STRIDE: Learnable Stepwise Language Feedback for LLM Reasoning

• URL: http://arxiv.org/abs/2605.18851v1
• Date: Wed, 13 May 2026 11:04:31 GMT
• Title: STRIDE: Learnable Stepwise Language Feedback for LLM Reasoning
• Abstract summary: We propose language-driven stepwise trajectory redirection, termed as STRIDE.<n>We co-train a generator and a generative verifier using only outcome-based rewards, eliminating external annotations.<n> Experiments on diverse reasoning benchmarks show that STRIDE significantly outperforms state-of-the-art baselines.
• Score: 80.78140312980484
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advances in Reinforcement Learning (RL) have underscored its potential for incentivizing reasoning capabilities of Large Language Models (LLMs). However, existing step-level efforts suffer from costly annotations that limit domain coverage, while scalar scores further impose an information bottleneck, offering insufficient semantic bandwidth to improve intermediate decisions. Alternative language-critique approaches, which rely on frozen or external critics, provide richer textual feedback but lack the scalability needed for sustained policy improvement. In this work, we propose language-driven stepwise trajectory redirection, termed as STRIDE, a novel training framework that shifts process supervision from scalar rewards to learnable stepwise language feedback. Specifically, we co-train a generator and a generative verifier using only outcome-based rewards, eliminating external annotations, while delivering sustained policy improvement through jointly aligned verifier training. The verifier's stepwise language critiques explicitly localize and explain failures, enabling the generator to redirect reasoning trajectories at intermediate steps toward alternative decisions. The trajectory redirection design guarantees harmless policy improvement, even under noisy or suboptimal verifier feedback. Experiments on diverse reasoning benchmarks show that STRIDE significantly outperforms state-of-the-art baselines, as well as achieving breakthroughs on zero-pass-rate problems where scalar methods yield no learning signal in our ablation studies, demonstrating the effectiveness of learnable stepwise language feedback for enhancing LLM reasoning.

Related papers

• Inference-Time Steering for Cross-Lingual Factual Consistency in LLMs [51.56484100374058]
  Large Language Models (LLMs) demonstrate remarkable multilingual fluency, but their internal knowledge representations remain disproportionately biased toward high-resource languages.<n>We investigate whether these biases can be mitigated at inference time, forcing an English-prompted model to answer as if it were queried in target languages.<n>Experiments on Gemma 3 12B Instruct reveal persona prompting to be the strongest overall intervention, balancing efficacy, safety, and out-of-domain generalization.
  arXiv  Detail & Related papers  (2026-07-21T16:15:05Z)
• Language-Critique Imitation Learning from Suboptimal Demonstrations [23.463660165952632]
  We propose a language-critique framework for imitation learning from suboptimal demonstrations.<n>Our method first constructs language labels from demonstrations that explicitly describe current progress.<n>We then introduce a language-critique loss that directly trains policies using these structured signals.
  arXiv  Detail & Related papers  (2026-07-01T17:57:22Z)
• Learning from Language Feedback via Variational Policy Distillation [47.05700353631563]
  Variational Policy Distillation (VPD) is a framework that formalizes learning from language feedback as a Variational Expectation-Maximization problem.<n>VPD consistently outperforms standard RLVR and existing self-distillation baselines.
  arXiv  Detail & Related papers  (2026-05-14T17:27:34Z)
• Moira: Language-driven Hierarchical Reinforcement Learning for Pair Trading [34.923407379006626]
  Many sequential decision-making problems exhibit hierarchical structure, where high-level semantic choices constrain downstream actions and feedback is delayed and ambiguous.<n>We study this challenge through pair trading, a domain that naturally combines long-horizon semantic reasoning for asset pair selection with short-horizon execution under partial observability.<n>We formulate pair trading as a hierarchical reinforcement learning problem and propose a language-driven optimization framework in which both high-level and low-level policies are parameterized by large language models (LLMs) and optimized exclusively through prompt updates.
  arXiv  Detail & Related papers  (2026-05-03T16:37:52Z)
• Let's Reward Step-by-Step: Step-Aware Contrastive Alignment for Vision-Language Navigation in Continuous Environments [38.97818584066075]
  Vision-Language Navigation in Continuous Environments (VLN-CE) requires agents to learn complex reasoning from long-horizon human interactions.<n>Current training paradigms struggle to balance generalization capability, error recovery and training stability.<n>We introduce Step-Aware Contrastive Alignment (SACA), a framework designed to extract dense supervision from imperfect trajectories.
  arXiv  Detail & Related papers  (2026-03-10T14:45:50Z)
• LANPO: Bootstrapping Language and Numerical Feedback for Reinforcement Learning in LLMs [73.27182315028021]
  LANPO is a framework that cleanly separates the roles of feedback: language guides exploration, while numerical rewards drive optimization.<n>Our work provides a robust method for integrating historical experiences into the LLM RL loop, creating more effective and data-efficient learning agents.
  arXiv  Detail & Related papers  (2025-10-18T15:51:19Z)
• Learning a Dense Reasoning Reward Model from Expert Demonstration via Inverse Reinforcement Learning [50.20267980386502]
  We learn a dense, token-level reward model for process supervision directly from expert demonstrations.<n>The learned reasoning reward serves two complementary roles: (i) it provides step-level feedback to optimise a reasoning policy during training; and (ii) it functions at inference as a critic to rerank sampled traces under fixed compute budgets.
  arXiv  Detail & Related papers  (2025-10-02T09:55:26Z)
• Dancing with Critiques: Enhancing LLM Reasoning with Stepwise Natural Language Self-Critique [66.94905631175209]
  We propose a novel inference-time scaling approach -- stepwise natural language self-critique (PANEL)<n>It employs self-generated natural language critiques as feedback to guide the step-level search process.<n>This approach bypasses the need for task-specific verifiers and the associated training overhead.
  arXiv  Detail & Related papers  (2025-03-21T17:59:55Z)
• Zero-shot Model-based Reinforcement Learning using Large Language Models [12.930241182192988]
  We investigate how pre-trained Large Language Models can be leveraged to predict in context the dynamics of continuous Markov decision processes.<n>We present proof-of-concept applications in two reinforcement learning settings: model-based policy evaluation and data-augmented off-policy reinforcement learning.
  arXiv  Detail & Related papers  (2024-10-15T15:46:53Z)
• Unconditional Truthfulness: Learning Unconditional Uncertainty of Large Language Models [104.55763564037831]
  We train a regression model that leverages attention maps, probabilities on the current generation step, and recurrently computed uncertainty scores from previously generated tokens.<n>Our evaluation shows that the proposed method is highly effective for selective generation, achieving substantial improvements over rivaling unsupervised and supervised approaches.
  arXiv  Detail & Related papers  (2024-08-20T09:42:26Z)
• Gaining Wisdom from Setbacks: Aligning Large Language Models via Mistake Analysis [127.85293480405082]
  The rapid development of large language models (LLMs) has not only provided numerous opportunities but also presented significant challenges. Existing alignment methods usually direct LLMs toward the favorable outcomes by utilizing human-annotated, flawless instruction-response pairs. This study proposes a novel alignment technique based on mistake analysis, which deliberately exposes LLMs to erroneous content to learn the reasons for mistakes and how to avoid them.
  arXiv  Detail & Related papers  (2023-10-16T14:59:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.