 51.0% on the two strongest generators (Gemini-3-Pro-Image and Flux-2-Max), indistinguishable from chance. MLLMs never misclassified an authentic image (100% specificity), but missed most synthetic outputs from the harder generators, with average MLLM detection at 5.9% on Gemini-3-Pro-Image outputs. Human and MLLM errors were largely uncorrelated, while the four MLLMs were strongly correlated with each other. Neither group is a reliable standalone authenticator. We argue that visual evidence in legal proceedings should be treated as inherently contestable, and that a workable procedural response must combine trained human review, MLLM screening, and provenance infrastructure such as C2PA Content Credentials. --- ### Position: Anthropomorphic Misalignment Research Needs Stronger Evidence Authors: Vansh Gupta, Peter Nutter, Samuel Stante, Andreas Krause, Florian Tramèr, Lukas Fluri, Xin Chen, Anna Hedström Published: 2026-05-29 Categories: cs.CY, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2606.07612v1 We argue that many Anthropomorphic Misalignment Research (AMR) studies need stronger evidence to ensure that they can provide a robust foundation for critical safety decisions, such as model deployment and regulation. By evaluating failure modes across different misalignment concepts, such as deception, emergent misalignment, and sycophancy, we show how conceptual ambiguity, non-robust datasets, experimental design, and insufficient causal interventions can lead to overinterpretation of model behaviors. This position paper aims to offer guidance on evidentiary considerations that can help improve methodological rigor in AMR. To achieve this, we provide a clear call to action through a proposed framework of evidence levels and a diagnostic checklist. These shared standards will enable more productive scientific discourse and ensure that claims about AI risks rest on solid empirical foundations. --- ### MIRAGE: Metadata-Integrated Repository Analysis and Guided Enhancement for MSR Datasets Authors: Aabia Ather, Muhammad Usayd Ather, Qurat-Ul-Ain Somroo, Muhammad Khuram Shahzad Published: 2026-05-29 Categories: cs.IR, cs.AI, cs.LG, cs.SE arXiv: https://arxiv.org/abs/2606.07611v1 This paper proposes an improved approach to the analysis of Mining Software Repositories (MSR) datasets via metadata enrichment, FAIRness assessment, and topic-driven analysis. This research expands upon an earlier dataset directory created specifically for the analysis of MSR datasets by adding new annotations to the datasets, enriching the metadata categories, and offering more advanced filtering options. The metadata of the MSR papers presented from 2013 to 2024 has been gathered using the Semantic Scholar API. The analysis is based on Latent Dirichlet Allocation (LDA) topic modeling and statistical analysis. Dataset-level attributes were included into the expanded dataset directory, namely repository hosting site, format, accessibility, reusability, and dataset quality. The study reveals that the choice of repository hosting sites and data formats influences citation patterns and dataset usability. Furthermore, the enhanced annotation approach improves the analysis and discoverability of MSR datasets, supporting more effective reuse and evaluation of research artifacts. --- ### LEAF: Growing Trees Without Branching for Speech-Aware Large Language Model Post-Training Authors: Argyrios Gerogiannis, Yekaterina Yegorova, Mark Hasegawa-Johnson, Venugopal V. Veeravalli Published: 2026-05-29 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2606.07610v1 State-of-the-art GRPO-style methods for speech-aware large language model post-training suffer from coarse credit assignment, broadcasting the same terminal-reward advantage to every token in a response. This ignores useful structure within rollout batches, where speech-conditioned completions often share prefixes before diverging at important decisions. We propose Low-rank Exploration with Adaptive Forking (LEAF), a retrospective tree-based RL method that recovers this structure without online branching or additional decoding. LEAF samples complete responses, selects high-surprisal boundaries, groups responses by shared prefixes, and assigns span-level advantages using descendant rewards. We theoretically justify LEAF's span-level credit assignment and boundary-selection design. Empirically, LEAF improves over GRPO across speech question answering and speech translation benchmarks under the same rollout and low-rank adaptation budget. Notably, smaller LEAF-trained models outperform current state-of-the-art, full-parameter baselines. --- ### Subtitle-Aligned Fine-Tuning of Whisper for Swiss German ASR: Benchmark Contamination, Convention Mismatch, and an Honest Baseline at 25.6% WER (13.8% cWER) Authors: Felix Akeret Published: 2026-05-29 Categories: cs.CL, cs.AI, cs.LG, cs.SD arXiv: https://arxiv.org/abs/2606.07608v1 We present a systematic study of fine-tuning OpenAI's Whisper large-v3 for Swiss German ASR, using 1,367 hours of broadcast speech paired with Standard German subtitles as weak supervision. Through 16 iterative training runs on an NVIDIA DGX Spark (Grace Blackwell, 128 GB unified memory, up to 1 PFLOP FP4), we compare LoRA and full fine-tuning of the 1.55B-parameter model, investigate hallucination root causes, and quantify the effect of data quality, subtitle alignment, and training strategy. Our best model achieves 25.6% measured WER on the All Swiss German Dialects Test Set (ASGDTS) in an honest evaluation on strictly disjoint data. A harmonized error analysis separating genuine errors from valid stylistic variation (tense, word order, Swiss orthography) yields a content WER (cWER) of 13.8%, counting only actual recognition failures. Bias-corrected estimation reduces this to 8.5%, suggesting the true error rate is roughly one third of measured WER. We demonstrate that published state-of-the-art Swiss German ASR results (17.1-17.5% WER) are inflated by benchmark contamination: a vanilla Whisper model self-trained on the ASGDTS test set with zero Swiss German data achieves 13.88% WER, surpassing all published systems. Experiments with Phi-4-multimodal show an even stronger memorization effect (3.9% WER), revealing that the benchmark primarily measures convention matching rather than dialectal comprehension. We release two models, a LoRA adapter (25.32% WER, 13.9% cWER) and a full fine-tuned model (25.60% WER, 13.8% cWER), among the few publicly available, honestly evaluated Whisper models for Swiss German, under Apache 2.0 with full reproducibility, requiring no institutional data agreements. --- ### Position: Genomic Model Research Must Move Beyond Anecdotal Evaluation of Interpretability Methods Authors: Shasha Zhou, Mingyu Huang, Ke Li Published: 2026-05-29 Categories: cs.LG, q-bio.GN arXiv: https://arxiv.org/abs/2606.07607v1 Advances in machine learning and computational power have unlocked the predictive potential of the human genome, yet biologists now demand that these models also elucidate the underlying biological mechanisms. While interpretable machine learning (IML) techniques have been increasingly applied to bridge this gap, there has been a pervasive reliance on anecdotal validation: the vast majority of research relies on a single IML method and reports only isolated successful instances. Through a benchmarking study on transcription factor binding, we demonstrate the risks of current practices. We show that different IML methods can often (1) yield contradictory explanations for the same predictions, (2) fail to localize known regulatory motifs, and (3) fail to faithfully reflect the model's internal decision process. In light of this, we argue for a validation framework analogous to clinical trials: just as trials require rigorous design and adverse-event reporting, genomic interpretability must move beyond cherry-picked plausibility toward systematic assessment of consistency, faithfulness, and biological validity. To facilitate this, we propose a tiered framework to guide rigorous evaluation and reporting of genomic IML methods. --- ### QDSP: An Interpretable Structured Learning Framework for Predicting Death or Cerebral Palsy in Very Low Birth Weight Infants Authors: Ling Wang, Xiaolong Li, Hui Zhou, Jing Shi, Fuhao Zhang, Dapeng Chen, Nan Mu Published: 2026-05-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2606.07606v1 Very low birth weight infants (VLBWI) are at high risk of mortality and severe neurodevelopmental impairment, including cerebral palsy, yet reliable discharge-time prognostic stratification remains challenging in high-dimensional and data-limited clinical settings. To address this problem, we propose QDSP, an interpretable structured learning framework that integrates Quota-guided Subspace Sampling (QSS) and Differentiable-decision-guided Structure Perception (DSP). The QSS module constructs stability-aware and low-redundancy feature subspaces through bootstrap-based feature consistency estimation, whereas the DSP module employs differentiable soft oblique decision structures to model nonlinear clinical interactions while preserving traceable decision evidence. The proposed framework was evaluated on a real-world VLBWI cohort comprising 51 infants and further validated on three public medical tabular datasets. On the primary cohort, QDSP achieved an accuracy of 0.9200 and an AUC of 0.9714, outperforming representative machine learning and deep tabular learning baselines, including XGBoost, TabNet, and TabPFN. Across external datasets, QDSP maintained competitive discrimination and calibration under varying sample sizes and clinical distributions. In addition, SHAP-based analyses and differentiable decision-path tracing identified clinically relevant predictors, including cystic periventricular leukomalacia (cPVL) and birth weight, consistent with established neonatal pathophysiological evidence. These results suggest that QDSP provides an interpretable and robust framework for discharge-time risk stratification in VLBWI and may support early individualized clinical decision-making in neonatal intensive care settings. --- ### SRT: Super-Resolution for Time Series via Disentangled Rectified Flow Authors: Jufang Duan, Shenglong Xiao, Yuren Zhang Published: 2026-05-29 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07605v2 Fine-grained time series data with high temporal resolution is critical for accurate analytics across a wide range of applications. However, the acquisition of such data is often limited by cost and feasibility. This problem can be tackled by reconstructing high-resolution signals from low-resolution inputs based on specific priors, known as super-resolution. While extensively studied in computer vision, directly transferring image super-resolution techniques to time series is not trivial. To address this challenge at a fundamental level, we propose Super-Resolution for Time series (SRT), a novel framework that reconstructs temporal patterns lost in low-resolution inputs via disentangled rectified flow. SRT decomposes the input into trend and seasonal components, aligns them to the target resolution using an implicit neural representation, and leverages a novel cross-resolution attention mechanism to guide the generation of high-resolution details. We further introduce SRT-large, a scaled-up version with extensive pre-training, which enables strong zero-shot super-resolution capability. Extensive experiments on nine public datasets demonstrate that SRT and SRT-large consistently outperform existing methods across multiple scale factors, showing both robust performance and the effectiveness of each component in our architecture. --- ### Contribution Weights: A Geometrical Analysis of Self-Attention Transformers Authors: Harry Jake Cunningham, Nicola Muca Cirone Published: 2026-05-29 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07604v1 Analyzing attention weights has become a standard approach for interpreting the information flow of Large Language Models (LLMs). However, this approach has significant limitations as it neglects the geometric properties of the value vectors being aggregated. To address this gap, we introduce \emph{Contribution Weights}, a projection-based metric that quantifies a token's influence by accounting for it's attention weight, value magnitude, and directional alignment with the layer output. We demonstrate that contribution weights provide a more faithful measure of token importance, consistently outperforming attention-based metrics in identifying semantically critical tokens across different decoder-only models, tasks, and datasets. Further, our metric enables novel mechanistic analysis of \emph{attention sinks}. While previous work characterized sinks as passive repositories for excess attention, we reveal they serve an active functional role, suppressing information through a convex relationship between sink rate and output norm, stabilizing representations by opposing the semantic drift of low-confidence tokens. --- ### MetaEvo: A Meta-Optimization Framework for Experience-Driven Agent Evolution Authors: Bowen Ren, Heyan Huang, Yinghao Li, Yang Gao Published: 2026-05-29 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.07603v1 Large language models (LLMs) exhibit strong reasoning capabilities, yet most LLM-based agents are statically deployed and unable to improve through task interactions. Existing experience-driven methods often rely on memory or heuristics without enhancing the model's ability to learn, treating it as a passive executor and leading to early performance plateaus and limited long-term improvement. To address this issue, we propose MetaEvo, a two-stage framework for continual agent evolution that focuses on improving how the model learns from tasks experience, rather than solely on what it stores. MetaEvo first applies preference-based optimization to enhance the model's ability of principle abstraction, then enables the accumulation and reuse of these principles within a modular agent architecture. Experimental results on diverse reasoning benchmarks demonstrate that MetaEvo consistently outperforms strong baselines, maintains reliable improvement across iterations. These findings validate the effectiveness of meta-optimization in enabling agents to learn from experience and continually enhance their reasoning capabilities. ---