RL for Language Model Reasoning

RL: Reinforcement Learning

Follow this topic Ask about this topic

Momentum

91 papers in the last four weeks, up 176% on the four weeks before. 0.9% of all new papers.

Jul 13Week of Sep 28

Latest papers 707

Jun 3, 2026cs.CL

Learning What to Learn: Stage-Specific Data Sets for SFT-then-RL in Small Language Model Reasoning

Post-training Small Language Models (SLMs) for reasoning typically follows an SFT-then-RL pipeline, yet existing work rarely considers what data should be learned at each stage. We argue that data strategy should be aligned with the distinct roles of SFT and RL: SFT is better suited for acquiring not-yet-mastered reasoning skills, while RL is better suited for consolidating skills that the model can already partially access. Based on this principle, we propose a difficulty-aware SFT-then-RL framework that organizes training data into stage-specific sets. For hard samples in the SFT stage, we introduce a Bridge mechanism that transforms raw teacher-generated reasoning traces into more learnable supervision for SLMs. For hard samples that remain unsolved during RL, we apply Critique Fine-Tuning by converting all-zero-reward failures into diagnostic, repair, and new reasoning trace supervision for the next SFT stage. Experiments on two SLMs across five reasoning benchmarks show that our method consistently improves over representative SFT, distillation, and RL baselines. Our results highlight the importance of coordinating data difficulty across SFT and RL for effective SLM reasoning post-training.

Chongyang He, Rui Zhang, Zixuan Wang +1

1Tsinghua University · 2National University of Singapore · 3DiDi +1

Supervised Fine-Tuning RL for Language Model Reasoning Small Language Models Curriculum Learning Language Model Post-Training

Jun 3, 2026cs.CL

Read the Trace, Steer the Path: Trajectory-Aware Reinforcement Learning for Diffusion Language Models

Diffusion large language models (dLLMs) generate responses by iteratively unmasking and revising many positions in parallel. This process leaves a rich denoising trace depicting which tokens become confident, which remain unstable, and when commitments form. Existing dLLM reinforcement learning methods use this signal only weakly. Flat rollouts are cheap, but assign a single outcome reward to the whole trajectory. Tree rollouts provide finer, verifiable training signals by branching partial trajectories and propagating leaf rewards upward, but are compute intensive. We ask whether the denoising trace itself can provide tree-like supervision without tree-level compute. We introduce CAPR (Cached-Amortized Path Refinement), a dLLM-RL algorithm that summarizes the denoising trace into a compact path state, uses cached trajectory states to generate cheap sibling continuations, and trains a block-level value head for local block-wise supervision. Under a block-wise unmasking schedule, CAPR records path-state and block-progress features, then redistributes the final outcome reward across blocks according to the tokens revealed in each block. This trains the value head to convert one sparse reward into block-level PPO weights. CAPR therefore recovers much of the granularity of tree search while avoiding full tree expansion, reducing rollout-generation cost to roughly 0.75x that of flat rollouts and 0.6x that of tree rollouts (under standard settings). Across 4x4 Sudoku, Countdown, GSM8K, and Math500, on dense and mixture-of-experts LLaDA backbones, CAPR sets a new state of the art for RL-tuned dLLMs at 256- and 512-token budgets. On Sudoku, it matches the strongest tree-structured baseline at less than one third of the per-step compute.

Anant Khandelwal, Manish Gupta

Microsoft AI, India

RL for Language Model Reasoning Credit Assignment in RL RL for Diffusion Models Diffusion Language Models

Jun 2, 2026cs.CL

QUBRIC: Co-Designing Queries and Rubrics for RL Beyond Verifiable Rewards

Rubric-based RL is a promising route for extending reinforcement learning beyond verifiable rewards, yet existing methods optimize rubrics while treating the query distribution as fixed. We identify a structural bottleneck: rubric quality is constrained by query structure. Open-ended queries yield vague rubrics; naively narrowing them introduces fabricated references that no model can verify, so all responses fail and training receives no reward signal. We present QUBRIC, a framework that co-designs queries and rubrics. Teacher-derived key points ground the rewriting of open-ended queries into scenario-based, evaluable questions. Contrastive rubric generation then turns teacher-policy gaps into query-level criteria, and learnability filtering retains only informative query-rubric pairs for GRPO training. QUBRIC achieves a +5.5 point gain on ArenaHard over the SFT baseline. Trained only on instruction-following data, it further transfers to three held-out benchmarks spanning legal, moral, and narrative reasoning (+6.3 points on average), with improvements concentrated in reasoning-related dimensions. These results provide evidence that co-designing queries and rubrics can make rubric-based RL a practical complement to RLVR beyond strictly verifiable tasks.

Rongzhi Zhang, Rui Feng, Zhihan Zhang +8

Amazon · Georgia Institute of Technology

Reinforcement Learning RL for Language Model Reasoning Rubric-Based RL Rubric-Based Evaluation Reward Design for RL

Jun 2, 2026cs.AI

PyraMathBench: Evaluating and Improving Mathematical Capability in Large Language Models

Despite the pivotal role of numerical reasoning as the cornerstone of mathematical capabilities in large language models (LLMs) across applications, few benchmarks evaluate LLMs by integrating numerical processing and mathematical reasoning, hindering the interpretability of failures in math tasks. We introduce PyraMathBench, a comprehensive hierarchical benchmark with 32,505 questions derived from 7,404 math word problems, spanning 4 key cognitive aspects, 14 subcategories, and 2 modalities. Experiments reveal that LLMs' performance is severely compromised by inadequate numerical computation and weak handling of abstract numerical questions. To address this, we propose the Smart Optimization & Learning-based VErsatile module (SOLVE) and Interactive Relative Policy Optimization (IRPO), which enhance LLMs' numerical-mathematical synergy via efficient tool calls (fuzzy matching and low-quality call rejection). Comparative experiments show Qwen-2.5 achieves a 5.0 score improvement with SOLVE and IRPO training.

Zetian Ouyang, Linlin Wang, Gerard de Melo +1

East China Normal University · Hasso Plattner Institute, University of Potsdam

Numerical Reasoning in Language Models Mathematical Reasoning Benchmarks RL for Language Model Reasoning Mathematical Reasoning LLM Tool Use LLM Mathematical Reasoning

Jun 2, 2026cs.LG

Tool-Aware Optimization with Entropy Guidance for Efficient Agentic Reinforcement Learning

Agentic reinforcement learning (RL) equips large language models (LLMs) with tool-use capabilities that substantially improve reasoning on complex tasks. However, integrating external tools often destabilizes training: over-reliance on tools can induce input distribution shift, while overly conservative tool use limits effective exploration. To address this issue, we propose a unified framework TAO-RL that couples tool-aware trajectory filtering with entropy-guided exploration for efficient policy optimization. Specifically, at the data level, TAO-RL filters rollout trajectories along two criteria: discarding those where all tool invocations fail to execute, and removing those where all rollouts are either correct or incorrect, as both cases yield degenerate advantage estimates that contribute no discriminative learning signal. This joint filtering retains data that are both tool-capable and informative, establishing a high-quality training distribution. At the algorithmic level, we introduce a tool-aware entropy-guided bonus that reshapes the advantage function at post-tool-call tokens, encouraging the policy to explore more diverse reasoning paths at critical decision points. These two components are mutually reinforcing: trajectory filtering establishes a clean and informative training foundation, while entropy-guided exploration drives stronger reasoning behaviors at critical tool-interaction junctures. Extensive experiments on 7 challenging reasoning benchmarks across 3 model scales demonstrate the superiority of TAO-RL over existing methods.

Hongye Cao, Nuo Yan, Haoyuan Deng +5

National Key Laboratory for Novel Software Technology, Nanjing University, Nanjing 210093, China · Nanyang Technological University, Singapore · China Mobile NineVerse Artificial Intelligence Technology (Beijing) Co., Ltd., and Institute of Artificial Intelligence, NineVerse, Beijing 100032, China

Agentic RL RL for Language Model Reasoning RL Exploration RL for Tool Use

Jun 2, 2026cs.LG

Exploiting Verification-Generation Gap: Test-Time Reinforcement Learning with Confidence-Conditioned Verification

Test-time reinforcement learning has emerged as a promising paradigm for enhancing the complex reasoning abilities of large language models in a completely label-free manner. Despite existing studies focusing on Pass@1 performance, optimizing Pass@k remains under-explored yet critical in label-free settings, which measures generation coverage for sustained exploration. Optimizing Pass@k in label-free setting is highly non-trivial, as directly applying the Pass@k advantage designs effective for RLVR yields unsatisfactory performance. Through in-depth empirical analysis, we discover the root causes hindering performance: pseudo-label estimations for low-confidence samples have a high probability of being incorrect, while candidate answers for high-confidence samples suffer from severe diversity collapse. To overcome these hurdles, we propose TTRL-CoCoV (Test-Time Reinforcement Learning with Confidence-Conditioned Verification), a novel confidence-adaptive framework that expands Pass@k coverage and improves Pass@1 performance. Based on our key insight that verification capability generally leads generation capability, TTRL-CoCoV employs a confidence-conditioned mechanism: for high-confidence samples, it bootstraps verifier and applies an exploration-enhancing reward to prevent diversity collapse; for low-confidence samples, it delegates pseudo-label selection to the verifier to filter incorrect pseudo-labels; and for medium-confidence samples, it bypasses verification entirely. Extensive experiments demonstrate that TTRL-CoCoV outperforms the best competing methods across 6 widely-recognized benchmarks, achieves average absolute gains of +9.8% in Pass@1 and +18.7% in Pass@16 over TTRL, and even achieves absolute Pass@1 improvements of up to +5.0% across multiple reasoning benchmarks when compared against fully supervised RL methods. Our code repository: https://github.com/shanjf666/CoCoV.

Jiahui Li, Jianfeng Shan, Wenpei Chen +5

1Sun Yat-Sen University · University of Science and Technology of China · 3National University of Singapore

Reinforcement Learning LLM Answer Verification RL for Language Model Reasoning Test-Time RL Reinforcement Learning with Verifiable Rewards

Jun 2, 2026cs.DB

CAPER: Clause-Aligned Process Supervision for Text-to-SQL

Text-to-SQL systems are typically evaluated by query-level execution correctness, but this terminal signal provides little guidance about which intermediate SQL decision caused success or failure. Token-level dense supervision is also ill-suited: SQL tokens do not align with complete semantic decisions, can penalize execution-equivalent queries, and are difficult to label reliably at scale. We therefore propose CAPER, which automatically derives clause-level supervision via counterfactual intervention on the SQL abstract syntax tree, enabling root-cause error localization for reward modeling; the resulting data is used to train CAPER-9B, a lightweight Clause-PRM that provides clause-boundary feedback for policy optimization and candidate verification. Experiments on BIRD and Spider show that clause-aligned supervision not only improves execution accuracy, achieving up to a 15.3% relative EX improvement over GPT-5.4, but also strengthens failure-localization capability, reaching 84.53% accuracy and 90.60% MRR on held-out failures. Our project page is at https://github.com/banrichard/RL-NL2SQL.

Lujie Ban, Jiasheng Shi, Jinyang Li +3

The Chinese University of Hong Kong, Shenzhen · The University of Hong Kong · Northwestern Polytechnical University +1

Process Reward Models RL for Language Model Reasoning Text-to-SQL Process Supervision

Jun 2, 2026cs.LG

Right Makes Might: Aligning Verified Hidden States Empowers RL Reasoning

Reinforcement Learning from Verifiable Rewards (RLVR) has become the dominant approach for improving mathematical reasoning in large language models, yet current methods reduce each correct rollout to a single reward bit, ignoring the geometric structure shared among their hidden states. Investigating this structure, we find that at the anchor token (the position immediately before the answer marker), correct rollouts converge naturally because they must produce the same answer (cosine similarity ~0.84), yet each retains residual variance from its unique reasoning path. Encouraging full alignment at this point pushes the model to extract a unified "correct decision" representation, reducing sensitivity to which reasoning path was taken. Based on this observation, we propose Hidden-Align, an auxiliary loss function that aligns the last-layer hidden states of correct rollouts at the anchor token during RL training, with zero overhead in both training and inference. On eight mathematical reasoning benchmarks, Hidden-Align improves average pass@1 over the DAPO baseline by 3.8, 6.2, and 5.4 percentage points on Qwen3-1.7B, 4B, and 14B respectively, with consistent pass@k gains across all three scales, supported by ablations on loss type, anchor position, layer depth, and loss weight.

Ziyue Wang, Aomufei Yuan, Yongfu Zhu +10

1Peking University · 2JINGDONG · 3Shanghai Innovation Institute +2

LLM Alignment RL for Language Model Reasoning Reinforcement Learning with Verifiable Rewards LLM Mathematical Reasoning

Jun 2, 2026cs.LG

FGRPO: Federated GRPO with Adaptive Aggregation on Non-IID Data

Recent advances in language models have established reinforcement learning as the primary paradigm for eliciting self-correction and long-chain reasoning. While group relative policy optimization (GRPO) offers superior scalability by eliminating the critic network, deploying it on a central infrastructure entails collecting a large volume of data from distributed owners, which poses significant privacy risks. To address these concerns, we introduce federated GRPO (FGRPO), a framework designed to decentralize the fine-tuning of reasoning models across heterogeneous data owners. To effectively mitigate the instability caused by divergent reward scales across heterogeneous tasks, FGRPO incorporates an adaptive aggregation mechanism based on relative performance gain. By characterizing each client's improvement relative to its personalized historical baseline, the framework dynamically prioritizes effective learning trajectories regardless of local task difficulty. FGRPO ensures robust convergence on non-IID data while preserving data privacy.

Pengyu Chen, Shaowei Li, Kai Wang +4

School of Computer Science and Technology, Shandong University, Qingdao China · School of Mathematical Science, Peking University, China · School of Computer Science and Artificial Intelligence, Shanghai University of Finance and Economics, Shanghai, China +1

Federated Learning Aggregation Federated RL RL for Language Model Reasoning Group Relative Policy Optimization Non-IID Federated Learning Federated Learning

Jun 2, 2026cs.LG

ASymPO: Asymmetric-Scale Policy Optimization for Asynchronous LLM Post-Training Without Behavior Information

Asynchronous reinforcement learning can improve language-model post-training throughput by decoupling response generation from policy optimization, but stale responses introduce distribution drift. Standard behavior-corrected methods control this drift with behavior-policy probabilities, importance ratios, or clipping, which requires token-aligned, versioned, and numerically consistent behavior log-probabilities across rollout and learner systems. We ask whether asynchronous group-relative RL can instead be stabilized using only current-policy probabilities. We identify a scale-imbalance failure mode: when stale responses are evaluated under the current policy, positive and negative loss terms can appear at different negative log-probability scales, so zero-sum advantages no longer imply balanced loss contributions. We propose Asymmetric-Scale Policy Optimization (ASymPO), which normalizes each response's token loss by its current average token negative log-probability. ASymPO requires no behavior-policy probabilities, restores response-level zero-sum balance, and preserves a nonzero learning signal. We also introduce Scaled Policy Optimization (SPO), a fixed negative-scaling baseline, and evaluate both current-policy-only objectives in asynchronous mathematical reasoning post-training.

Zehua Liu, Yuxuan Yao, Xiaojin Fu +2

Huawei Technologies

Asynchronous RL Reinforcement Learning RL for Language Model Reasoning Group Relative Policy Optimization RL Post-Training Off-Policy RL

Jun 2, 2026cs.CL

Hint-Guided Diversified Policy Optimization for LLM Reasoning

Recent developments in Large Language Models (LLMs) have showcased impressive reasoning capabilities, with Reinforcement Learning with Verifiable Rewards (RLVR) being a promising enhancement strategy. However, existing reward mechanisms are constrained to the outcome-level correctness and lack explicit signals to guide the model to consider diverse solutions. In contrast, human problem solving typically involves evaluating multiple potential approaches and selecting the most reliable solution, a cognitive process that current RLVR frameworks do not explicitly incentivize. Inspired by this, we propose Hint-Guided Diversified Policy Optimization (HDPO), allowing the model to first list all potential candidate solution outlines as hints and then select the most reliable one for further reasoning. HDPO comprises two stages of Cold Start for Structured Reasoning and Hint-Guided Diversified Reinforcement Learning to incentivize the model to generate diverse and reliable solutions following the ``propose-select-think'' trajectory. Experimental results show that HDPO effectively boosts LLM reasoning and enhances the diversity of candidate solutions as well as the LLM's ability to identify reliable solutions.

Zhiyu Cao, Kaixin Wu, Mingjie Zhong +4

School of Computer Science and Technology, Soochow University, Suzhou, China · Ant Group, Hangzhou, China

Reasoning Diversity RL for Language Model Reasoning Reinforcement Learning with Verifiable Rewards LLM Reasoning

Jun 1, 2026cs.CL

Learning When to Translate for Multilingual Reasoning

Reasoning language models (RLMs) achieve strong performance on complex reasoning tasks, but still exhibit substantial multilingual reasoning gaps, largely due to language-understanding failures in non-English inputs. English translation can mitigate these failures by expressing non-English inputs in a form that RLMs can more reliably interpret, yet translating every input is unnecessary when the model can reason reliably from the original query. To address this challenge, we propose Luar, a Language Understanding Boundary-aware Reinforcement Learning framework that trains RLMs to selectively invoke translation when direct understanding is unreliable. Luar trains the model to choose between solving the original input directly and reasoning over its English translation, encouraging translation only when translator-augmented reasoning is expected to substantially outperform direct reasoning. Across multilingual reasoning benchmarks, Luar outperforms standard GRPO and other training-based baselines, with particularly large gains on low-resource languages. Further analysis shows that Luar avoids unnecessary translation in cases where direct reasoning is sufficient, while extending its translator-call behavior to unseen low-resource languages. Together, our work suggests a selective approach to multilingual reasoning: RLMs can learn to invoke translation only when their direct understanding is unreliable. The project will be made publicly available at https://github.com/deokhk/LUAR

Deokhyung Kang, Hyounghun Kim, Gary Geunbae Lee

1Graduate School of Artificial Intelligence · POSTECH, South Korea · Department of Computer Science & Engineering

Multilingual Language Models Cross-Lingual Reasoning in Language Models RL for Language Model Reasoning Low-Resource Language Processing

Jun 1, 2026cs.AI

Learning When Not to Act: Mitigating Tool Abuse in Agentic Reinforcement Learning

Agentic reinforcement learning can induce tool abuse, where models overuse external tools even for queries solvable by internal reasoning. Existing approaches mitigate this issue with uniform tool-use penalties or hard limits, which reduce tool frequency but may also suppress useful tool-assisted exploration. We propose EAPO, an Efficient Agentic Policy Optimization framework that learns selective tool use. EAPO introduces tool-free trajectories into each rollout group, applies difficulty-aware reward shaping to penalize redundant tool calls mainly on easier queries, and uses confidence-aware token reweighting to improve policy learning. Across nine mathematical and knowledge-intensive reasoning benchmarks, EAPO consistently improves the accuracy efficiency trade-off on Qwen2.5-3B, Qwen2.5-7B, and Llama3.1-8B. Compared with GRPO, EAPO improves average performance by 10.45%, 7.27%, and 9.69%, while reducing average tool calls by 18.33%, 18.33%, and 24.59%, respectively. These results show that agents can learn when not to use tools without compromising tool-integrated reasoning.

Liuji Chen, Dianxing Tang, Xing Shi +4

1NLPR, Institute of Automation, Chinese Academy of Sciences · 3Zhejiang University · 2ByteDance

Agentic RL RL for Language Model Reasoning Tool Use RL for Tool Use

Jun 1, 2026cs.LG

HMPO: Hybrid Median-length Policy Optimization for Chain-of-Thought Compression

Large language models achieve remarkable performance via extended chain-of-thought (CoT) reasoning, yet this lengthy process incurs substantial inference overhead. Existing CoT compression methods struggle with inflexible manual length budgets, computationally expensive multi-stage training pipelines, and fragile scalability restricted to small models. We propose HMPO (Hybrid Median-length Policy Optimization), a cost-effective, single-stage reinforcement learning framework. HMPO efficiently compresses CoT via three synergistic components: an adaptive median-based budget derived from successful rollouts to eliminate manual tuning, a cosine-decay token reward for smooth length penalization, and a multiplicative reward formulation that substantially mitigates trivial reward hacking by strictly prioritizing answer correctness. Trained exclusively on mathematical data, HMPO generalizes seamlessly across math, code, science, and instruction-following tasks. Extensive experiments scaling from 9B to 122B parameters across dense and Mixture-of-Experts (MoE) architectures demonstrate that HMPO achieves 19%--46% token compression with negligible accuracy degradation, all while drastically reducing training costs compared to existing multi-stage baselines.

Minghui Zheng, Hongxu Chen, Huimin Ren +8

Li Auto Inc.

LLM Compression RL for Language Model Reasoning LLM Inference Acceleration CoT Compression

Jun 1, 2026cs.AI

CAPF: Guiding Search-Agent Rollouts with Credit-Attenuated Privileged Feedback

Recent LLM search agents use reinforcement learning with verifiable rewards (RLVR) to learn search-augmented reasoning from outcome rewards. On hard problems, these agents rarely sample end-to-end successful rollouts, leaving outcome-only RLVR with few positive-reward trajectories. We argue that improving learning on such problems requires additional guidance during training, and RLVR already contains verifier-side information that can provide it. This information can identify errors or omissions in the agent's submitted answer and guide revision within the rollout. We propose a training-time mechanism called \textbf{Credit-Attenuated Privileged Feedback} (CAPF), which makes this verifier-side information available through a Privileged Feedback call during training. CAPF lets the policy revise zero-reward attempts into positive-reward repair trajectories and attenuates credit for the feedback call and earlier actions to accommodate deployment without this call. Empirical research demonstrates that CAPF improves Qwen3-4B's average exact-match score from 44.7% under outcome-only RLVR to 48.5% on seven open-domain QA benchmarks.

Bin Chen, Xinye Liao, Yiming Liu +2

RL for Language Model Reasoning Credit Assignment in RL Reinforcement Learning with Verifiable Rewards Learning Using Privileged Information Open-Domain QA

May 31, 2026cs.LG

OmniOPD: Logit-Free On-Policy Distillation via Speculative Verification

On-Policy Distillation (OPD) trains a student model on its own generative trajectories under dense token-level feedback from a stronger teacher, mitigating both the off-policy distribution shift of Supervised Fine-Tuning (SFT) and the sparse credit assignment of Reinforcement Learning (RL). However, standard OPD faces two coupled limitations. First, it requires direct access to the teacher's token-level logits, excluding a broad class of capable proprietary models from serving as teachers. Second, the token-level logit signal itself is brittle, depending on a narrow overlap of plausible next tokens between teacher and student, and prone to amplifying degenerate patterns such as repetition loops. In this paper, we introduce OmniOPD, a novel framework that addresses both limitations through a logit-free, chunk-level supervision signal. OmniOPD replaces deterministic logit matching with Monte Carlo rollouts that approximate the teacher's local preferences through a continuous semantic similarity metric over multi-token chunks, and concentrates this supervision via a peak-entropy scheduler that audits the student only at its high-uncertainty reasoning forks. A Dirichlet-Multinomial Bayesian prior and a base-model KL anchor further bound the variance of discrete sampling and prevent policy collapse across unaudited tokens. Across competitive benchmarks, OmniOPD surpasses the standard OPD approach by up to +28.64% on math, confirming that chunk-level semantic verification extracts a more reliable learning signal than token-level logit matching, whose high information density is offset by significant noise and brittleness. Furthermore, when paired with stronger black-box teachers such as Claude-4.5-Haiku and Gemini-2.5-Flash, OmniOPD achieves an additional +9.54% relative on math over its open-weight teacher counterpart, advancing the student past the performance of self-exploratory RL.

Yuhang Zhou, Lizhu Zhang, Yifan Wu +5

Meta AI

RL for Language Model Reasoning Language Model Distillation On-Policy Distillation Teacher-Student Learning

May 31, 2026cs.CL

Cross-lingual Self-Consistency for Multilingual Reasoning with Language Models

Despite expanding their multilingual coverage, the advanced reasoning capabilities of LLMs remain largely confined to a few high-resource languages like English. To address this, we propose an unsupervised Reinforcement Learning (RL) approach to enhance multilingual reasoning by enforcing cross-lingual self-consistency: the principle that a model should produce the same final answer for equivalent problems in different languages. Existing methods are limited by the scarcity of multilingual reasoning data and show weak generalization to unseen languages. Our approach requires neither gold answers nor parallel data, and it achieves average gains of up to 21.7% on MGSM across 10 languages. In addition, our method demonstrates strong generalization, with an 18.2% mean improvement on MGSM languages unseen during training, and up to 6.2% gain on 3 out-of-distribution benchmarks. These results show the potential of consistency-based methods to improve the multilingual capabilities of LLMs without requiring supervised data.

Ahmed Elhady, Eneko Agirre, Mikel Artetxe

HiTZ Center, University of the Basque Country (UPV/EHU) · Reka AI

Multilingual Language Models Cross-Lingual Reasoning in Language Models RL for Language Model Reasoning Cross-Lingual Consistency in Language Models Cross-Lingual Transfer

May 31, 2026cs.LG

RLVR without Ineffective Samples: Group Prioritized Off-Policy Optimization for LLM Reasoning

Reinforcement learning with verifiable rewards (RLVR) has emerged as a powerful paradigm for enhancing the reasoning capabilities of large language models (LLMs). However, its effectiveness is substantially hindered by the prevalence of ineffective training data: many sampled prompts yield response groups that are either entirely correct or entirely incorrect, resulting in zero-variance rewards and limited learning signals. Recent state-of-the-art methods address this issue through extensive LLM rollouts to filter ineffective samples, but at the cost of considerable computational overhead. Alternative approaches, including predictive sampling and trajectory replay, aim to improve data efficiency but often remain insufficient and may introduce additional issues such as systematic bias or suboptimal constraints. To address these limitations, we propose Group Prioritized Off-Policy Optimization (POPO), a simple yet effective framework that fully exploits effective training batches without additional rollout overhead. POPO comprises two key components: prioritized group replay and decoupled off-policy optimization. The former replaces ineffective on-policy groups with effective off-policy groups via a recency-based replay mechanism that jointly considers sample quality and the degree of off-policiness. To further mitigate the off-policy gap, POPO employs decoupled importance sampling to correct off-policy bias while maintaining stable policy updates under consistent trust-region constraints. Empirical evaluations across diverse reasoning tasks, including mathematics, planning, and visual geometry, demonstrate that POPO substantially accelerates RL finetuning and achieves strong reasoning performance with significantly fewer rollouts.

Yixiu Mao, Yun Qu, Qi Wang +2

Department of Automation, Tsinghua University

Reinforcement Learning RL for Language Model Reasoning Reinforcement Learning with Verifiable Rewards Sample-Efficient RL Off-Policy RL

May 31, 2026cs.CL

Deep Research as Rubric for Reinforcement Learning

Open-ended reasoning and long-form generation tasks lack reliable automatic verification signals for reward-based policy optimization. Rubrics offer a promising alternative, but existing approaches treat them as given artifacts -- either hand-crafted or prompt-generated -- and often miss the task-specific, knowledge-intensive dimensions that matter most, distorting the reward signal. Our key observation is that rubric construction is itself a research problem: identifying what makes a response correct or insightful requires discovering and synthesizing external knowledge. We propose Deep Research as Rubric (DR-rubric), a two-stage framework for constructing such rubrics. Stage I elicits domain facts, structural constraints, and failure modes through iterative multi-turn agentic search; Stage II distills this evidence into atomic, independently verifiable constraints for GRPO-based policy optimization. Because the model under training can serve as its own rubric generator, DR-rubric-8B supports bootstrap rubric generation without frontier-model assistance. We evaluate on 6 benchmarks spanning agentic research and expert reasoning. Experiments show that DR-Rubric achieves strong competitive performance with only 1K -- 3K training instances, where GPT-5-generated rubrics particularly benefit breadth coverage on agentic tasks, Gemini-generated rubrics yield the most balanced performance across agentic and expert reasoning tasks, and bootstrap rubrics exhibit a specialization-to-rebalancing evolution achieving the best overall performance at the third iteration. Results demonstrate that reframing rubric construction from static evaluation templates into an evidence-driven research process yields more scalable, fine-grained reward signals for open-ended tasks.

Wangyi Mei, Zhouhong Gu, Zhenhan Bai +9

Fudan University · Xiaohongshu Inc. · Beijing University of Posts and Telecommunications

Open-Ended Generation Reinforcement Learning RL for Language Model Reasoning Rubric-Based RL Reinforcement Learning with Verifiable Rewards Rubric-Guided Generation+1

May 30, 2026cs.LG

Enhancing LLM Metacognition via Cognitive Pairwise Training

Reinforcement learning with verifiable rewards (RLVR) has become central to LLM reasoning, but its outcome-level rewards can make models more willing to give confident answers when evidence or reasoning is unreliable. Existing SFT or RL methods mainly teach LLMs to refuse or express uncertainty at the response level, which can overfit abstention behavior rather than improve reasoning reliability. To address this limitation, we propose Cognitive Pairwise Training (CPT), a cognitive mid-training alignment stage that turns pairwise comparisons over reasoning traces into a reusable alignment signal. By learning to distinguish trustworthy from flawed reasoning, CPT encourages the model to internalize a reasoning-quality discrimination boundary rather than memorize surface refusal patterns. Across five model scales and three model families, CPT improves the reasoning--metacognition trade-off. At 14B, CPT+RL outperforms the standard SFT+RL pipeline by +2.2 math-average points and +5.2 abstention-F1 points. Further analyses show that CPT improves trace quality and exhibits strong robustness and scalability across evaluation and training settings. Code and models are released at https://github.com/Tsinghua-dhy/CPT.

Weitao Li, Hao Zhou, Xuanyu Lei +11

Dept. of Comp. Sci. & Tech., Institute for AI, Tsinghua University, Beijing, China · Institute for AI Industry Research (AIR), Tsinghua University, Beijing, China · WeChat AI, Tencent Inc., China

Pairwise Preference Learning RL for Language Model Reasoning LLM Post-Training Preference Learning Metacognition in Language Models

May 30, 2026cs.CL

Internalize the Temperature: On-Policy Self-Distillation as Policy Reheater for Reinforcement Learning

Reinforcement learning from verifiable rewards improves the reasoning ability of large language models, but often suffers from entropy collapse, in which increasingly concentrated policies reduce rollout diversity and useful learning signals. Existing remedies either constrain the RL objective (e.g., entropy regularization) or adjust sampling temperature during rollout collection, but these interventions remain external to the model parameters. We propose Temperature-Scaled On-Policy Self-Distillation (TS-OPSD), a lightweight policy reheating method that internalizes the exploratory effect of temperature into model parameters. Starting from an entropy-collapsed RL checkpoint, TS-OPSD constructs a self-teacher by applying high-temperature scaling to the model's own logits, then distills the resulting smoother distribution back into the student. This policy reheating requires no external teacher, privileged data, or additional inference cost. Experiments on Qwen3-4B-Base and Qwen3-8B-Base show that policy reheating yields a stronger initialization for continued RL than both standard continued RL and rollout-level temperature reheating. Further analyses show that TS-OPSD mainly reduces output sharpness while preserving intermediate representations, top candidate sets, and reasoning capability. These results suggest that entropy restoration can serve as a simple post-collapse intervention for extending reasoning-oriented RL.

Xuewei Yang, Jiachen Yu, Jie Wu +3

Tsinghua University

On-Policy Self-Distillation RL for Language Model Reasoning RL Post-Training

May 30, 2026cs.LG

The Paradox of Outcome Optimization: A Causal Information-Theoretic Bound on Reasoning Shortcuts in LLMs

Large Language Models (LLMs) aligned via outcome-based Reinforcement Learning (RL) frequently exhibit a critical failure mode: they achieve high performance on in-distribution benchmarks while demonstrating brittle reasoning capabilities on out-of-distribution (OOD) tasks. We term this phenomenon Reward-Induced Manifold Collapse. We establish a theoretical framework bridging Structural Causal Models (SCM) and the Information Bottleneck (IB) principle to explain this paradox. We define reasoning as a high-complexity causal process and shortcut learning as the exploitation of low-complexity spurious correlations. Under the implicit inductive bias of Stochastic Gradient Descent (SGD), models optimized for outcome rewards are biased toward shortcut solutions whenever the training distribution allows for a ``Markovian Screening'' of the true causal mechanism. We derive a new generalization bound based on Semantic Coverage Measure (

η

) rather than sample size, showing why data scaling on homogeneous distributions may fail to correct reasoning flaws. We also show that Process Reward Models (PRMs) function as Topological Filters, enforcing step-wise mutual information constraints that render the low-complexity shortcut manifold inadmissible. These results provide a mathematical grounding for the role of process supervision beyond simple credit assignment.

Zihan Chen, Yiming Zhang, Wenxiang Geng +2

1HFIPS, Chinese Academy of Sciences · University of Science and Technology of China

Process Reward Models RL for Language Model Reasoning OOD Generalization Causal Reasoning in Language Models

May 29, 2026cs.LG

PR2: Predictive Routing Replay for MoE-Based LLM Reinforcement Learning

Mixture of Experts (MoE) Large Language Models (LLMs) achieve strong performance at scale. However, reinforcement learning (RL) on MoE-based LLMs often suffers from training instability. A root cause is router drift, i.e., expert activations can change drastically across model updates and differ between disaggregated rollout and training phases, causing large rollout--training mismatch and unstable importance sampling weights in PPO-style RL algorithms. Routing replay mitigates this issue by freezing the replay route within each reasoning trajectory, but it ignores how the router evolves under off-policy updates and thus causes router staleness. To address this limitation, we propose Predictive Routing Replay (PR2), which augments each router with a lightweight evolution predictor that learns to anticipate short-horizon router evolution. During the rollout phase, we use the predictive routing distribution to apply top-

k

routing, enabling gradients to reach experts that are likely to become active after updates. During the training phase, we replay the resulting predicted route to retain consistency for stable importance estimation. Theoretical analysis and experiments support that PR2 reduces routing-induced mismatch, improves RL stability, and yields stronger performance across various reasoning benchmarks.

Daize Dong, Junlin Chen, Haolong Jia +9

♣Rutgers University · ♦AMD · ⋆MBZUAI

RL for Language Model Reasoning Expert Routing LLM Routing RL Post-Training Off-Policy RL

May 29, 2026cs.LG

ARCA: Adapter-Residual Credit Assignment When Token Signals Degenerate

Token-level credit assignment for language-model reinforcement learning is usually formulated as if the policy were fully trainable, while practical LLM-RL pipelines often rely on parameter-efficient fine-tuning, especially LoRA. We argue that this separation hides a structural failure mode. Under LoRA, the policy is restricted to a low-rank neighborhood of the reference model, so the per-token output-distribution differences used by common intrinsic credit signals, surprisal, entropy reduction, and policy divergence, can become degenerate after within-trajectory normalization, either approaching uniform weights or concentrating on a small set of task-agnostic positions. We formalize this behavior and propose measuring it directly with concentration diagnostics such as weight Gini and effective-token ratio. We then introduce \emph{Adapter-Residual Credit Assignment} (ARCA), a lightweight alternative that derives token salience from the adapter's own hidden-state residual,

\|h^{\text{adapted}}_t - h^{\text{base}}_t\|_2

. ARCA asks where the adapter actually changes the model, rather than where the output distribution appears uncertain or shifted, and requires no learned reward model, value head, or tree construction. In a compact MATH/Qwen3-1.7B GRPO sweep, ARCA exhibits the predicted non-degenerate middle-regime credit distribution under matched rollout budgets and remains competitive with rank-matched baselines.

Rodney Lafuente-Mercado

Scale AI

Token-Level Credit Assignment RL for Language Model Reasoning Low-Rank Adaptation Credit Assignment Fine-Grained Credit Assignment

May 29, 2026cs.CL

LongTraceRL: Learning Long-Context Reasoning from Search Agent Trajectories with Rubric Rewards

Long-context reasoning remains a central challenge for large language models, which often fail to locate and integrate key information in extensive distracting content. Reinforcement learning with verifiable rewards (RLVR) has shown promise for this task, yet existing methods are limited by low-confusability distractors and sparse, outcome-only reward signals that cannot supervise intermediate reasoning steps. To address these issues, we introduce \textsc{LongTraceRL}. For data construction, we generate multi-hop questions via knowledge graph random walks and leverage search agent trajectories to build \emph{tiered distractors}: documents the agent read but did not cite (high confusability) and documents that appeared in search results but were never opened (low confusability), producing training contexts that are far more challenging than those built by random sampling or one-shot search. For reward design, we propose a \emph{rubric reward} that uses the gold entities along each reasoning chain as fine-grained, entity-level process supervision. This rubric reward is applied only to responses with correct final answers (positive-only strategy), distinguishing the reasoning quality among correct responses and preventing reward hacking. Experiments on three reasoning LLMs (4B--30B) across five long-context benchmarks demonstrate that \textsc{LongTraceRL} consistently outperforms strong baselines and encourages comprehensive, evidence-grounded reasoning. Codes, datasets and models are available at https://github.com/THU-KEG/LongTraceRL.

Nianyi Lin, Jiajie Zhang, Lei Hou +1

Tsinghua University

LLM Grounding RL for Language Model Reasoning Reinforcement Learning with Verifiable Rewards Process Supervision Long-Context Inference

May 29, 2026cs.CL

Are Full Rollouts Necessary for On-Policy Distillation?

On-policy distillation (OPD) provides dense teacher feedback along student-generated rollouts rather than fixed teacher traces and has emerged as a promising post-training paradigm. However, standard OPD typically generates full rollouts during training, which is computationally expensive and may expose the student to unreliable teacher feedback at late rollout positions, especially during early training. We identify the rollout horizon as a key bottleneck in OPD that substantially impacts training efficiency. Unlike Reinforcement Learning with Verifiable Rewards (RLVR), OPD does not require a final answer reward to provide learning signals. Therefore, full rollouts may not always be necessary for OPD. Motivated by this insight, we propose two simple horizon-control strategies: Progressive OPD (POPD), which gradually expands the rollout horizon during training, and Truncated OPD (TOPD), which permanently performs distillation on reliable truncated rollouts. Experiments on mathematical reasoning show that POPD improves the training efficiency of OPD by up to 3

\times

, while TOPD matches OPD performance using only 10% of the rollout horizon, leading to substantial wall-clock and memory reductions. These results demonstrate that controlling the rollout horizon offers a simple and practical path to more efficient OPD.

Yaocheng Zhang, Jiajun Chai, Yuqian Fu +7

Institute of Automation, Chinese Academy of Sciences · School of Advanced Interdisciplinary Sciences, University of Chinese Academy of Sciences · Meituan +1

RL for Language Model Reasoning Language Model Distillation On-Policy Distillation

May 29, 2026cs.AI

CAST: Non-Privileged Clipped Asymmetric Self-Teaching with Advantage Flipping for GRPO

Reinforcement learning with verifiable rewards (RLVR), especially Group Relative Policy Optimization (GRPO), has been widely used to improve reasoning in large language models. However, outcome-level rewards provide only sparse supervision, and group-relative advantages vanish when all sampled trajectories for a prompt are either correct or incorrect. On-Policy Self-Distillation (OPSD) offers dense token-level guidance, but its token preferences are not necessarily aligned with trajectory correctness; empirical diagnostics show that OPSD signals behave differently on correct and incorrect rollouts, with teacher-positive and teacher-negative gap signals exhibiting different noise profiles. These diagnostics are conducted under an OPSD-style privileged teacher context for analysis only, whereas CAST training uses answer-free self-teacher scoring.Motivated by these observations, this work proposes CAST, an answer-free self-distillation method for GRPO-style RLVR. CAST keeps the verifier-grounded GRPO objective, but uses a stop-gradient self-teacher to shape token-level advantages according to trajectory correctness. Unlike prior self-distilled RLVR methods, CAST does not require reference-solution-conditioned teacher scoring, keeps the self-teacher log-probability gap active throughout training, and applies bidirectional local advantage sign reversal: teacher-negative tokens in correct trajectories can receive negative token-level advantages, while teacher-positive tokens in incorrect trajectories can receive bounded positive local advantages. For zero-variance all-correct and all-wrong groups, CAST assigns bounded sign-constrained base advantages, so these otherwise zero-gradient groups can contribute verifier-signed token feedback. Experiments on mathematical reasoning show that CAST improves RLVR training while retaining a lightweight, verifier-grounded trajectory-level objective.

Yang Li, Gongle Xue, Yijia Guo +3

School of Software and Microelectronics, Peking University, Beijing · School of Artificial Intelligence, Peking University, Beijing · School of Computer Science, Peking University, Beijing +1

RL for Language Model Reasoning Group Relative Policy Optimization Reinforcement Learning with Verifiable Rewards On-Policy Distillation Teacher-Student Learning

May 29, 2026cs.LG

EchoRL: Reinforcement Learning via Rollout Echoing

Reinforcement Learning with Verifiable Rewards is an effective route for post-training to strengthen the reasoning capability of large language models. However, as training proceeds, the learning signal can collapse thus makes the training gain become marginal and ineffective. Specifically, a growing fraction of prompts' rollouts become advantage-degenerated: all the self-generated rollouts show verified-success, making the standard deviation over their rewards be zero; accordingly each rollout's advantage becomes degenerated (zero) as well. Given such rollouts' advantages, the policy-gradient for model optimization eventually vanishes, capping the training performance. We argue that some of these rollouts still contain valuable learning signals but unfortunately omitted with the existing RLVR methods. In this paper, inspired through analyzing the entropy pattern behind golden trajectories produced by external expert models, we propose EchoRL for better exploiting the advantage-degenerated rollouts to further improve the training performance. EchoRL is a lightweight module that first identifies an EchoClip from verified-success rollouts based on their step-level entropy values, and then feeds this clip back as an auxiliary supervision signal in the RL objective. Extensive experiments across 10 benchmarks, 5 LLM backbones, and 4 popular RLVR post-training methods demonstrate that EchoRL consistently improves RLVR post-training with minimal overhead.

Jinhe Bi, Aniri, Minglai Yang +9

Huawei Heisenberg Research Center · LMU Munich · Munich Center for Machine Learning +5

Reinforcement Learning RL for Language Model Reasoning Reinforcement Learning with Verifiable Rewards RL Post-Training

May 29, 2026cs.CL

AdaptR1: Reinforcement Learning Based Adaptive Interleaved Thinking in Multi-hop Question Answering

Large Language Models (LLMs) have achieved remarkable performance in complex reasoning tasks through Chain-of-Thought (CoT) prompting. However, this approach often leads to ``over-thinking,'' where models generate unnecessarily long reasoning traces for simple queries and incur avoidable inference cost. While recent work has explored adaptive reasoning, existing methods typically make a single query-level decision about whether to reason. This overlooks the dynamic nature of multi-step tasks, where the need for explicit reasoning varies across intermediate stages. To address this limitation, we introduce AdaptR1, a Reinforcement Learning (RL) based framework for adaptive interleaved thinking in multi-hop Question Answering (QA). Unlike previous approaches that require Supervised Fine-Tuning (SFT) for cold-start initialization, AdaptR1 uses a fully RL-based strategy with a quality-gated efficiency reward to dynamically allocate reasoning budgets at each step. Under the Graph-R1 setting, AdaptR1 reduces average think tokens by 69.71%, with a 90.35% reduction on HotpotQA, while maintaining performance comparable to or better than standard baselines. Furthermore, our analysis reveals that overthinking in multi-hop reasoning is not uniformly distributed but occurs predominantly during the initial planning stages, highlighting the effectiveness of step-wise adaptive budget allocation.

Yuxin Wang, Jiahao Lu, Qifeng Wu +5

1Computer Science, Fudan University · Institute of Modern Languages and Linguistics, Fudan University · 3Shanghai Innovation Institute +1

Multi-Hop QA RL for Language Model Reasoning LLM Inference Acceleration Large Reasoning Models

May 29, 2026cs.LG

Sample-Efficient Post-Training for LEGO Spatial-Physics Reasoning

LLM-based LEGO assembly generation requires both semantic grounding and physical feasibility. We identify a data-induced failure mode, PhysHack, in which the assemblies satisfy physical-validity constraints while producing structures that are geometrically misaligned, semantically inconsistent, or poorly calibrated. To address this challenge, we propose a model-based data selection approach that uses only a small fraction of the training data while improving physically grounded LEGO assembly generation. Building on the selected trajectories, we introduce PVPO, a sample-efficient reinforcement learning method that couples physical feasibility with voxel-space geometric rewards. Our results show that physical validity alone is an insufficient proxy for reliable physical reasoning: models can learn to generate valid structures without preserving semantic or geometric fidelity. Experiments across model backbones and test-time scaling settings demonstrate that PVPO improves structural and semantic alignment, physical validity, structural stability, and calibration, while reducing reliance on extensive post-hoc rejection sampling. In particular, results on calibration show that PVPO mitigates PhysHack by making test-time selection more predictive of semantic and structural quality.

Yuhuan Yuan, Zhouliang Yu, Minghao Liu +2

HKUST(GZ) · CUHK · ZODA

Physical Reasoning Visual Spatial Reasoning RL for Language Model Reasoning Data Selection Training Data Selection Sample-Efficient RL