Reasoning & Chain-of-Thought - Weekly Roundup

The value of curated, capability-balanced supervision for adapting general language models to educational tasks spanning problem solving, curriculum understanding, and instructional support is demonstrated.

This method introduces Instance-level Difficulty-Aware Control (IDAC), a reward-shaping mechanism that leverages pre-computed reference statistics (accuracy and token usage) to regulate reasoning depth and enables stable critic-free optimization without learned reward models or online reference-model queries.

The value of curated, capability-balanced supervision for adapting general language models to educational tasks spanning problem solving, curriculum understanding, and instructional support is demonstrated.

This method introduces Instance-level Difficulty-Aware Control (IDAC), a reward-shaping mechanism that leverages pre-computed reference statistics (accuracy and token usage) to regulate reasoning depth and enables stable critic-free optimization without learned reward models or online reference-model queries.

ALIBI is presented, a semantic cover story attack against frontier LLM-based malware analyzers that adds a small, non-executed read-only section to a compiled binary, containing a coherent but false security product narrative, without altering imports or executable behavior.

Qualitative analysis shows that S4L captures implicit causal relations, deontic modality, and exception structure more reliably than the baselines, demonstrating that structured reasoning prompts can substantially improve the reliability of natural-language-to-logic translation for legal and safety-critical applications.

The refined proposal enables a sparse encoding that magnifies evidence and excludes background tokens, and is called Vision-RL2, which improves accuracy over the base model at every token budget and surpasses its largest-budget accuracy with about 4 times fewer visual tokens.

An architecture extending unanticipated-fault recovery from detection to mitigation and an ensemble methodology for evaluating LLM-assisted mission management on low-power AUVs are made.

Initial evidence is provided that greCAPTCHA can assess manuscript-specific understanding under proctored conditions and the appropriate construct validity for author understanding is remarked on, while also suggesting important changes to be made before deployment.

AMPLE-Math, a reusable suite of 5,319 mathematical problems with six reasoning views that share the same answer, is constructed and AMPLE-Math, a reusable suite of 5,319 mathematical problems with six reasoning views that share the same answer, is compared, suggesting that OPSD can improve access to existing reasoning capabilities through parameters shared by direct-response and thinking-enabled inference.

This work takes an RSI-inspired approach at the harness layer, scaling auto-research loops across increasingly numerous and diverse environments for harness rollouts, and yields reusable improvements that transfer beyond their development setting.

Unified Latent Diffusion Reasoner (Unified Latent Diffusion Reasoner), a framework that brings teacher reasoning steps from different modalities into shared thought tokens, trained to preserve the information needed for later reasoning steps and the final answer or action.

Simulated consensus did not track collective accuracy, and belief-anchored agent groups were biased estimators of the human group-outcome distribution in this setting, providing a scoring-explicit basis for assessing simulated-group estimates of human deliberative outcomes.

Inspired by recent findings that self-attention acts as a low-pass filter and that its low-frequency components are informative, AttnPrint for white-box provenance is developed, which achieves strong derivative-model detection performance while remaining robust to five downstream modification techniques.

This tutorial aims to walk the reader through the journey from words to network traffic, shedding light on how language is translated into communication flows within HPC training systems, and tries to demystify how words travel across the network.

It is shown that the workspace token can be used as a drop-in replacement for observations during deployment, enabling policies to solve memory-intensive tasks without the need for VLM reasoning in-the-loop.

This paper introduces a hybrid qualitative model designed for Camelot Jr, a block-puzzle game that requires constructing multi-level bridges to connect two avatars stationed on separate towers, and integrates a mathematical center-of-mass stability logic to guide the qualitative solver.

This article investigates neuro-symbolic agentic AI (NSAAI) as a framework for combining neural grounding, symbolic reasoning, and closed-loop agentic interaction to support more reliable and adaptive UAV autonomy.

Findings show that semantic reference-text overlap should not be treated as a proxy for temporal localization, and that AVTrace can identify task-specific weaknesses while providing a testbed for temporal post-training.

The Profiling, Investigation, and Judgment (PIJ), comprising 2,500 real homicide cases from five countries, is introduced and it is found that performance degrades systematically as tasks shift from explicit fact extraction to implicit reasoning over unknown suspect profiles.

CliniCIRCA is the first to temporally classify clinical events across unstructured discharge summaries without event-level timestamps and instruction tuning generally improves five open-weight models on event extraction, temporal tagging, and summarization across silver and clinician-verified evaluations.

This empirical study is an independent reproduction of the dissociation Zhao reported in 2026, where the shape of a large language model's chain-of-thought entropy trajectory predicts whether the final answer is correct, while the magnitude of its total entropy drop does not.

EviRCA is presented, a framework for LLM-based RCA that decouples deterministic evidence extraction from LLM reasoning and substantially outperforms prior OpenRCA baselines that achieve up to 15.2%, while reducing token consumption by 15-26x and execution time by 3-20x.

The paper presents DeltaSelect, an open-source method that identifies tasks whose one-run results consistently track full-benchmark performance using Pearson correlation, maps fractional verifier results to a common score using linear regression, and selects a fixed task set within a dollar budget.

LAMP is proposed, the first foreground-guided framework for polyp image synthesis based on lesion-guided adaptive mucosal context propagation and substantially outperforms existing methods in overall generation quality and consistently improves five downstream segmentation models.

Through a series of worked examples, it is shown how formally representing a theory of mind as a memo-like programming language can lend insight into a variety of theoretical issues in theory-of-mind research.

Trivet is presented, a framework combining large language models (LLMs) and Lean for automated translation validation of LLVM transformations, which generates structured proof scaffolds based on source and target functions, automatically discharges obligations amenable to deterministic reasoning, and delegates transformationspecific obligations to LLMs.

The reformulate city-scale 3D grounding as structured constraint reasoning, where description semantics are organized into computable cross-modal constraints over open-vocabulary 3D entities, attributes, and spatial relations, and introduces CitySTAR-3D, an enhanced benchmark that improves semantic coverage, instance completeness, bounding-box fidelity, and spatial-relation complexity in city-scale 3D grounding.

CoRef-GS is proposed, a cooperative referring Gaussian splatting framework that constructs local open-vocabulary instance-aware Gaussian maps, then aligns partially overlapping maps with a cross-agent alignment module by geometric and semantic consistency, and grounds queries using a view-conditioned mask relation graph.

It is revealed that 2D image and 3D shape encoders exhibit largely disjoint failure patterns and rarely share identical wrong labels, whereas two 2D image encoders frequently repeat the same errors, which makes the 2D and 3D pair inherently complementary.

Experimental results show that SOLAR consistently outperforms state-of-the-art vision-only, vision-language, and task-specific models across all tasks, demonstrating superior accuracy, robustness, and multimodal reasoning.

VideoResearcher achieves state-of-the-art performance among self-improving agents and approaches the human-designed upper bound, demonstrating a training-free paradigm for long-video understanding that expands agent capabilities through autonomous tool development while reducing costly manual engineering.

The effectiveness of the YOLOv12-m-seg model for real-time cotton boll detection, segmentation, and selective picking under varying confidence levels is validated.

This work proposes StageGuard, an agentic distillation framework for accurate and efficient stage-transition decisions, which combines teacher-model reasoning with demonstration trajectories to generate structured explanations of subtask completion and policy switching.

HEROIC is a decentralized heterogeneous multi-agent open-vocabulary search coordination framework that requires agents to communicate in natural language only and reaches the target 84% of the time across all 6 scenes, compares to 35-54% for vision-language frontier baselines, frontier-based search, lawnmower, and random-walk running the same perception.

This work tests a model's ability to rank conformers by stability, suggesting that it emerged unintendedly from models training on linked but conceptually different tasks, and suggests that current frontier models offer a realistic starting point for AI-driven molecular design.

By centering smart glasses as wearable-intelligence platforms, this survey provides a unified framework for organizing technologies, applications, and open challenges in this emerging area.

Inspired by human cognition, CRCD reformulates the task as counterfactual latent change captioning to bypass affirmation bias, and contrasts a synthesized safe expectation with reality to identify semantic omissions.

An autonomous AI agent integrating Reasoning and Acting with Retrieval-Augmented Generation with Retrieval-Augmented Generation is developed to classify 10,211 Human Phenotype Ontology terms, enabling standardized panel design by providing reliable, automated classification supported by direct evidence.

SALR is proposed, a schema-anchored latent reasoning method for LF construction that performs multi-step reasoning by generating continuous thoughts in the model's hidden states, thereby delaying the explicit commitment to LF decisions.

NeuSOGA3D (Neuro-Symbolic Geometric Abstraction in 3D), a hybrid framework that combines learned perceptual priors inherited from NeuSOGA with explicit symbolic geometric reasoning for explainable geometric intelligence, is presented.

UnifiedPlayers, a cooperative framework comprising a Planning Player that generates tasks, an Execution Player that produces multi-turn trajectories with Python tool calls, and an Evaluation Player that constructs executable verifiers, highlights cooperation among specialized players as a promising path toward self-enhanced tool-integrated agents.

The latency with respect to the resource dynamics of processing multiple requests and tasks concurrently is characterized and new optimization opportunities that exploit the resource dynamics of tasks are demonstrated: CPU-aware tool admission and task-aware CPU allocation.

PetriBench is introduced, a compact, fully self-contained, and scalable benchmark for evaluating LLM reasoning over dynamic state spaces using Petri nets, a mature formalism for modeling real-world concurrent and distributed systems.

The results argue that useful self-reports remain sensitive to elicitation, correlated errors, and benchmark noise, and that useful self-reports remain sensitive to elicitation, correlated errors, and benchmark noise.

These findings support long-context inference in which pretrained models guide their own access to the information they retain, by introducing On-Demand Attention (ODA), a local-first decoding method that uses a lightweight recall head to selectively invoke global attention as its predicted benefit changes during generation.

This work defines the task of safety as a highly pragmatic and critical cognitive dimension, and proposes the Cognitive Expected Scene Graph (CESG) Score, a structure-grounded, polarity-aware evaluation metric.

This work proposes Abstract Token Curriculum (ATC), a novel curriculum learning framework that elicits effective continuous intermediate representations without direct supervision or manual scratchpad design and shows ATC's effectiveness on graph reachability and arithmetic learning tasks.

This work introduces History-Conditioned Minimax Trajectory Search (HC-MTS), which combines temporal occlusion reasoning with response-aware search and solves a bilevel minimax problem.

The results show that reliable scenario generation requires verifying realized behavior, not merely executable code, and demonstrate the value of explicit intermediate representations for verifiable and repairable language-driven co-simulation.

SnapPhysics enables physically interactive MR experiences without manual parameter tuning by combining instance-level 3D reconstruction and spatial alignment with a physics-aware scene graph that encodes these relationships and per-object metric geometry as structured context for VLM-based property reasoning.

It is argued that chess provides a useful bridge between cognitive and computational perspectives on strategic reasoning, and identifies explicit planning, grounded and faithful explanation, metacognitive calibration, and human--AI complementarity as directions for future research.

This work proposes DualSQL, a new Text-to-SQL system consisting of two agents powered by a single model backbone, enabling joint optimization through a robust multi-agent reinforcement learning (RL) framework, and introduces a new SQL correctness metric, robust execution match (REX), to more accurately judge SQL correctness and assign reward signals.

Rosetta is presented, a multi-agent LLM pipeline that automatically generates first-principles analytical models from research paper PDFs, and four design decisions address failure modes of na\"ive LLM-based generation.

This work introduces agentic-eCAL, generalizing the Energy Cost of AI Lifecycle (eCAL) metric to directed multi-agent workflows by coupling a closed-form two-rate single-call energy model (compute-bound prefill and memory-bound decode) with 7-layer OSI data transport.

The findings show that even the presence of a small percentage of biased agents in such a system leads to significant shifts in the opinions of non-biased agents, and semantic analysis demonstrates that rhetorical consistency in textual explanations increases systematically with biased exposure and is partially decoupled from numerical convergenumericalutral agents.

It is shown that epistemic uncertainty leaves a measurable trace within intermediate layers, where spikes in attention entropy are associated with reasoning breakdowns, and that attention dispersion is a promising complement to traditional hallucination detection methods.

This work uses teacher-successful problems to define a cheap reference for what the student can learn and measures how the likelihood of each observed token in trajectories from teacher-failed problems changes as an operational learnability signal, which can be computed once from stored trajectories and model checkpoints.

This work introduces rubric-based preference reasoning, which uses automated rubrics to guide the model toward producing explicit preference states, and proposes a preference-conditioned optimization to align response generation with dual objectives: user preference satisfaction and situation consistency.

GuardEn (Guarding by Safety Rule Entailment), an executable safeguard framework that decomposes safety policies into atomic propositions through Safety-Rule Compilation, modeling their composition as executable code, is proposed.

This work formalizes a Wiki Graph schema that seamlessly bridges fine-grained structures with dense contexts, maintaining explicit topologies alongside continuous semantics and engineer an infrastructural NCCL boundary exchange protocol that hoists static partition indices and leverages fixed-shape GPU-to-GPU collectives, bypassing CPU serialization and memory copy overheads.

This study shows that without explicit network-layer indication, MCP traffic structurally and temporally resembles Command and Control beaconing behavior, specifically the polling architectures used by advanced persistent threats like Cobalt Strike, and proposes an agent-native network indication standard including Agent-Native ALPN and standardized out-of-band headers.

This work introduces a comprehensive evaluation framework organized around four complementary dimensions of physical world reasoning, and constructs a diverse set of novel tasks that require models to integrate complementary information across modalities.

It is argued that harnessing physical-digital AI agents, through persistent coupling of digital intelligence with physical sensing, dynamics, action, and feedback, provides a system-level pathway toward adaptive and continuously evolving UAV autonomy.

The model family shows gains in held-out scientific-code repair and across selected general-purpose benchmarks in code, reasoning, and knowledge, providing evidence of positive transfer from scientific experience to broader capabilities.

Teaching Memory is proposed, a design approach that treats instructional reasoning as a primary form of knowledge and preserves instructional reasoning in a lightweight manner within instructors' existing workflows, enabling continuity and reuse.

MUSE decouples image annotation from question generation, enabling diverse tasks with controllable difficulty while reducing annotation effort, and it is hoped MUSE will serve as a standardized benchmark for advancing multi-modal understanding in situated educational applications.

EviGen is proposed, a three-layer framework for verifiable clinical rationale generation that improves prediction performance and rationale faithfulness over full-context LLM and RAG baselines, and is preferred by clinical reviewers in a usability evaluation.

This work introduces a structured framework for claim-level discourse analysis in dense health narratives through tuples linking atomic claims with thematic aspects, stance, and multidimensional pragmatic discourse attributes and evaluates automated structured discourse analysis under different discourse context settings.

Experiments demonstrate that Code-as-Auditor delivers more accurate and evidence-backed evaluations, enabling automated compliance regulation checking grounded in explicit regulatory criteria and improves the logical consistency of the generated code and the traceability of outcomes.

Echo is presented, a matching decompilation system based on trusted back-translation that uses compilation not only for verification, but also as trusted feedback to guide iterative search in exact matching for optimized binaries under unknown compilation configurations.

The results show that candidate count alone is not enough to describe the systems cost of multi-candidate test-time scaling and Evaluations should report not only candidate count and accuracy, but also generation schedule and GPU-level systems metrics.

This article categorizes existing UAV inspection architectures, identifies their key system challenges and architectural requirements, and experimentally assesses the feasibility of semantic edge intelligence on NVIDIA Jetson UAV-class hardware using the COCO-Bridge dataset.

This work proposes ASPIRE, a non-synchronized batched self-speculative decoding framework built on three components, which achieves speedup in decoding throughput over autoregressive baselines and improves average speedup by approximately $27\% over the strongest prior self-speculative baselines.

AutoTuring hands the same agent the same 15-dimensional accelerator space twice: once as named architectural knobs with simulator counters, once as anonymous variables on [0,1], with the evaluator, the legal space and the reachable optima held identical, so that the only thing that varies is whether the problem means anything.