arXiv AI
Jun 11
ProcessThinker: Enhancing Multi-modal Large Language Models Reasoning via Rollout-based Process Reward
arXiv:2606. 11209v1 Announce Type: cross Abstract: Visual question answering increasingly requires multi-step reasoning.
Giving Credit Where It's Due: Redundancy-Aware Learning for Efficient Reasoning
The paper introduces RECAP, a redundancy-aware credit assignment method that improves reasoning efficiency in large language models by assigning credit to each reasoning step based on its downstream role and contribution to the correct answer. RECAP uses a semantic dependency graph to measure structural responsibility and evaluates step efficacy via changes in gold-answer log-likelihood, enabling step-specific updates without requiring a separate reward model or concise trajectories. Experiments on two 7B models across four mathematical reasoning benchmarks show that RECAP enhances the accuracy-efficiency trade-off, boosting pass@1 by 2.0–3.7 percentage points while cutting reasoning tokens by 8–31% compared to GRPO.
Related stories
arXiv AI
Jun 9
Correct Is Not Enough: Training Reasoning Planners with Executor-Grounded Rewards
arXiv:2605. 03862v4 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards has become a common way to improve explicit reasoning in large language models, but final-answer correctness alone does not reveal whether the reasoning trace is faithful, reliable, or useful to the model that consumes it.
arXiv Machine Learning
Jun 10
How Does Reasoning Flow? Tracing Attention-Induced Information Flow for Targeted RL in LLMs
arXiv:2606. 10646v1 Announce Type: new Abstract: Token-level credit assignment remains a key obstacle for reinforcement learning (RL) in large language models (LLMs), where RL recipes typically treat all tokens equally, failing to distinguish decisive reasoning steps from routine formatting or fluent filler.
arXiv AI
Oct 2
Interpreting Reasoning of Large Language Models via Partial Information Decomposition
arXiv:2610.00571v1 Announce Type: cross Abstract: Large reasoning models (LRMs) have achieved substantial improvements in solving complex mathematical problems, but often produce lengthy, repetitive,...
arXiv AI
Jun 30
Reconsidering Overthinking: Penalizing Internal and External Redundancy in CoT Reasoning
arXiv:2508. 02178v3 Announce Type: replace Abstract: Large reasoning models (LRMs) often exhibit overthinking, producing verbose Chain-of-Thought (CoT) traces that increase inference cost and obscure the underlying reasoning process.
arXiv Computation and Language
6d ago
Enhancing Language Model Reasoning with Structured Multi-Level Modeling
The paper introduces Multi-Level Reasoning (MLR), a two-level stochastic framework that separates long chain-of-thought generation into a high-level planner producing structured step descriptors and a low-level executor generating detailed reasoning. MLR replaces single-policy reinforcement learning with an iterative Step‑DPO preference optimization that uses Twisted Sequential Monte Carlo for scalable, stepwise supervision, improving training stability and accuracy. Experiments on math, science, and logic benchmarks show that MLR outperforms both supervised fine‑tuning and strong RL baselines while maintaining robustness over extended reasoning horizons.
arXiv AI
Sep 28
Stepwise Intrinsic Rewards for Reasoning in Large Language Models
The paper introduces Stepwise Marginal Information Gain (MIG), an intrinsic process reward that evaluates how each reasoning step of a large language model (LLM) or vision-language model (VLM) improves the likelihood of the reference answer. MIG rewards only new likelihood maxima, preventing duplicate credit, and is combined with outcome, format, and self‑distillation objectives to guide training. Experiments on eight benchmarks show that this method outperforms outcome‑only reinforcement learning and improves accuracy by up to 4.8 points over binary‑reward training, including a 12.6‑point gain on MathVerse and a 12.9‑point advantage on vision‑language transfer at 7B parameters.
arXiv Computation and Language
Sep 1
Every Token Leaves a Ripple in the Stream of Thought: Eliciting Model-Internal Token Saliency for Chain-of-Thought Compression
arXiv:2608.31066v1 Announce Type: new Abstract: Chain-of-thought (CoT) reasoning improves multi-step problem solving, but long reasoning traces inflate inference cost. Token-level CoT compression red...
arXiv AI
Jun 2
Process Reward Agents for Steering Knowledge-Intensive Reasoning
arXiv:2604. 09482v2 Announce Type: replace Abstract: Reasoning in knowledge-intensive domains remains challenging as intermediate steps are often not locally verifiable: unlike math or code, evaluating step correctness may require synthesizing clues across large external knowledge sources.
arXiv AI
Oct 21
TrajEvolve: Trajectory Evolution for Reinforcement Learning with Hindsight Credit Assignment
TrajEvolve introduces a trajectory evolution paradigm with hindsight credit assignment to improve reinforcement learning for large language models on long-horizon tasks. It refines low-quality steps in historical trajectories to expose the policy to high-quality compositions that are rarely sampled, and reformulates step-level continuous credit assignment as a binary classification problem to identify redundant steps. Experiments on two datasets show that this approach outperforms existing methods.
Hugging Face Trending Papers
Jun 3
GRAIL: Gradient-Reweighted Advantages for Reinforcement Learning with Verifiable Rewards
Reinforcement learning with verifiable rewards (e. g.
arXiv AI
Oct 2
Dependency-Aware Reward Shaping for Agentic Reinforcement Learning
The paper introduces Dependency‑Aware Reward Shaping (DARS), a method that assigns step‑level credit in reinforcement learning by modeling task progress as a graph of predicates with prerequisite relations. Annotators mark each step’s effect on predicates, and DARS discounts verified predicates based on distance from broken prerequisites while preserving independent ones, converting these annotations into signed per‑step rewards. Experiments on five task families with models ranging from 1.5B to 8B show that DARS improves success rates by up to 10 points over GiGPO, boosts WebShop and Search‑R1 QA scores, complements AEPO on AIME24/25, and outperforms OmniOPD in tool‑free reasoning, with ablations confirming the contribution of step‑level credit, dependency attenuation, and graph topology.