LLM Reasoning

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

68 papers in the last four weeks, up 143% on the four weeks before. 0.7% of all new papers.

Jul 13Week of Sep 28

Latest papers 650

Apr 17, 2026cs.AI

LLM Reasoning Is Latent, Not the Chain of Thought

This position paper argues that large language model (LLM) reasoning should be studied as latent-state trajectory formation rather than as faithful surface chain-of-thought (CoT). This matters because claims about faithfulness, interpretability, reasoning benchmarks, and inference-time intervention all depend on what the field takes the primary object of reasoning to be. We ask what that object should be once three often-confounded factors are separated and formalize three competing hypotheses: H1, reasoning is primarily mediated by latent-state trajectories; H2, reasoning is primarily mediated by explicit surface CoT; and H0, most apparent reasoning gains are better explained by generic serial compute than by any privileged representational object. Reorganizing recent empirical, mechanistic, and survey work under this framework, and adding compute-audited worked exemplars that factorize surface traces, latent interventions, and matched budget expansions, we find that current evidence most strongly supports H1 as a default working hypothesis rather than as a task-independent verdict. We therefore make two recommendations: the field should treat latent-state dynamics as the default object of study for LLM reasoning, and it should evaluate reasoning with designs that explicitly disentangle surface traces, latent states, and serial compute.

Wenshuo Wang

School of Future Technology, South China University of Technology, China

LLM Interpretability CoT Reasoning LLM Reasoning Latent Reasoning

Apr 16, 2026cs.CL

From Tokens to Steps: Verification-Aware Speculative Decoding for Efficient Multi-Step Reasoning

Speculative decoding (SD) accelerates large language model inference by allowing a lightweight draft model to propose outputs that a stronger target model verifies. However, its token-centric nature allows erroneous steps to propagate. Prior approaches mitigate this using external reward models, but incur additional latency, computational overhead, and limit generalizability. We propose SpecGuard, a verification-aware speculative decoding framework that performs step-level verification using only model-internal signals. At each step, SpecGuard samples multiple draft candidates and selects the most consistent step, which is then validated using an ensemble of two lightweight model-internal signals: (i) an attention-based grounding score that measures attribution to the input and previously accepted steps, and (ii) a log-probability-based score that captures token-level confidence. These signals jointly determine whether a step is accepted or recomputed using the target, allocating compute selectively. Experiments across a range of reasoning benchmarks show that SpecGuard improves accuracy by 3.6% while reducing latency by ~11%, outperforming both SD and reward-guided SD.

Kiran Purohit, Ramasuri Narayanam, Soumyabrata Pal

IIT Kharagpur · Adobe Research

LLM Answer Verification Speculative Decoding LLM Inference Acceleration LLM Reasoning

Apr 16, 2026cs.AI

IG-Search: Step-Level Information Gain Rewards for Search-Augmented Reasoning

Reinforcement learning has emerged as an effective paradigm for training large language models to perform search-augmented reasoning. However, existing approaches rely on trajectory-level rewards that cannot distinguish precise search queries from vague or redundant ones within a rollout group, and collapse to a near-zero gradient signal whenever every sampled trajectory fails. In this paper, we propose IG-Search, a reinforcement learning framework that introduces a step-level reward based on Information Gain (IG). For each search step, IG measures how much the retrieved documents improve the model's confidence in the gold answer relative to a counterfactual baseline of random documents, thereby reflecting the effectiveness of the underlying search query. This signal is fed back to the corresponding search-query tokens via per-token advantage modulation in GRPO, enabling fine-grained, step-level credit assignment within a rollout. Unlike prior step-level methods that require either externally annotated intermediate supervision or shared environment states across trajectories, IG-Search derives its signals from the policy's own generation probabilities, requiring no intermediate annotations beyond standard question-answer pairs. Experiments on seven single-hop and multi-hop QA benchmarks demonstrate that IG-Search achieves an average EM of 0.430 with Qwen2.5-3B, outperforming the strongest trajectory-level baseline (MR-Search) by 1.6 points and the step-level method GiGPO by 0.9 points on average across benchmarks, with particularly pronounced gains on multi-hop reasoning tasks. Despite introducing a dense step-level signal, IG-Search adds only ~6.4% to per-step training wall-clock time over the trajectory-level baseline and leaves inference latency unchanged, while still providing a meaningful gradient signal even when every sampled trajectory answers incorrectly.

Zihan Liang, Yufei Ma, Ben Chen +6

Kuaishou Technology

Retrieval-Augmented Generation RL for Language Model Reasoning Credit Assignment in RL Step-Level Credit Assignment LLM Reasoning

Apr 16, 2026cs.CL

IE as Cache: Information Extraction Enhanced Agentic Reasoning

Information Extraction aims to distill structured, decision-relevant information from unstructured text, serving as a foundation for downstream understanding and reasoning. However, it is traditionally treated merely as a terminal objective: once extracted, the resulting structure is often consumed in isolation rather than maintained and reused during multi-step inference. Moving beyond this, we propose \textit{IE-as-Cache}, a framework that repurposes IE as a cognitive cache to enhance agentic reasoning. Drawing inspiration from hierarchical computer memory, our approach combines query-driven extraction with cache-aware reasoning to dynamically maintain compact intermediate information and filter noise. Experiments on challenging benchmarks across diverse LLMs demonstrate significant improvements in reasoning accuracy, indicating that IE can be effectively repurposed as a reusable cognitive resource and offering a promising direction for future research on downstream uses of IE.

Hang Lv, Sheng Liang, Hongchao Gu +5

University of Science and Technology of China · Huawei Technologies Co., Ltd.

Agentic Reasoning LLM Information Extraction LLM Reasoning Memory-Augmented Language Model Agents

Apr 16, 2026cs.AI

Governing Reflective Human-AI Collaboration: A Framework for Epistemic Scaffolding and Traceable Reasoning

Large language models have advanced rapidly, from pattern recognition to emerging forms of reasoning, yet they remain confined to linguistic simulation rather than grounded understanding. They can produce fluent outputs that resemble reflection, but lack temporal continuity, causal feedback, and anchoring in real-world interaction. This paper proposes a complementary approach in which reasoning is treated as a relational process distributed between human and model rather than an internal capability of either. Building on recent work on "System-2" learning, we relocate reflective reasoning to the interaction layer. Instead of engineering reasoning solely within models, we frame it as a cognitive protocol that can be structured, measured, and governed using existing systems. This perspective emphasizes collaborative intelligence, combining human judgment and contextual understanding with machine speed, memory, and associative capacity. We introduce "The Architect's Pen" as a practical method. Like an architect who thinks through drawing, the human uses the model as an external medium for structured reflection. By embedding phases of articulation, critique, and revision into human-AI interaction, the dialogue itself becomes a reasoning loop: human abstraction -> model articulation -> human reflection. This reframes the question from whether the model can think to whether the human-AI system can reason. The framework enables auditable reasoning traces and supports alignment with emerging governance standards, including the EU AI Act and ISO/IEC 42001. It provides a practical path toward more transparent, controllable, and accountable AI use without requiring new model architectures.

Rikard Rosenbacke, Carl Rosenbacke, Victor Rosenbacke +1

Faculty of Medicine, Lund University, Sweden · Department of Economics, Lund University School of Economics and Management, Sweden · Department of Health Services Research and Policy, London School of Hygiene & Tropical Medicine, UK

AI Governance Human-AI Collaboration LLM Reasoning

Apr 16, 2026cs.AI

Dissecting Failure Dynamics in Large Language Model Reasoning

Large Language Models (LLMs) achieve strong performance through extended inference-time deliberation, yet how their reasoning failures arise remains poorly understood. By analyzing model-generated reasoning trajectories, we find that errors are not uniformly distributed but often originate from a small number of early transition points, after which reasoning remains locally coherent but globally incorrect. These transitions coincide with localized spikes in token-level entropy, and alternative continuations from the same intermediate state can still lead to correct solutions. Based on these observations, we introduce GUARD, a targeted inference-time framework that probes and redirects critical transitions using uncertainty signals. Empirical evaluations across multiple benchmarks confirm that interventions guided by these failure dynamics lead to more reliable reasoning outcomes. Our findings highlight the importance of understanding when and how reasoning first deviates, complementing existing approaches that focus on scaling inference-time computation.

Wei Zhu, Jian Zhang, Lixing Yu +2

School of Information Science and Engineering, Yunnan University, Kunming, China · Yunnan Key Laboratory of Intelligent Systems and Computing, Kunming, China

Inference-Time Guidance LLM Reasoning

Apr 16, 2026cs.AI

Quantifying Cross-Query Contradictions in Multi-Query LLM Reasoning

Large language models frequently produce mutually inconsistent answers when reasoning over multiple related queries. We study case-file logical consistency: maintaining a globally satisfiable belief state across interdependent queries. We introduce a benchmark of 390 multi-query reasoning instances with entailment/contradiction/unknown labels and propose set-level metrics including Case Satisfiability Rate, Contradiction Density and Revision Cost. Our solver-augmented approach extracts commitments, verifies global satisfiability and performs counterexample-guided repair. Across four reasoning domains, our method substantially reduces cross-query contradictions (SetCons: 0.56 to 0.94) while preserving per-query accuracy, demonstrating that global coherence is critical for robust multi-query reasoning.

Rohit Kumar Salla, Ramya Manasa Amancherla, Manoj Saravanan

Virginia Tech · Columbia University

Apr 15, 2026cs.CL

Correct Prediction, Wrong Steps? Consensus Reasoning Knowledge Graph for Robust Chain-of-Thought Synthesis

Large language models (LLMs) have become increasingly used for various tasks, often coupled with Chain-of-Thought (CoT) prompting to boost accuracy. Recent work has shown that high label-prediction accuracy does not guarantee correct intermediate reasoning, and the causes of reasoning flaws vary from sample to sample, yet existing remedies either focus on a single domain or assume that one flaw type applies uniformly across samples. A simple mitigation method is to provide the model with the correct answer, but we show that this yields no consistent improvement in reasoning quality. This indicates that the problem cannot be fixed by LLMs' awareness of answers, and must instead be addressed through the structure of reasoning. Motivated by this, we propose CRAFT (Consensus Reasoning-knowledge-graph Aggregation for Flaw-aware Trace synthesis), which aggregates the consensus components shared across multiple candidate reasoning traces to synthesize improved ones. CRAFT consistently improves label-prediction accuracy on both logical and mathematical reasoning benchmarks, outperforming most baselines, while its post-processed traces achieve higher quality under fine-grained benchmark evaluation.

Zipeng Ling, Shuliang Liu, Seonil Son +4

The Hong Kong University of Science and Technology (Guangzhou) · University of Alberta · Alberta Machine Intelligence Institute +3

Logical Reasoning Self-Consistency Decoding CoT Reasoning Mathematical Reasoning Structured Reasoning LLM Reliability+2

Apr 2, 2026cs.AI

How LLMs Might Think

Do large language models (LLMs) think? Daniel Stoljar and Zhihe Vincent Zhang have recently developed an argument from rationality for the claim that LLMs do not think. We contend, however, that the argument from rationality not only falters, but leaves open an intriguing possibility: that LLMs engage only in arational, associative forms of thinking, and have purely associative minds. Our positive claim is that if LLMs think at all, they likely think precisely in this manner.

Joseph Gottlieb, Ethan Kemp, Matthew Trager

Dept. of Philosophy, Texas Tech University · Dept. of Philosophy, New York University · Dept. of History and Philosophy of Science, University of Pittsburgh, Matthew Trager

Apr 1, 2026cs.LG

Reasoning Shift: How Context Silently Shortens LLM Reasoning

Large language models (LLMs) exhibiting test-time scaling behavior, such as extended reasoning traces and self-verification, have demonstrated remarkable performance on complex, long-term reasoning tasks. However, the robustness of these reasoning behaviors remains underexplored. To investigate this, we conduct a systematic evaluation of multiple reasoning models across three scenarios: (1) problems augmented with lengthy, irrelevant context; (2) multi-turn conversational settings with independent tasks; and (3) problems presented as a subtask within a complex task. We observe an interesting phenomenon: reasoning models tend to produce much shorter reasoning traces (up to 74%) for the same problem under different context conditions compared to the traces produced when the problem is presented in isolation. A finer-grained analysis reveals that this compression is associated with a decrease in self-verification and uncertainty management behaviors, such as double-checking. Importantly, we show that even when additional self-checks are forced, their efficiency depends not only on the content of the reasoning traces, but also on the presence of redundant context. We hope our findings draw additional attention to both the robustness of reasoning models and the problem of context management for LLMs.

Gleb Rodionov, Roman Garipov, George Yakushev

Yandex · HSE University

Language Model Robustness LLM Context Management LLM Reasoning Large Reasoning Models

Mar 26, 2026cs.CL

SafeMath: Safe Solutions for Unsafe Math Word Problems

Recent research points toward LLMs being manipulated through adversarial and seemingly benign inputs, resulting in harmful, biased, or policy-violating outputs. In this paper, we study an underexplored issue concerning harmful and toxic mathematical word problems. We show that math questions, particularly those framed as natural language narratives, can serve as a subtle medium for propagating biased, unethical, or psychologically harmful content, with heightened risks in educational settings involving children. To support a systematic study of this phenomenon, we introduce ToxicGSM, a dataset of 1.9k arithmetic problems in which harmful or sensitive context is embedded while preserving mathematically well-defined reasoning tasks. Using this dataset, we audit the behaviour of existing LLMs and analyse the trade-offs between safety enforcement and mathematical correctness. We further propose SafeMath -- a safety alignment technique that reduces harmful outputs while maintaining, and in some cases improving, mathematical reasoning performance. Our results highlight the importance of disentangling linguistic harm from math reasoning and demonstrate that effective safety alignment need not come at the cost of accuracy.

Sagnik Basu, Subhrajit Mitra, Aman Juneja +3

1Indian Institute of Technology Kharagpur · 2Cisco Systems · 3National University of Singapore

LLM Safety Alignment LLM Safety Evaluation LLM Reasoning

Mar 23, 2026cs.LG

Thinking Deeper, Not Longer: Memory-Efficient Test-Time Reasoning with Depth-Recurrent Transformers for Compositional Generalization

Standard Transformers have a fixed computational depth, limiting their ability to generalize to tasks that require variable-depth reasoning. The usual remedy, Chain-of-Thought (CoT), spends tokens to reason, inflating the key--value cache and making latency grow with the step count, so memory becomes the limiting cost when reasoning is served over large query batches. We study a depth-recurrent Transformer that decouples computational depth from parameter count by iterating a shared-weight block, so that each added reasoning step costs flat memory and linear latency, with no token generation. Three ingredients keep the recurrence stable for 20+ thinking steps: a silent thinking objective that supervises only the final output, LayerScale initialization, and an identity-biased gate that opens a gradient highway across steps. We characterize it on three compositional domains with decreasing structural bias: graph reachability (adjacency masking), nested boolean logic (relative positioning), and unstructured relational text (no positional cue). We find a \emph{computational frontier}: accuracy climbs once the thinking-step count meets the task's complexity, reaching near-perfect performance on the two structured tasks and a lower plateau on unstructured text. How it climbs depends on the structural bias---abruptly from chance on the graph task, gradually on the other two. Depth recurrence extrapolates beyond the training range: it succeeds on the graph task where fixed-depth models barely extrapolate, and on the two sequence tasks comes within two points of fixed-depth Transformers that use

4

--

6.4\times

more parameters. On the graph task, whose adjacency mask makes propagation depth verifiable, intermediate per-step supervision---a standard recipe for deep iterative models---consistently \emph{harms} this extrapolation. We release the code for reproducibility.

Hung-Hsuan Chen

Computer Science and Information Engineering National Central University Taoyuan, Taiwan

Efficient Transformer Inference Recurrent Transformers Structured Reasoning Algorithmic Reasoning Compositional Generalization LLM Reasoning

Mar 22, 2026cs.CL

Beyond Memorization: Distinguishing Between Pattern-Based and Epistemic Reasoning in LLMs Using Epistemic Puzzles

Epistemic reasoning requires agents to infer the state of the world from partial observations and information about other agents' knowledge. Prior work evaluating LLMs on epistemic puzzles often frames failures as memorization rather than reasoning. We argue that this dichotomy is too coarse for newer models: memorization is a limiting case of pattern-based reasoning, where a model matches a task to a familiar template and applies the corresponding solution. We introduce a two-dimensional benchmark over DEL-style puzzles, separating narrative familiarity from inference complexity, allowing us to distinguish pattern-based from epistemic reasoning. We find that models are substantially more robust to surface form changes than prior work suggested, yet consistently struggle in asymmetric settings where familiar patterns no longer apply and success requires tracking fragmented epistemic states.

Adi Gabay, Gabriel Stanovsky, Liat Peterfreund

School of Computer Science and Engineering The Hebrew University of Jerusalem

Epistemic Logic LLM Reasoning

Mar 20, 2026cs.AI

Grounded Chess Reasoning in Language Models via Master Distillation

Language models often lack grounded reasoning capabilities in specialized domains where training data is scarce but bespoke systems excel. We introduce a general framework for distilling expert system reasoning into natural language chain-of-thought explanations, enabling compact models to acquire domain expertise and the ability to generate faithful, grounded explanations. Rather than distilling only final outputs, we capture the full reasoning process, transforming opaque expert computations into transparent, step-by-step explanations. We demonstrate this approach in chess, a canonical reasoning domain where language models continue to underperform. Our 4B parameter model, C1, advances from a near-zero baseline to 48.1% accuracy, outperforming all open-source models and most frontier proprietary systems. Notably, C1 surpasses its distillation teacher and generates solutions in two orders of magnitude fewer tokens than baselines. Unlike prior neural chess approaches that predict only best moves, C1 generates explainable solutions revealing strategic reasoning. Our pipeline combines supervised fine-tuning and reinforcement learning with theme-balanced data sampling for comprehensive tactical coverage. Master Distillation demonstrates how to inject expert-level knowledge into compact models for under-optimized domains, offering a recipe for unlocking RLVR where LLMs lack sufficient base capabilities.

Zhenwei Tang, Qianfeng Wen, Seth Grief-Albert +4

Department of Computer Science, University of Toronto · Queen’s University · Coolwei AI Lab

LLM Grounding CoT Distillation Language Model Distillation Chess LLM Reasoning

Mar 20, 2026cs.AI

Learning Dynamic Belief Graphs for Theory-of-mind Reasoning

Theory of Mind (ToM) reasoning with Large Language Models (LLMs) requires inferring how people's implicit, evolving beliefs shape what they seek and how they act under uncertainty -- especially in high-stakes settings such as disaster response, emergency medicine, and human-in-the-loop autonomy. Prior approaches either prompt LLMs directly or use latent-state models that treat beliefs as static and independent, often producing incoherent mental models over time and weak reasoning in dynamic contexts. We introduce a structured cognitive trajectory model for LLM-based ToM that represents mental state as a dynamic belief graph, jointly inferring latent beliefs, learning their time-varying dependencies, and linking belief evolution to information seeking and decisions. Our model contributes (i) a novel projection from textualized probabilistic statements to consistent probabilistic graphical model updates, (ii) an energy-based factor graph representation of belief interdependencies, and (iii) an ELBO-based objective that captures belief accumulation and delayed decisions. Across multiple real-world disaster evacuation datasets, our model significantly improves action prediction and recovers interpretable belief trajectories consistent with human reasoning, providing a principled module for augmenting LLMs with ToM in high-uncertainty environment. https://anonymous.4open.science/r/ICML_submission-6373/

Ruxiao Chen, Xilei Zhao, Thomas J. Cova +2

Department of Civil and Systems Engineering, Johns Hopkins University, Baltimore, MD, USA · Civil and Coastal Engineering, University of Florida, Gainesville, FL, USA · School of the Environment, Society & Sustainability, University of Utah, Salt Lake City, UT, USA +1

Computational Cognitive Modeling LLM Reasoning with Graphs Theory of Mind Probabilistic Graphical Models Disaster Response LLM Reasoning

Mar 19, 2026cs.RO

Can LLMs Prove Robotic Path Planning Optimality? A Benchmark for Research-Level Algorithm Verification

Robotic path planning problems are often NP-hard, and practical solutions typically rely on approximation algorithms with provable performance guarantees for general cases. While designing such algorithms is challenging, formally proving their approximation optimality is even more demanding, which requires domain-specific geometric insights and multi-step mathematical reasoning over complex operational constraints. Recent Large Language Models (LLMs) have demonstrated strong performance on mathematical reasoning benchmarks, yet their ability to assist with research-level optimality proofs in robotic path planning remains under-explored. In this work, we introduce the first benchmark for evaluating LLMs on approximation-ratio proofs of robotic path planning algorithms. The benchmark consists of 34 research-grade proof tasks spanning diverse planning problem types and complexity levels, each requiring structured reasoning over algorithm descriptions, problem constraints, and theoretical guarantees. Our evaluation of state-of-the-art proprietary and open-source LLMs reveals that even the strongest models struggle to produce fully valid proofs without external domain knowledge. However, providing LLMs with task-specific in-context lemmas substantially improves reasoning quality, a factor that is more effective than generic chain-of-thought prompting or supplying the ground-truth approximation ratio as posterior knowledge. We further provide fine-grained error analysis to characterize common logical failures and hallucinations, and demonstrate how each error type can be mitigated through targeted context augmentation.

Zhengbang Yang, Md. Tasin Tazwar, Minghan Wei +1

George Mason University. · Florida Atlantic University.

Mathematical Reasoning Benchmarks Automated Theorem Proving Approximation Algorithms Robot Motion Planning LLM Reasoning

Mar 19, 2026cs.AI

ZEBRAARENA: A Diagnostic Simulation Environment for Studying Reasoning-Action Coupling in Tool-Augmented LLMs

Tool-augmented large language models (LLMs) must tightly couple multi-step reasoning with external actions, yet existing benchmarks often confound this interplay with complex environment dynamics, memorized knowledge or dataset contamination. In this paper, we introduce ZebraArena, a procedurally generated diagnostic environment for studying reasoning-action coupling in tool-augmented LLMs, with controllable difficulty and a knowledge-minimal design, which limits gains from memorization or dataset contamination. Each task in ZebraArena requires a set of critical information which is available only through targeted tool use, yielding an interpretable interface between external information acquisition and deductive reasoning. This design provides deterministic evaluation via unique solutions, and a theoretical optimal query count for measuring efficient tool use. We show that ZebraArena requires a combination of in-depth reasoning and accurate external tool calling, which remains a challenge as frontier reasoning models such as GPT-5 and Gemini 2.5 Pro only achieves 60% accuracy on the hard instances. We also observe a persistent gaps between theoretical optimality and practical tool usage. For example, GPT-5 uses 70-270% more tool calls than the theoretical optimum. We highlight the key findings in our evaluation, and hope ZebraArena stimulates further research on the interplay between internal reasoning and external action.

Wanjia Zhao, Ludwig Schmidt, Yejin Choi +3

Stanford University · Microsoft Research

LLM Evaluation AI Agent Benchmarks LLM Tool Use LLM Reasoning

Mar 18, 2026cs.CL

RAWR: Reward Assignment Without Rollouts in Verifiable Domains

Understanding and evaluating multi-step reasoning in LLMs at the level of individual steps remains a key challenge. Process reward models (PRMs) provide a solution by scoring each step, enabling fine-grained supervision and improved reliability. However, training them requires costly human annotation or computationally intensive rollout-based labeling. To solve this, we introduce MCNIG, a scalable method for automatically labeling the quality of individual reasoning steps in any verifiable domain. Its step score, net information gain (NetIG), improves upon single-reference information gain (IG) by comparing the most-supported correct answer against the most-supported incorrect one, yielding a robust signal even for long and structured outputs like code and SQL, where IG fails. We show that the signal produced by MCNIG correlates with human judgments of step quality, and we apply MCNIG labels to train PRMs that achieve the best average best-of-K accuracy across eight benchmarks spanning mathematics, code generation, text-to-SQL, and scientific QA. Crucially, MCNIG generates no rollouts, cutting labeling complexity to O(N) and making it up to X times cheaper than rollout-based methods at comparable label quality, which makes large-scale process supervision practical.

Corentin Royer, Anna Hedström, Debarun Bhattacharjya +3

International Business Machines · Department of Computer Science, ETH Zurich, 8092 Zurich, Switzerland · ETH AI Center +2

Process Reward Models Reinforcement Learning with Verifiable Rewards Process Supervision Step-Level Credit Assignment LLM Reasoning

Mar 16, 2026cs.CL

Can LLMs Model Incorrect Student Reasoning? A Case Study on Distractor Generation

Modeling student misconceptions in a realistic manner is critical for AI in education. In this work, we examine how large language models (LLMs) reason about misconceptions when generating distractor answers for multiple-choice questions (MCQs), a task that requires producing answers that are incorrect, yet plausible. We introduce a taxonomy over reasoning strategies for distractor generation that is grounded in learning-science literature and empirical observation, which we apply to LLM-generated reasoning traces across math and science MCQs. On the math dataset, we find that models follow a misconception-based process with potentially high diagnostic value: they recover the correct solution, articulate student errors, simulate them, and select plausible candidates. On the science dataset, on the other hand, they tend to follow a less robust approach based on semantic similarity to the correct answer. We find the most frequent failure modes to be that the model is unable to generate a correct solution or that it discards plausible distractor candidates when performing selection. Providing the correct solution in the prompt yields a relative improvement of 6.4% in alignment with human-authored distractors, highlighting the critical role of anchoring distractor generation to the correct solution. Together, our findings offer an interpretable view of how LLMs model incorrect student reasoning.

Yanick Zengaffinen, Andreas Opedal, Donya Rooein +3

1ETH Zürich · 2Bocconi University · University of Central Florida

LLM Prompting Generative AI in Education User Modeling Text Generation LLM Reasoning

Mar 13, 2026cs.LG

RetroReasoner: A Reasoning LLM for Strategic Retrosynthesis Prediction

Retrosynthesis prediction aims to identify reactants that can synthesize a given product molecule. Although molecular large language models (LLMs) have recently shown promising results, most existing methods either generate reactants directly or provide only generic product-level analysis, without explicitly reasoning about bond-disconnection strategies that justify specific reactant choices. This paper proposes RetroReasoner, a retrosynthetic reasoning model that captures chemists' strategic disconnection-based thinking. RetroReasoner is trained with supervised fine-tuning and reinforcement learning. For supervised fine-tuning, SyntheticRetro generates structured disconnection rationales paired with reactant predictions. For reinforcement learning, a round-trip reward evaluates predicted reactants by passing them through a forward synthesis model and rewarding predictions that reconstruct the original product. RetroReasoner can also be applied to multi-step retrosynthetic planning by incorporating it into a parallelized Monte Carlo tree search framework, reducing search time while increasing the number and diversity of valid synthetic pathways. Experimental results show that RetroReasoner outperforms prior baselines, including not only molecular LLMs but also retrosynthesis-specific expert models, and generates a broader range of feasible reactant proposals, especially for challenging reaction instances. The code is available at https://github.com/KU-AGI/RetroReasoner.

Hanbum Ko, Chanhui Lee, Ye Rin Kim +4

Department of Artificial Intelligence, Korea University · Materials Intelligence Lab, LG AI Research · Department of Statistics, Korea University

AI-Assisted Scientific Research Language Modeling LLM Reasoning

Mar 11, 2026cs.AI

HEAL: Hindsight Entropy-Assisted Learning for Reasoning Distillation

Distilling reasoning capabilities from Large Reasoning Models (LRMs) into smaller models is typically constrained by the limitations of rejection sampling. Standard methods treat the teacher as a static filter, discarding complex "corner-case" problems where the teacher fails to explore valid solutions independently, thereby creating an artificial "Teacher Ceiling" for the student. In this work, we propose Hindsight Entropy-Assisted Learning (HEAL), an RL-free framework designed to bridge this reasoning gap. Drawing on the educational theory of the Zone of Proximal Development (ZPD), HEAL synergizes three core modules: (1) Guided Entropy-Assisted Repair (GEAR), an active intervention mechanism that detects critical reasoning breakpoints via entropy dynamics and injects targeted hindsight hints to repair broken trajectories; (2) Perplexity-Uncertainty Ratio Estimator (PURE), a ratio-based filtering heuristic that reduces high-anomaly shortcut-like rationales; and (3) Progressive Answer-guided Curriculum Evolution (PACE), a three-stage distillation strategy that organizes training from foundational alignment to hard-case adaptation. Extensive experiments on multiple benchmarks demonstrate that HEAL significantly outperforms traditional SFT distillation and other baselines.

Wenjing Zhang, Jiangze Yan, Jieyun Huang +7

Data Science & AI Research Institute, China Unicom · Unicom Data Intelligence, China Unicom

CoT Distillation Language Model Distillation LLM Reasoning Large Reasoning Models

Mar 10, 2026cs.CV

EXPLORE-Bench: Egocentric Scene Prediction with Long-Horizon Reasoning

Multimodal large language models (MLLMs) are increasingly considered as a foundation for embodied agents, yet it remains unclear whether they can reliably reason about the long-term physical consequences of actions from an egocentric viewpoint. We study this gap through a new task, Egocentric Scene Prediction with LOng-horizon REasoning: given an initial-scene image and a sequence of atomic action descriptions, a model is asked to predict the final scene after all actions are executed. To enable systematic evaluation, we introduce EXPLORE-Bench, a benchmark curated from real first-person videos spanning diverse scenarios. Each instance pairs long action sequences with structured final-scene annotations, including object categories, visual attributes, and inter-object relations, which supports fine-grained, quantitative assessment. Experiments on a range of proprietary and open-source MLLMs reveal a significant performance gap to humans, indicating that long-horizon egocentric reasoning remains a major challenge. We further analyze test-time scaling via stepwise reasoning and show that decomposing long action sequences can improve performance to some extent, while incurring non-trivial computational overhead. Overall, EXPLORE-Bench provides a principled testbed for measuring and advancing long-horizon reasoning for egocentric embodied perception.

Chengjun Yu, Xuhan Zhu, Chaoqun Du +4

University of Science and Technology of China · Foundation Model Team, Li Auto Inc. · Tsinghua University

Multimodal Large Language Models Egocentric Video Understanding Egocentric Vision Multimodal Model Evaluation LLM Reasoning

Mar 9, 2026cs.CL

Learning When to Sample: Confidence-Aware Selective Sampling for Efficient Chain-of-Thought Reasoning

Large language models (LLMs) can achieve strong reasoning performance through chain-of-thought (CoT) reasoning, yet they often generate unnecessarily long reasoning paths that incur high inference cost. Self-consistency-based approaches push accuracy higher still, but they require sampling and aggregating multiple reasoning trajectories, leading to substantial computational overhead. In this paper, we introduce a confidence-aware selective sampling framework that, at inference time, analyzes a single reasoning trajectory to adaptively determine whether to rely on that trajectory alone or trigger multi-path sampling. The framework uses trajectory-level numeric features and sentence-level linguistic features extracted from reasoning states to guide selective multi-path reasoning. We train it on MedQA and evaluate it in-domain on MedQA and under calibration-only transfer on MathQA, MedMCQA, and MMLU, without further fine-tuning. Experimental results show that the proposed framework maintains comparable performance to full and efficient multi-path reasoning baselines, with accuracy changes of

-0.41 \pm 0.58

and

-0.31 \pm 0.58

percentage points, respectively, while reducing token usage by

71.7 \pm 5.0%

and

36.6 \pm 9.1%

. These findings demonstrate that reasoning trajectories contain rich signals for uncertainty estimation, enabling a simple, transferable mechanism to balance accuracy and efficiency in LLM reasoning.

Juming Xiong, Kevin Guo, Congning Ni +7

1Vanderbilt University · 2Vanderbilt University Medical Center · 3Intuit AI Research

CoT Reasoning Efficient Language Model Reasoning LLM Inference Acceleration Adaptive Sampling LLM Uncertainty Estimation LLM Reasoning+1

Mar 5, 2026cs.LG

CRISP: Compressed Reasoning via Iterative Self-Policy Distillation

Reasoning models often generate far more tokens than a task requires, which raises inference cost and can compound errors. We introduce CRISP (Compressed Reasoning via Iterative Self-Policy Distillation), an on-policy self-distillation method that teaches a model to reason more concisely by distilling its own concise behavior back into itself. The method uses a single idea: condition the same model on a "be concise" instruction to obtain teacher logits, then minimize the per-token reverse KL divergence between the student and this teacher on the student's own rollouts. It requires no ground-truth answers, no token budgets, and no difficulty estimators. The reverse-KL objective is naturally difficulty-adaptive: it compresses easy problems aggressively while preserving the reasoning steps that hard problems require. On Qwen3-14B, CRISP cuts reasoning length by up to 56% on MATH-500 and 38% on the harder AIME 2024, while improving MATH-500 accuracy by up to 3.3 points over the base model and holding AIME 2024 accuracy within about one point. This behavior generalizes across model sizes and families: Qwen3-8B shows the same compression with accuracy preserved, and DeepSeek-R1-Distill-Llama-8B improves accuracy on all five benchmarks while shortening its responses. General capabilities are preserved across all three models. Code is available at https://github.com/HJSang/OPSD_Reasoning_Compression.

Hejian Sang, Yuanda Xu, Zhengze Zhou +3

CoT Distillation Efficient Language Model Reasoning On-Policy Distillation CoT Compression LLM Reasoning

Mar 3, 2026cs.LG

Verify to Amplify: Improving Reasoning via Learned Chain-of-Thought Verification

Large Language Models (LLMs) using chain-of-thought have demonstrated great potential for solving complex reasoning and planning tasks. Despite these advances, LLM-generated outputs remain susceptible to errors, making verification important for reliable reasoning systems. Learned verifiers can increase trust, enforce safety constraints, and ensure alignment with personal preferences, while also providing feedback to improve generation. This raises a central challenge: when learned verifiers are used to guide generation, the feedback loop between generator and verifier may induce a distribution shift. This is particularly salient for process reward models, a prominent class of learned verifiers that score or classify individual steps in a chain-of-thought reasoning trace. Motivated by this challenge, we propose a new online learning framework for chain-of-thought verifiers that, given a problem statement and a reasoning trace, check the correctness of each reasoning step given the preceding steps. Highlighting the asymmetric role of soundness errors (accepting an incorrect reasoning step) and completeness errors (flagging a correct step as wrong), we introduce novel notions of dimension that characterize their optimal tradeoff. We then show how our learned verifiers can boost the accuracy of a weak generator. Assuming that the generator can produce a correct next step with a small success probability, we show how to learn a strong generator with small error and abstention rates. Our results also allow learning from offline data when queries to an expert verifier can be simulated from a small set of correct reasoning traces. However, we establish a separation between our approach and learning from offline expert demonstrations: we show that learning from offline demonstrations cannot in general achieve the soundness-completeness guarantees produced by our interactive learning approach.

Maria-Florina Balcan, Avrim Blum, Kiriaki Fragkia +2

Carnegie Mellon University · Toyota Technological Institute at Chicago · Northwestern University

LLM Evaluation LLM Answer Verification Process Reward Models LLM Reliability LLM Reasoning

Mar 3, 2026cs.LG

Step-Level Sparse Autoencoder for Reasoning Process Interpretation

Large Language Models (LLMs) have achieved strong complex reasoning capabilities through Chain-of-Thought (CoT) reasoning. However, their reasoning patterns remain too complicated to analyze. While Sparse Autoencoders (SAEs) have emerged as a powerful tool for interpretability, existing approaches predominantly operate at the token level, creating a granularity mismatch when capturing more critical step-level information, such as reasoning direction and semantic transitions. In this work, we propose step-level sparse autoencoder (SSAE), which serves as an analytical tool to disentangle different aspects of LLMs' reasoning steps into sparse features. Specifically, by precisely controlling the sparsity of a step feature conditioned on its context, we form an information bottleneck in step reconstruction, which splits incremental information from background information and disentangles it into several sparsely activated dimensions. Experiments on multiple base models and reasoning tasks show the effectiveness of the extracted features. By linear probing, we can easily predict surface-level information, such as generation length and first token distribution, as well as more complicated properties, such as the correctness and logicality of the step. These observations indicate that LLMs should already at least partly know about these properties during generation, which provides the foundation for the self-verification ability of LLMs. Our code is available at https://github.com/Miaow-Lab/SSAE.

Xuan Yang, Jiayu Liu, Yuhang Lai +3

Department of Data Science, City University of Hong Kong · Hong Kong Institute of AI for Science, City University of Hong Kong · Li Auto Inc +1

LLM Interpretability Sparse Autoencoders LLM Reasoning

Mar 2, 2026cs.LG

Recursive Models for Long-Horizon Reasoning

Modern language models reason within bounded context, an inherent constraint that poses a fundamental barrier to long-horizon reasoning. We identify recursion as a core principle for overcoming this barrier, and propose recursive models as a minimal realization, where the model can recursively invoke itself to solve subtasks in isolated contexts. We prove that any computable problem admits a recursive decomposition of reasoning in which each subtask requires only exponentially smaller active context than standard autoregressive models; this strictly surpasses any context management approach confined to a single sequence, such as summarization. We further generalize our framework to modern agentic systems with arbitrary context processing and control flows, and prove that recursive models can achieve optimal power within this broader class. Experimentally, we test two settings: fine-tuning a pretrained base model for recursive SAT solving, and training a small model from scratch on Go traces generated by exact game-tree search. Both show improved long-horizon accuracy with small active contexts.

Chenxiao Yang, Nathan Srebro, Zhiyuan Li

Toyota Technological Institute at Chicago

Long-Context Modeling Language Modeling Algorithmic Reasoning LLM Reasoning Recursive Language Models

Feb 18, 2026cs.AI

Framework of Thoughts: A Foundation Framework for Dynamic and Optimized Reasoning based on Chains, Trees, and Graphs

Prompting schemes such as Chain of Thought, Tree of Thoughts, and Graph of Thoughts can significantly enhance the reasoning capabilities of large language models. However, most existing schemes require users to define static, problem-specific reasoning structures that lack adaptability to dynamic or unseen problem types. Additionally, these schemes are often under-optimized in terms of hyperparameters, prompts, runtime, and prompting cost. To address these limitations, we introduce Framework of Thoughts (FoT)--a general-purpose foundation framework for building and optimizing dynamic reasoning schemes. FoT comes with built-in features for hyperparameter tuning, prompt optimization, parallel execution, and intelligent caching, unlocking the latent performance potential of reasoning schemes. We demonstrate FoT's capabilities by implementing three popular schemes--Tree of Thoughts, Graph of Thoughts, and ProbTree--within FoT. We empirically show that FoT enables significantly faster execution, reduces costs, and achieves better task scores through optimization. We release our codebase to facilitate the development of future dynamic and efficient reasoning schemes.

Felix Fricke, Simon Malberg, Georg Groh

School of Computation, Information and Technology Technical University of Munich

Inference-Time Optimization Tree of Thoughts LLM Prompting Efficient Language Model Inference LLM Reasoning

Feb 13, 2026cs.CL

Think Deep, Not Just Long: Measuring LLM Reasoning Effort via Deep-Thinking Tokens

Large language models (LLMs) have demonstrated impressive reasoning capabilities by scaling test-time compute via long Chain-of-Thought (CoT). However, recent findings suggest that raw token counts are unreliable proxies for reasoning quality: increased generation length does not consistently correlate with accuracy and may instead signal "overthinking," leading to performance degradation. In this work, we quantify inference-time effort by identifying deep-thinking tokens -- tokens where internal predictions undergo significant revisions in deeper model layers prior to convergence. Across four challenging mathematical and scientific benchmarks (AIME 24/25, HMMT 25, and GPQA-diamond) and a diverse set of reasoning-focused models (GPT-OSS, DeepSeek-R1, and Qwen3), we show that deep-thinking ratio (the proportion of deep-thinking tokens in a generated sequence) exhibits a robust and consistently positive correlation with accuracy, substantially outperforming both length-based and confidence-based baselines. Leveraging this insight, we introduce Think@n, a test-time scaling strategy that prioritizes samples with high deep-thinking ratios. We demonstrate that Think@n matches or exceeds standard self-consistency performance while significantly reducing inference costs by enabling the early rejection of unpromising generations based on short prefixes.

Wei-Lin Chen, Liqian Peng, Tian Tan +5

University of Virginia · Google

LLM Inference Efficiency Inference-Time Scaling Efficient Language Model Reasoning LLM Reasoning

Feb 10, 2026cs.AI

Discovering High Level Patterns from Simulation Traces

Large Language Models (LLMs) are unable to reliably reason about specific physical systems. Attempts to imbue LLMs with knowledge of the necessary physics concepts have shown great promise, but explainability and validation remain open challenges. An emerging alternative is tooling, where LLMs can query physical simulators and use the resulting simulation traces as context for validation. This approach suffers from poor scalability since simulation traces contain large volumes of fine-grained numerical and semantic data. We show that translating simulation traces to a sparse representation of "high-level" structural patterns leads to more effective interpretation by LLMs. We propose an unsupervised learning scheme to perform this translation, or annotation, via program synthesis. Our learning results in a library of programs that act as pattern detectors which can translate simulation traces to sparse, annotated pattern sequences. The detected patterns may optionally be guided by human experts via string labels (rigid collision, stretching spring, etc.). We show, using a recent physics benchmark, that such annotated representations are more amenable to natural language reasoning about specific physical systems. The synthesized programs serve as transparent, explainable functions that map system states to a sparse and efficient annotation space. As an example application, we show how goals within physical systems that are specified in natural language may be converted to reward programs which are maximized to find solutions.

Sean Memery, Kartic Subr

University of Edinburgh, United Kingdom

LLM Grounding Program Synthesis Physics-Based Simulation LLM Reasoning