Large Reasoning Models

Follow this topic Ask about this topic

Large reasoning models study how language models can solve difficult problems by allocating computation across intermediate steps, tools, and alternative solution paths. Common approaches include reinforcement learning with verifiable rewards, supervised reasoning-trace distillation, tool use, search over sequential or parallel paths, and latent recurrent computation. Evaluation spans mathematics, coding, science question answering, logic, and multimodal tasks, measuring not only accuracy but also compute use, robustness, and the quality of intermediate reasoning. A central tension is that extra thinking can help on hard problems but also produce loops, errors, and safety risks, so researchers are developing adaptive stopping and compression while testing whether reasoning traces reliably reflect how answers are formed.

Latest papers

339 papers25 papers in the last four weeks, up 4% on the four weeks before. 0.1% of all new papers.

May 4, 2026

Evaluating Reasoning Models for Queries with Presuppositions

Millions of users turn to AI models for their information needs. It is conceivable that a large number of user queries contain assumptions that may be factually inaccurate. Prior work notes that large language models (LLMs) often fail to challenge such erroneous assumptions, and can reinforce users' misinformed opinions. However, given the recent advances, especially in model's reasoning capabilities, we revisit whether large reasoning models (LRMs) can reason about the underlying assumptions and respond to user queries appropriately. We construct queries with varying degrees of presuppositions spanning health, science, and general knowledge, and use it to evaluate several widely-deployed models When compared to non-reasoning models, we find that reasoning models achieve a slightly higher accuracy (2-11%), but they still fail to challenge a large fraction (26-42%) of false presuppositions. Further, reasoning models remain susceptible to how strongly the presupposition is expressed.

Rose Sathyanathan, Kinshuk Vasisht, Danish Pruthi

Indian Institute of Science Bengaluru, KA, India

LLM Evaluation Language Model Robustness Large Reasoning Models Language Model Reasoning Evaluation

May 4, 2026

Distilling Long-CoT Reasoning through Collaborative Step-wise Multi-Teacher Decoding

Distilling large reasoning models is essential for making Long-CoT reasoning practical, as full-scale inference remains computationally prohibitive. Existing curation-based approaches select complete reasoning traces post-hoc, overlooking collaboration among heterogeneous teachers and lacking dynamic exploration, which leads to redundant sampling and missed complementary reasoning. We introduce CoRD, a collaborative multi-teacher decoding framework that performs step-wise reasoning synthesis guided by predictive perplexity-based scoring and beam search. This enables heterogeneous LRMs to jointly construct coherent reasoning trajectories while efficiently preserving diverse, high-potential hypotheses. Experiments show that CoRD produces higher-quality reasoning data and achieves near teacher-level student performance with fewer, structured supervision signals, without substantial efficiency overhead. CoRD further generalizes well to out-of-domain and open-ended settings. The dataset and model are available at https://github.com/DISL-Lab/CoRD.

Taewon Yun, Jisu Shin, Jeonghwan Choi +2

KAIST · UNIST

CoT Distillation Efficient Language Model Reasoning Multi-Teacher Knowledge Distillation Teacher-Student Learning Multi-LLM Collaboration Large Reasoning Models

May 4, 2026

Towards Understanding Specification Gaming in Reasoning Models

Specification gaming is a critical failure mode of LLM agents. Despite this, there has been little systematic research into when it arises and what drives it. To address this, we build and open source a diverse suite of tasks where models can score highly by taking unintended actions. We find that all tested models exploit their specifications at non-negligible rates in most of our eight settings, including five non-coding settings. We see the highest rates of specification gaming in Grok 4 and the lowest rates in Claude models. We use our evaluation suite to study what drives specification gaming, and find that: 1. RL reasoning training substantially increases the rate at which models exploit their specifications, 2. Increasing RL reasoning budget has a weakly positive effect on exploit rate, and 3. Test-time mitigations reduce but do not eliminate the rate of specification gaming. Our results suggest that specification gaming is a fundamental challenge arising from RL reasoning training; we release our evaluation suite to support further work on this problem.

Kei Nishimura-Gasparian, Robert McCarthy, David Lindner

1MATS · University College London, London, UK · 3Google Deep- Mind.

Reward Hacking RL for Language Model Reasoning AI Agent Benchmarks Large Reasoning Models

May 3, 2026

Spatiotemporal Hidden-State Dynamics as a Signature of Internal Reasoning in Large Language Models

Large reasoning models (LRMs) generate extended solutions, yet it remains unclear whether these traces reflect substantive internal computation or merely verbosity and overthinking. Although recent hidden-state analyses suggest that internal representations carry correctness-related signals, their coarse aggregations may obscure the token and layer structure underlying reasoning computation. We investigate hidden-state transitions across decoding steps and layers, and identify a distinct spatiotemporal pattern in LRMs: successful trajectories exhibit broad temporal dynamics with localized layer-wise concentration, while this structure is weaker in non-reasoning models and knowledge-heavy domains. We formalize this characteristic as Spatiotemporal Amplitude of Latent Transition (StALT), a training-free trajectory statistic that summarizes temporal changes between adjacent tokens weighted by within-token layer saliency. Across diverse models and benchmarks, StALT reliably separates correct from incorrect trajectories in reasoning-intensive regimes, providing a competitive label-free correctness signal alongside strong output-space and length-based baselines. Intervention analyses further show that this spatiotemporal amplitude responds systematically to manipulations that increase or reduce the demand for internal reasoning, supporting its association with latent reasoning dynamics in LRMs. These findings provide empirical evidence that LRMs exhibit measurable hidden-state dynamics and offer a practical probe for understanding internal computation beyond output-based evaluation.

Kotaro Furuya, Takahito Tanimura

Research and Development Group, Hitachi, Ltd. 1-280 Higashi-Koigakubo, Kokubunji-shi, Tokyo 185-8601, Japan

Transformer Interpretability LLM Evaluation LLM Interpretability Large Reasoning Models

May 3, 2026

Only Say What You Know: Calibration-Aware Generation for Long-Form Factuality

Large Reasoning Models achieve strong performance on complex tasks but remain prone to hallucinations, particularly in long-form generation where errors compound across reasoning steps. Existing approaches to improving factuality, including abstention and factuality-driven optimization, follow a coupled exploration-commitment paradigm, in which intermediate reasoning is unconditionally propagated to the final output, limiting fine-grained control over information selection and integration. In this paper, we propose an Exploration-Commitment Decoupling paradigm that disentangles knowledge exploration from final commitment, enabling models to explore with awareness while answering cautiously. We instantiate the paradigm with Calibration-Aware Generation (CAG), a framework that equips models with end-to-end, calibration-aware generation capabilities, by augmenting intermediate reasoning with calibrated reliability estimates and prioritizing reliable content in final outputs. Across five long-form factuality benchmarks and multiple model families, CAG improves factuality by up to 13%, while reducing decoding time by up to 37%. Overall, our work highlights decoupling as a principled approach for more reliable long-form generation, offering directions for trustworthy and self-aware generative systems.

Wen Luo, Guangyue Peng, Liang Wang +7

State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University Research Asia · MSRA · Microsoft

Long-Form Document Generation LLM Hallucination Mitigation Hallucination in Language Models Language Model Calibration LLM Reliability Large Reasoning Models

May 1, 2026

When Less is Enough: Efficient Inference via Collaborative Reasoning

In this work, we introduce DUET (Dual-model Efficient Two-stage inference), a collaborative inference framework in which a capable model and a lightweight model work together to solve a task. Relying on a single large model to perform end-to-end reasoning and prediction often incurs substantial inference cost. In contrast, DUET decomposes inference into two stages: the capable model produces a reasoning signal, and the lightweight model interprets this signal to generate the final answer, allowing reasoning-intensive computation to be handled by the capable model while non-reasoning-intensive components are delegated to the lightweight model without sacrificing task performance. To achieve this objective, we propose a length-penalized joint training objective that encourages the capable model to transmit only the information that is sufficient for the lightweight model to solve the task. As a result, DUET maintains strong reasoning performance with substantially lower inference cost than end-to-end inference using a large model alone, saving up to 60% of the large model's output tokens on challenging reasoning benchmarks, including AIME and GPQA.

Yilei Chen, Sharut Gupta, Yannis Paschalidis +2

Boston University · MIT CSAIL · Biohub

Efficient Language Model Reasoning Efficient Language Model Inference LLM Reasoning Large Reasoning Models

Apr 30, 2026

Retrieval-Augmented Reasoning for Chartered Accountancy

The inception of Large Language Models (LLMs) has catalyzed AI adoption in the finance sector, yet their reliability in complex, jurisdiction-specific tasks like Indian Chartered Accountancy (CA) remains limited. The models display difficulty in executing numerical tasks which require multiple steps while also needing advanced knowledge about legal regulations and the method of scaling their operations is not feasible in settings which have limited access to resources. We present CA-ThinkFlow as a parameter-efficient Retrieval-Augmented Generation (RAG) framework which operates with a 14B, 4-bit-quantized reasoning model, 14B-DeepSeek-R1, and a layout-aware Docling extraction system which maintains document structure during extraction. CA-ThinkFlow uses a basic RAG method which automatically adds retrieved information into the prompt, while it depends on the model's built-in Chain-of-Thought (CoT) functions to create context and produce correct answers. The system we developed system operates at performance levels which match large proprietary models when we tested it on the multi-level CA-Ben benchmark, achieving Scholastic Reliability Coefficient (SRC) results which equal 68.75% of GPT-4o and Claude 3.5 Sonnet. The framework shows high efficiency and strength in handling parameters, but essential reasoning abilities fail to process complex regulatory texts which exist in fields such as Taxation.

Jatin Gupta, Akhil Sharma, Saransh Singhania +1

Dept. of Computer Science and Engineering Sharda University, Greater Noida, India · Sharda University, Greater Noida, India

Retrieval-Augmented Generation Financial QA LLM Reasoning Large Reasoning Models

Apr 30, 2026

State Stream Transformer (SST) V2: Parallel Training of Nonlinear Recurrence for Latent Space Reasoning

Current transformers discard their rich latent residual stream between positions, reconstructing latent reasoning context at each new position and leaving potential reasoning capacity untapped. The State Stream Transformer (SST) V2 enables parameter-efficient reasoning in continuous latent space through an FFN-driven nonlinear recurrence at each decoder layer, where latent states are streamed horizontally across the full sequence via a learned blend. This same mechanism supports continuous latent deliberation per position at inference time, dedicating additional FLOPs to exploring abstract reasoning before committing to a token. A two-pass parallel training procedure approximates the sequential recurrence, making co-training computationally practical. Hidden state analysis shows that the state stream facilitates reasoning through sharp, content-dependent reorganisations in continuous latent space; the LM head exposes the resulting latent belief states through the output distribution, while the state stream carries them forward to influence future positions. A learned probe shows that at the first generated token, the latent state already predicts whether the eventual answer will survive or break under additional latent computation for every subsequent position. Co-trained into an existing 27B backbone using only a small dataset of GSM8K examples and evaluated using an oracle to route each question to a depth of one to four recurrent forward passes per token, the SST achieves an architectural capacity bound of 61.11% on out-of-distribution GPQA-Diamond, a +15.15 point gain over a fine-tuning-matched baseline, and cuts that same baseline's remaining GSM8K errors by 46%. Together, these results provide a method for efficiently training a nonlinear recurrence and show that the state stream provides additional reasoning capacity beyond that of an otherwise training-matched transformer.

Thea Aviss

Fifth Dimension

Recurrent Neural Networks Recurrent Transformers Large Reasoning Models Latent Reasoning

Apr 29, 2026

Path-Lock Expert: Separating Reasoning Mode in Hybrid Thinking via Architecture-Level Separation

Hybrid-thinking language models expose explicit think and no-think modes, but current designs do not separate them cleanly. Even in no-think mode, models often emit long and self-reflective responses, causing reasoning leakage. Existing work reduces this issue through better data curation and multi-stage training, yet leakage remains because both modes are still encoded in the same feed-forward parameters. We propose Path-Lock Expert (PLE), an architecture-level solution that replaces the single MLP in each decoder layer with two semantically locked experts, one for think and one for no-think, while keeping attention, embeddings, normalization, and the language-model head shared. A deterministic control-token router selects exactly one expert path for the entire sequence, so inference preserves the dense model's per-token computation pattern and each expert receives mode-pure updates during supervised fine-tuning. Across math and science reasoning benchmarks, PLE maintains strong think performance while producing a substantially stronger no-think mode that is more accurate, more concise, and far less prone to reasoning leakage. On Qwen3-4B, for example, PLE reduces no-think reflective tokens on AIME24 from 2.54 to 0.39 and improves no-think accuracy from 20.67% to 40.00%, all while preserving think-mode performance. These results suggest that controllable hybrid thinking is fundamentally an architectural problem, and separating mode-specific feed-forward pathways is a simple and effective solution.

Shouren Wang, Wang Yang, Chuang Ma +7

1Case Western Reserve University · 2NII LLMC, Japan · 3Michigan State University

Mixture-of-Experts Language Models LLM Reasoning Large Reasoning Models

Apr 29, 2026

When to Retrieve During Reasoning: Adaptive Retrieval for Large Reasoning Models

Large reasoning models such as DeepSeek-R1 and OpenAI o1 generate extended chains of thought spanning thousands of tokens, yet their integration with retrieval-augmented generation (RAG) remains fundamentally misaligned. Current RAG systems optimize for providing context before reasoning begins, while reasoning models require evidence injection during multi-step inference chains. We introduce ReaLM-Retrieve, a reasoning-aware retrieval framework that addresses this mismatch through three key innovations: (1) a step-level uncertainty detector that identifies knowledge gaps at reasoning-step granularity rather than token or sentence level; (2) a retrieval intervention policy that learns when external evidence maximally benefits ongoing reasoning; and (3) an efficiency-optimized integration mechanism that reduces per-retrieval overhead by 3.2x compared to naive integration. Experiments on MuSiQue, HotpotQA, and 2WikiMultiHopQA demonstrate that ReaLM-Retrieve achieves on average 10.1% absolute improvement in answer F1 over standard RAG (range: 9.0-11.8% across the three benchmarks) while reducing retrieval calls by 47% compared to fixed-interval approaches like IRCoT (all improvements significant at p<0.01, paired bootstrap). On the challenging MuSiQue benchmark requiring 2-4 hop reasoning, our method achieves 71.2% F1 with an average of only 1.8 retrieval calls per question. Analysis shows that ReaLM-Retrieve also improves retrieval quality itself, achieving 81.3% Recall@5 with consistently higher precision and MRR than fixed-interval baselines on supporting evidence, establishing new state-of-the-art efficiency-accuracy trade-offs for reasoning-intensive retrieval tasks.

Dongxin Guo, Jikun Wu, Siu Ming Yiu

The University of Hong Kong Hong Kong, China · Brain Investing Limited Hong Kong, China · Stellaris AI Limited Hong Kong, China

Retrieval-Augmented Generation Retrieval-Augmented Reasoning Multi-Hop Reasoning Large Reasoning Models

Apr 29, 2026

When to Vote, When to Rewrite: Disagreement-Guided Strategy Routing for Test-Time Scaling

Large Reasoning Models (LRMs) achieve strong performance on mathematical reasoning tasks but remain unreliable on challenging instances. Existing test-time scaling methods, such as repeated sampling, self-correction, and tree search, improve performance at the cost of increased computation, yet often exhibit diminishing returns on hard problems. We observe that output disagreement is strongly correlated with instance difficulty and prediction correctness, providing a useful signal for guiding instance-level strategy selection at test time. Based on this insight, we propose a training-free framework that formulates test-time scaling as an instance-level routing problem, rather than allocating more computation within a single strategy, dynamically selecting among different scaling strategies based on output disagreement. The framework applies lightweight resolution for consistent cases, majority voting for moderate disagreement, and rewriting-based reformulation for highly ambiguous instances. Experiments on seven mathematical benchmarks and three models show that our method improves accuracy by 3% - 7% while reducing sampling cost compared to existing approaches.

Zhimin Lin, Yixin Ji, Jinpeng Li +5

School of Computer Science and Technology, Soochow University · Department of Foundation Model, 2012 Labs, Huawei · 3Harbin Institute of Technology, Shenzhen (HITSZ)

Test-Time Scaling LLM Routing LLM Mathematical Reasoning Large Reasoning Models

Apr 27, 2026

Stabilizing Efficient Reasoning with Step-Level Advantage Selection

Large language models (LLMs) achieve strong reasoning performance by allocating substantial computation at inference time, often generating long and verbose reasoning traces. While recent work on efficient reasoning reduces this overhead through length-based rewards or pruning, many approaches are post-trained under a much shorter context window than base-model training, a factor whose effect has not been systematically isolated. We first show that short-context post-training alone, using standard GRPO without any length-aware objective, already induces substantial reasoning compression-but at the cost of increasingly unstable training dynamics and accuracy degradation. To address this, we propose Step-level Advantage Selection (SAS), which operates at the reasoning-step level and assigns a zero advantage to low-confidence steps in correct rollouts and to high-confidence steps in verifier-failed rollouts, where failures often arise from truncation or verifier issues rather than incorrect reasoning. Across diverse mathematical and general reasoning benchmarks, SAS improves average Pass@1 accuracy by 0.86 points over the strongest length-aware baseline while reducing average reasoning length by 16.3%, yielding a better accuracy-efficiency trade-off.

Han Wang, Xiaodong Yu, Jialian Wu +4

UNC Chapel Hill · Advanced Micro Devices, Inc.

RL for Language Model Reasoning Efficient Language Model Reasoning Efficient Language Model Inference Fine-Grained Credit Assignment Large Reasoning Models

Apr 25, 2026

UpstreamQA: A Modular Framework for Explicit Reasoning on Video Question Answering Tasks

Video Question Answering (VideoQA) demands models that jointly reason over spatial, temporal, and linguistic cues. However, the task's inherent complexity often requires multi-step reasoning that current large multimodal models (LMMs) perform implicitly, leaving their internal decision process opaque. In contrast, large reasoning models (LRMs) explicitly generate intermediate logical steps that enhance interpretability and can improve multi-hop reasoning accuracy. Yet, these models are not designed for native video understanding, as they typically rely on static frame sampling. We propose UpstreamQA, a modular framework that disentangles and evaluates core video reasoning components through explicit upstream reasoning modules. Specifically, we employ multimodal LRMs to perform object identification and scene context generation before passing enriched reasoning traces to downstream LMMs for VideoQA. We evaluate UpstreamQA on the OpenEQA and NExTQA datasets using two LRMs (o4-mini, Gemini 2.5 Pro) and two LMMs (GPT-4o, Gemini 2.5 Flash). Our results demonstrate that introducing explicit reasoning can significantly boost performance and interpretability of downstream VideoQA, but can also lead to performance degradation when baseline performance is sufficiently high. Overall, UpstreamQA offers a principled framework for combining explicit reasoning and multimodal understanding, advancing both performance and diagnostic transparency in VideoQA in several scenarios.

Jason Nguyen, Ameet Rao, Alexander Chang +2

Lincoln North Star High School · The Charter School of Wilmington · Greenwich High School +2

Multimodal Large Language Models Video QA Multimodal CoT Reasoning Large Reasoning Models

Apr 22, 2026

TRACES: Tagging Reasoning Steps for Adaptive Cost-Efficient Early-Stopping

The field of Language Reasoning Models (LRMs) has been very active over the past few years with advances in training and inference techniques enabling LRMs to reason longer, and more accurately. However, a growing body of studies show that LRMs are still inefficient, over-generating verification and reflection steps. Additionally, the high-level role of each reasoning step and how different step types contribute to the generation of correct answers, is largely underexplored. To address this challenge, we introduce TRACES (Tagging of the Reasoning steps enabling Adaptive Cost-Efficient early-Stopping), a lightweight framework that tags reasoning steps in real-time, and enable adaptive, cost-efficient early stopping of large-language-model inferences. Building on this framework we monitor reasoning behaviors during inferences, and we find that LRMs tend to shift their reasoning behavior after reaching a correct answer. We demonstrate that the monitoring of the specific type of steps can produce effective interpretable early stopping criteria. We evaluate the TRACES framework on three mathematical reasoning benchmarks, namely, MATH500, GSM8K, AIME and two knowledge and reasoning benchmarks, MMLU and GPQA respectively. We achieve 20 to 50% token reduction while maintaining comparable accuracy to standard generation.

Yannis Belkhiter, Seshu Tirupathi, Giulio Zizzo +1

IBM Research Europe · Trinity College Dublin · Dublin, Ireland +1

Reasoning Trace Analysis Early Stopping Efficient Language Model Reasoning LLM Inference Efficiency Efficient Language Model Inference LLM Reasoning+2

Apr 21, 2026

Binary Agreement for Efficient Unsupervised Confidence Calibration of Reasoning LLMs

Reasoning language models often lack the calibrated confidence needed for reliable deployment, particularly when correctness labels are unavailable and sampling budgets are limited. We introduce Binary Agreement for Self-consistency Calibration (BASC), which uses self-consistency as an unlabeled proxy for correctness without constructing costly self-consistency targets for each calibration query. Our key observation is that agreement between an anchor response and a single independent probe is a Bernoulli observation whose conditional mean is the anchor's latent self-consistency. Pooling these observations across unlabeled queries, BASC supports both recalibrating existing confidence scores and learning more discriminative scores before recalibration, using as few as two responses per calibration query and only one at deployment. The use of binary isotonic recalibration provides a distribution-free calibration guarantee with respect to latent self-consistency. Across 5 tasks and 9 reasoning models, BASC substantially improves calibration with respect to correctness.

Thomas Zollo, Jimmy Wang, Richard Zemel

Columbia University

Confidence Estimation in Language Models Language Model Calibration LLM Reliability LLM Uncertainty Estimation Self-Consistency in Language Models Large Reasoning Models

Apr 21, 2026

How Much Thinking is Enough? Quantifying and Understanding Redundancy in LLM Reasoning

Reasoning-capable large language models solve hard problems by emitting long chains of thought, paying heavily in latency, GPU time, and energy. Casual inspection of their traces reveals extensive reformulation, verification, and circular self-reflection, yet how much of this deliberation is actually necessary has never been measured at scale or explained from first principles. This paper closes both gaps. We formalise reasoning redundancy directly in terms of the reasoning model itself: the redundancy of a correct trace is the largest fraction of its trailing segmented steps that can be truncated while

π

, forced to terminate thinking and emit a final answer, still produces the correct answer. A large-scale quantification across four frontier reasoning models and two mathematical benchmarks shows that step-level redundancy is consistently high -- between 61% and 93% across the 8 (model, benchmark) conditions we study, with the median critical prefix equal to a single segmented step in six of the eight conditions -- that the finding is robust to the choice of judge family, and that although

ρ

decreases with problem difficulty on MATH-500, all four models remain substantially redundant (

ρ\in [46\%, 85\%]

) even on the hardest Level-5 problems. We then prove that this redundancy is a structural consequence of length-agnostic outcome rewards, not a model-specific artefact: under any such reward, no finite expected stopping time is optimal. The result holds regardless of RL algorithm, base model, data distribution, or whether the policy is obtained via RL or distillation; over-thinking is therefore not a bug to be patched in individual models but a structural property of how current reasoning models are trained. Code: https://github.com/zhiyuanZhai20/how-much-thinking-is-enough

Zhiyuan Zhai, Xinkai You, Wenjing Yan +1

Fudan University · The Chinese University of Hong Kong

LLM Inference Efficiency RL for Language Model Reasoning Efficient Language Model Reasoning LLM Mathematical Reasoning LLM Reasoning Large Reasoning Models+1

Apr 21, 2026

More Thinking, More Bias: Length-Driven Position Bias in Reasoning Models

Chain-of-thought (CoT) reasoning and reasoning-tuned models such as DeepSeek-R1 are commonly assumed to reduce shallow heuristic biases by thinking carefully. We test this on position bias in multiple-choice QA and find a different story: within any reasoning-capable model, per-question position bias scales with the length of the reasoning trajectory. Across thirteen reasoning-mode configurations (two R1-distilled 7-8B models, two base models prompted with CoT, and DeepSeek-R1 at 671B) on MMLU, ARC-Challenge, and GPQA, twelve show a positive partial correlation between trajectory length and Position Bias Score (PBS) after controlling for accuracy, ranging from 0.11 to 0.41 (all p < 0.05). All twelve open-weight reasoning-mode configurations show monotonically increasing PBS across length quartiles. A truncation intervention provides causal evidence: continuations resumed from later points in the trajectory are increasingly likely to shift toward position-preferred options (16% to 32% for R1-Qwen-7B across absolute-position buckets). At 671B, aggregate PBS collapses to 0.019, but the length effect still manifests in the longest quartile (PBS = 0.071), suggesting that accuracy gates the expression of length-driven bias rather than eliminating the underlying mechanism. We additionally find that direct-answer position bias is a distinct phenomenon with a different footprint (strong in Llama-Instruct-direct, weak in Qwen-Instruct-direct, and uncorrelated with trajectory length): CoT reasoning replaces this baseline bias with length-accumulated bias. Our results argue that reasoning-capable models should not be treated as order-robust by default in MCQ evaluation pipelines, and offer a diagnostic toolkit (PBS, commitment change point, effective switching, truncation probes) for auditing position bias in reasoning models.

Xiao Wang

Multiple-Choice Question Answering CoT Reasoning Language Model Bias Evaluation Position Bias Large Reasoning Models

Apr 21, 2026

When Safety Fails Before the Answer: Benchmarking Harmful Behavior Detection in Reasoning Chains

Large reasoning models (LRMs) produce complex, multi-step reasoning traces, yet safety evaluation remains focused on final outputs, overlooking how harm emerges during reasoning. When jailbroken, harm does not appear instantaneously but unfolds through distinct behavioral steps such as suppressing refusal, rationalizing compliance, decomposing harmful tasks, and concealing risk. However, no existing benchmark captures this process at sentence-level granularity within reasoning traces -- a key step toward reliable safety monitoring, interventions, and systematic failure diagnosis. To address this gap, we introduce HarmThoughts, a benchmark for step-wise safety evaluation of reasoning traces. \ourdataset is built on our proposed harm taxonomy of 16 harmful reasoning behaviors across four functional groups that characterize how harm propagates rather than what harm is produced. The dataset consists of 56,931 sentences from 1,018 reasoning traces generated by four model families, each annotated with fine-grained sentence-level behavioral labels. Using HarmThoughts, we analyze harm propagation patterns across reasoning traces, identifying common behavioral trajectories and drift points where reasoning transitions from safe to unsafe. Finally, we systematically compare white-box and black-box detectors on the task of identifying harmful reasoning behaviours on HarmThoughts. Our results show that existing detectors struggle with fine-grained behavior detection in reasoning traces, particularly for nuanced categories within harm emergence and execution, highlighting a critical gap in process-level safety monitoring. HarmThoughts is available publicly at: https://huggingface.co/datasets/ishitakakkar-10/HarmThoughts

Ishita Kakkar, Enze Zhang, Rheeya Uppaal +1

Department of Computer Sciences, University of Wisconsin-Madison

AI Safety Evaluation LLM Safety Evaluation Large Reasoning Models Reasoning Trace Analysis

Apr 21, 2026

Reasoning Structure Matters for Safety Alignment of Reasoning Models

Large reasoning models (LRMs) achieve strong performance on complex reasoning tasks but often generate harmful responses to malicious user queries. This paper investigates the underlying cause of these safety risks and shows that the issue lies in the reasoning structure itself. Based on this insight, we claim that effective safety alignment can be achieved by altering the reasoning structure. We propose AltTrain, a simple yet effective post training method that explicitly alters the reasoning structure of LRMs. AltTrain is both practical and generalizable, requiring no complex reinforcement learning (RL) training or reward design, only supervised finetuning (SFT) with a lightweight 1K training examples. Experiments across LRM backbones and model sizes demonstrate strong safety alignment, along with robust generalization across reasoning, QA, summarization, and multilingual setting.

Yeonjun In, Wonjoong Kim, Sangwu Park +1

KAIST

LLM Safety Alignment Large Reasoning Models

Apr 21, 2026

Fine-Tuning Small Reasoning Models for Quantum Field Theory

Despite the growing application of large language models (LLMs) to theoretical physics, there has been little academic exploration of how domain-specific physics reasoning ability develops during training. To investigate this, we perform the first academic comparison of reinforcement learning and supervised fine-tuning for LLMs on verifiable problem-solving tasks in theoretical physics. Because the open-source verifiable training data required to develop such capabilities is scarce, we built a robust data generation pipeline that can both create synthetic problems and make existing human-authored problems suitable for model training. Selecting quantum field theory (QFT) as our primary domain, we generated over 2,500 synthetic problems alongside a curated collection of human-adapted problems sourced from arXiv and standard pedagogical resources. We conduct both reinforcement learning (RL) and supervised fine-tuning (SFT) experiments, benchmarking performance gains as well as generalization to other physics domains. Although both methods improve over the base model on every dataset, SFT gains are concentrated on in-distribution synthetic tasks, while RL generalizes better, outperforming SFT on the human-adapted datasets and TPBench. We perform an extensive analysis of model chain-of-thought (CoT) reasoning before and after fine-tuning to understand how reasoning errors evolve during RL and SFT. Finally, we publicly release our data pipeline, verifiable QFT training data, and ~200M tokens of QFT reasoning traces.

Nathaniel S. Woodward, Zhiqi Gao, Yurii Kvasiuk +3

Department of Physics, University of Wisconsin-Madison · Department of Computer Science, University of Wisconsin-Madison · Perimeter Institute for Theoretical Physics

Physical Reasoning LLM Fine-Tuning RL for Language Model Reasoning Large Reasoning Models

Apr 20, 2026

Reasoning Models Know What's Important, and Encode It in Their Activations

Language models often solve complex tasks by generating long reasoning chains, consisting of many steps with varying importance. While some steps are crucial for generating the final answer, others are removable. Determining which steps matter most, and why, remains an open question central to understanding how models process reasoning. We investigate if this question is best approached through model internals or through tokens of the reasoning chain itself. We find that model activations contain more information than tokens for identifying important reasoning steps. Crucially, by training probes on model activations to predict importance, we show that models encode an internal representation of step importance, even prior to the generation of subsequent steps. The internal representations of importance in different models yield high agreement on which steps are important. The representation is distributed across layers, and does not correlate with surface-level features, such as a step's relative position or its length. Our findings suggest that analyzing activations can reveal aspects of reasoning that surface-level approaches fundamentally miss, indicating that reasoning analyses should look into model internals.

Yaniv Nikankin, Martin Tutek, Tomer Ashuach +2

Technion · University of Zagreb, FER · MIT +1

LLM Interpretability LLM Reasoning Large Reasoning Models Reasoning Trace Analysis

Apr 18, 2026

Playing Psychic: Using Thought Trees to Predict Reasoning Models Accuracy on Coding Tasks

Recent advances in large language models (LLMs) have shown that test-time scaling can substantially improve model performance on complex tasks, particularly in the coding domain. Under this paradigm, models use a larger token budget during inference to generate intermediate reasoning traces before producing a final answer. However, current evaluations primarily rely on competitive programming benchmarks, which may not capture the full range of reasoning abilities. In this work, we perform a systematic study of frontier reasoning models to understand their performance on real-world coding benchmarks. To gain more insights into the performance of such models, we devise a programmatic way to {\em automatically generate} coding tasks of arbitrary difficulty and structure from existing benchmarks. Using this framework, our analysis reveals that the structure of a reasoning trace, not just its contents, is a strong predictor of correctness. Motivated by this, we propose structured thought-trees as means to represent reasoning traces. To illustrate their use, we train a lightweight classifier on features extracted from thought-trees to predict trace correctness, and demonstrate that flagging and retrying structurally anomalous traces based on the extracted features yields consistent gains at lower complexity levels.

Jiaxin Fang, Runyuan He, Sahil Bhatia +2

University of California, Berkeley

Structured Reasoning Code Generation Evaluation Code Generation Benchmarks Large Reasoning Models Reasoning Trace Analysis

Apr 18, 2026

x1: Learning to Think Adaptively Across Languages and Cultures

Languages encode distinct abstractions and inductive priors, yet most large language models (LLMs) overlook this diversity by reasoning in a single dominant language. In this work, we introduce x1, a family of reasoning models that can adaptively reason in an advantageous language on a per-instance basis. To isolate the effect of reasoning-language choice, x1 is constructed without expanding the model's knowledge boundaries and is trained by contrasting linguistically distinct reasoning trajectories for the same input. Our extensive experiments demonstrate the benefits of adaptive multilingual reasoning across multilingual mathematical reasoning and culturally grounded tasks. Moreover, our results challenge a simplistic view of scaling laws: while scaling reduces cross-lingual disparities in procedural domains such as math reasoning, it does not eliminate the advantages of culture-associated languages in culturally grounded tasks, as we empirically show that such reasoning enables more efficient and accurate cultural knowledge recall. Overall, our findings establish language choice as a functional component of reasoning, with implications for building more generalist and globally competent reasoning models.

Yangfan Ye, Xiaocheng Feng, Xiachong Feng +8

Harbin Institute of Technology · Peng Cheng Laboratory · The University of Hong Kong +1

Multilingual Language Models Cross-Lingual Reasoning in Language Models Cross-Lingual Transfer Large Reasoning Models Adaptive Reasoning in Language Models

Apr 18, 2026

Step-GRPO: Internalizing Dynamic Early Exit for Efficient Reasoning

Large reasoning models that use long chain-of-thought excel at problem-solving yet waste compute on redundant checks. Curbing this overthinking is hard: training-time length penalties can cripple ability, while inference-time early-exit adds system overhead. To bridge this gap, we propose Step-GRPO, a novel post-training framework that internalizes dynamic early-exit capabilities directly into the model. Step-GRPO shifts the optimization objective from raw tokens to semantic steps by utilizing linguistic markers to structure reasoning. We introduce a Dynamic Truncated Rollout mechanism that exposes the model to concise high-confidence trajectories during exploration, synergized with a Step-Aware Relative Reward that dynamically penalizes redundancy based on group-level baselines. Extensive experiments across three model sizes on diverse benchmarks demonstrate that Step-GRPO achieves a superior accuracy-efficiency trade-off. On Qwen3-8B, our method reduces token consumption by 32.0% compared to the vanilla model while avoiding the accuracy degradation observed in traditional length-penalty methods.

Benteng Chen, Weida Wang, Shufei Zhang +2

East China Normal University · Shanghai AI Laboratory · Fudan University +1

RL for Language Model Reasoning LLM Inference Acceleration Early-Exit Neural Networks Large Reasoning Models

Apr 17, 2026

RankGuide: Tensor-Rank-Guided Routing and Steering for Efficient Reasoning

Large reasoning models (LRMs) enhance problem-solving capabilities by generating explicit multi-step chains of thought (CoT) reasoning; however, they incur substantial inference latency and computational overhead. To mitigate this issue, recent works have explored model collaboration paradigms, where small reasoning models (SRMs) generate intermediate reasoning steps to achieve a better accuracy--latency trade-off. Despite recent progress, effectively and efficiently detecting and mitigating SRM failures in collaborative systems remains a key challenge. To address this issue, we analyze SRM inference in both the generated text and hidden-state spaces, and identify three types of failure modes: overconfidence, uncertainty, and heavy revalidation. Building on these insights, we propose RankGuide, a framework that improves the efficiency and effectiveness of SRM--LRM collaboration through tensor-rank-guided routing and steering. Specifically, RankGuide leverages a routing signal that incorporates tensor-rank signals derived from consecutive hidden states to detect when SRMs are likely to fail and selectively invoke LRMs. In addition, we introduce a tensor-rank-filtered steering vector extraction method to modulate the reasoning trajectory of SRMs, thereby improving their generation quality. By improving both routing and steering through tensor-rank signals, RankGuide enables SRM--LRM collaborative systems to achieve more efficient reasoning with fewer steps and improved accuracy. Experiments on multiple reasoning benchmarks demonstrate the efficacy of RankGuide in reducing latency by up to

1.75\times

compared to LRM, while maintaining competitive accuracy relative to prior methods.

Jiayi Tian, Yupeng Su, Ryan Solgi +2

University of California, Santa Barbara · Intel Labs

LLM Routing Efficient Language Model Inference Activation Steering Multi-LLM Collaboration Large Reasoning Models

Apr 17, 2026

Cut Your Losses! Learning to Prune Paths Early for Efficient Parallel Reasoning

Parallel reasoning enhances Large Reasoning Models (LRMs) but incurs prohibitive costs due to futile paths caused by early errors. To mitigate this, path pruning at the prefix level is essential, yet existing research remains fragmented without a standardized framework. In this work, we propose the first systematic taxonomy of path pruning, categorizing methods by their signal source (internal vs. external) and learnability (learnable vs. non-learnable). This classification reveals the unexplored potential of learnable internal methods, motivating our proposal of STOP (Super TOken for Pruning). Extensive evaluations across LRMs ranging from 1.5B to 20B parameters demonstrate that STOP achieves superior effectiveness and efficiency compared to existing baselines. Furthermore, we rigorously validate the scalability of STOP under varying compute budgets - for instance, boosting GPT-OSS-20B accuracy on AIME25 from 84% to nearly 90% under fixed compute budgets. Finally, we distill our findings into formalized empirical guidelines to facilitate optimal real-world deployment. Code, data and models are available at https://bijiaxihh.github.io/STOP

Jiaxi Bi, Tongxu Luo, Wenyu Du +2

The Chinese University of Hong Kong, Shenzhen · Shenzhen Loop Area Institute · DualityRL

LLM Inference Acceleration Large Reasoning Models Parallel Reasoning in Language Models

Apr 17, 2026

CiPO: Counterfactual Unlearning for Large Reasoning Models through Iterative Preference Optimization

Machine unlearning has gained increasing attention in recent years, as a promising technique to selectively remove unwanted privacy or copyrighted information from Large Language Models that are trained on a massive scale of human data. However, the emergence of Large Reasoning Models (LRMs), which emphasize long chain-of-thought (CoT) reasoning to address complex questions, presents a dilemma to unlearning: existing methods either struggle to completely eliminate undesired knowledge from the CoT traces or degrade the reasoning performances due to the interference with the reasoning process. To this end, we introduce Counterfactual Unlearning through iterative Preference Optimization (CiPO), a novel framework that redefines unlearning as the targeted intervention of the CoT reasoning in LRMs. More specifically, given a desired unlearning target answer, CiPO instructs LRMs to generate a logically valid counterfactual reasoning trace for preference tuning. As the LRM adjusts to the counterfactual trace, CiPO iteratively updates the preference learning data to increase the discrepancy from the original model. This iterative loop ensures both desirable unlearning and smooth optimization, effectively mitigating the dilemma. Experiments on challenging benchmarks demonstrate that CiPO excels at unlearning, completely removing knowledge from both the intermediate CoT steps and the final answer, while preserving the reasoning abilities of LRMs.

Junyi Li, Yongqiang Chen, Ningning Ding

The Hong Kong University of Science and Technology (Guangzhou) · The Chinese University of Hong Kong

Machine Unlearning LLM Unlearning Large Reasoning Models

Apr 17, 2026

Reasoning-targeted Jailbreak Attacks on Large Reasoning Models via Semantic Triggers and Psychological Framing

Large Reasoning Models (LRMs) have demonstrated strong capabilities in generating step-by-step reasoning chains alongside final answers, enabling their deployment in high-stakes domains such as healthcare and education. While prior jailbreak attack studies have focused on the safety of final answers, little attention has been given to the safety of the reasoning process. In this work, we identify a novel problem that injects harmful content into the reasoning steps while preserving unchanged answers. This type of attack presents two key challenges: 1) manipulating the input instructions may inadvertently alter the LRM's final answer, and 2) the diversity of input questions makes it difficult to consistently bypass the LRM's safety alignment mechanisms and embed harmful content into its reasoning process. To address these challenges, we propose the Psychology-based Reasoning-targeted Jailbreak Attack (PRJA) Framework, which integrates a Semantic-based Trigger Selection module and a Psychology-based Instruction Generation module. Specifically, the proposed PRJA automatically selects manipulative reasoning triggers via semantic analysis and leverages psychological theories of obedience to authority and moral disengagement to generate adaptive instructions for enhancing the LRM's compliance with harmful content generation. Extensive experiments on five question-answering datasets demonstrate that PRJA achieves an average attack success rate of 83.6% against several commercial LRMs, including DeepSeek R1, Qwen2.5-Max, and OpenAI o4-mini.

Zehao Wang, Lanjun Wang

College of Intelligence and Computing, Tianjin University, Tianjin 300350, China. · School of New Media and Communication, Tianjin University, Tianjin 300072, China. · Shanghai Key Laboratory of Data Science,[cs.LG] Shanghai 200438, China.

Adversarial Prompt Generation Jailbreak Attacks LLM Jailbreak Attacks Large Reasoning Models

Apr 16, 2026

LACE: Lattice Attention for Cross-thread Exploration

Current large language models reason in isolation. Although it is common to sample multiple reasoning paths in parallel, these trajectories do not interact, and often fail in the same redundant ways. We introduce LACE, a framework that transforms reasoning from a collection of independent trials into a coordinated, parallel process. By repurposing the model architecture to enable cross-thread attention, LACE allows concurrent reasoning paths to share intermediate insights and correct one another during inference. A central challenge is the absence of natural training data that exhibits such collaborative behavior. We address this gap with a synthetic data pipeline that explicitly teaches models to communicate and error-correct across threads. Experiments show that this unified exploration substantially outperforms standard parallel search, improving reasoning accuracy by over 7 points. Our results suggest that large language models can be more effective when parallel reasoning paths are allowed to interact.

Yang Li, Zirui Zhang, Yang Liu +1

Department of Computer Science Rutgers University · Amazon AGI Labs

Parallel Reasoning in Language Models LLM Inference Attention Mechanisms Synthetic Data Generation for Language Models Large Reasoning Models

Apr 16, 2026

Think Multilingual, Not Harder: A Data-Efficient Framework for Teaching Reasoning Models to Code-Switch

Recent developments in reasoning capabilities have enabled large language models to solve increasingly complex mathematical, symbolic, and logical tasks. Interestingly, while reasoning models are often trained to generate monolingual text, these models have also been observed to code-switch (i.e., mix languages). Prior works have either viewed code-switching as an undesirable error, attempted to control code-switching through modifications to input prompts or the output decoding process, or focus on narrow subsets of languages, domains, tasks, and models. We address these gaps by introducing the first linguistically and behaviorally motivated fine-tuning framework for identifying beneficial code-switched reasoning behaviors in large language models and teaching these models to code-switch more effectively for reasoning. First, we create and systematically analyze a dataset of reasoning traces from diverse models, languages, tasks, and domains to understand the types of code-switching behaviors found in existing reasoning models. Then, we develop fine-tuning interventions that teach reasoning models to code-switch based on our observations of helpful behaviors in existing models. We find that our framework can significantly increase beneficial code-switched reasoning behaviors in a data-efficient manner. Interestingly, we also find that code-switching behaviors in reasoning models can be modified by fine-tuning for tasks that do not directly demonstrate code-switching in reasoning (e.g., machine translation). Our work suggests that data-efficient interventions can instill helpful forms of code-switching behavior in reasoning models.

Eleanor M. Lin, David Jurgens

University of Michigan

Multilingual Language Models Code-Switching Cross-Lingual Reasoning in Language Models LLM Fine-Tuning Large Reasoning Models