Momentum for Reasoning: Dense Intrinsic Signals in Policy Optimization

• URL: http://arxiv.org/abs/2606.08815v1
• Date: Sun, 07 Jun 2026 20:08:36 GMT
• Title: Momentum for Reasoning: Dense Intrinsic Signals in Policy Optimization
• Abstract summary: Reinforcement learning with verifiable rewards (RLVR) has emerged as a powerful paradigm for eliciting long-chain reasoning in large language models.<n>We propose ISPO (Intrinsic Signal Policy Optimization), which combines a sequence-level signal measuring how informative the thinking trajectory is for the final answer, with a token-level directional reward whose hallucinated-certainty hinge penalizes confidently-wrong predictions at critical decision tokens.<n>ISPO consistently outperforms competitive baselines, with the largest gains on the hardest benchmarks where zero-advantage collapse is most frequent, and training-dynamics diagnostics confirm that both failure modes are decreased.
• Score: 31.547614109943158
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning with verifiable rewards (RLVR) has emerged as a powerful paradigm for eliciting long-chain reasoning in large language models. However, existing methods based on Group Relative Policy Optimization (GRPO) rely on a binary outcome reward, which induces two structural failure modes: Zero-Advantage Collapse, in which all rollouts in a group share the same outcome and the gradient vanishes, and Hallucinated Certainty, in which the model becomes increasingly confident on incorrect rollouts late in training. We address both modes by densifying the reward with intrinsic signals computed entirely from the policy's own conditional probabilities, and propose ISPO (Intrinsic Signal Policy Optimization, which combines a sequence-level signal measuring how informative the thinking trajectory is for the final answer, with a token-level directional reward whose hallucinated-certainty hinge penalizes confidently-wrong predictions at critical decision tokens. Across three base models and five mathematical reasoning benchmarks, ISPO consistently outperforms competitive baselines, with the largest gains on the hardest benchmarks where zero-advantage collapse is most frequent, and training-dynamics diagnostics confirm that both failure modes are decreased.

Related papers

• Joint Training of Multi-Token Prediction in Reinforcement Learning via Optimal Coefficient Calibration [61.46060073417047]
  Multi-Token Prediction (MTP) has been a widely adopted module in pretraining.<n>We show that the per-step effect of MTP on the RL objective can be decomposed into two terms: a first-order correlation and a second-order penalty.<n>We propose an adaptive scheme that tracks the optimal coefficient online via a log-probability proxy at negligible cost.
  arXiv  Detail & Related papers  (2026-05-27T09:07:06Z)
• Learning from Failures: Correction-Oriented Policy Optimization with Verifiable Rewards [73.44333771806282]
  Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as an effective paradigm for improving the reasoning capabilities of large language models.<n>We propose Correction-Oriented Policy Optimization (CIPO), a simple and effective extension to RLVR that converts on-policy failed trajectories into correction-oriented supervision.<n>CIPO improves learning effectiveness while explicitly enhancing the model's ability to correct its own errors.
  arXiv  Detail & Related papers  (2026-05-14T08:22:21Z)
• Multimodal Learning on Low-Quality Data with Conformal Predictive Self-Calibration [72.0672328514289]
  Multimodal learning often grapples with the challenge of low-quality data.<n>We propose a unified framework, termed Conformal Predictive Self-Calibration.<n>Our framework consistently outperforms existing state-of-the-art methods.
  arXiv  Detail & Related papers  (2026-05-05T14:48:52Z)
• Too Correct to Learn: Reinforcement Learning on Saturated Reasoning Data [55.84428098924793]
  We propose Constrained Uniform Top-K Sampling (CUTS), a parameter-free decoding strategy enforcing structure-preserving exploration.<n>We integrate this into Mixed-CUTS, a training framework synergizing exploitative and exploratory rollouts to amplify intra-group advantage variance.<n> Notably, Mixed-CUTS improves Pass@1 accuracy on the challenging AIME25 benchmark by up to 15.1% over standard GRPO.
  arXiv  Detail & Related papers  (2026-04-20T16:43:28Z)
• ERPO: Token-Level Entropy-Regulated Policy Optimization for Large Reasoning Models [3.463914032107119]
  We show that Group Relative Policy Optimization (GRPO) typically assigns a uniform, sequence-level advantage to all tokens.<n>We propose Entropy-Regulated Policy Optimization (ERPO), which transitions the optimization focus from coarse sequences to fine-grained token dynamics.
  arXiv  Detail & Related papers  (2026-03-30T09:20:25Z)
• Unifying Stable Optimization and Reference Regularization in RLHF [64.16830602324345]
  This paper introduces a unified regularization approach that balances objectives of preventing reward hacking and maintaining stable policy updates.<n>Our simple yet principled alignment objective yields a weighted supervised fine-tuning loss with a superior trade-off, which demonstrably improves both alignment results and implementation complexity.
  arXiv  Detail & Related papers  (2026-02-12T03:31:19Z)
• Difficulty-Estimated Policy Optimization [38.86673795561421]
  We propose Difficulty-Estimated Policy Optimization (DEPO), a novel framework designed to optimize the efficiency and robustness of reasoning alignment.<n>Our approach significantly lowers the computational barrier for training high-performance reasoning models, offering a more sustainable path for reasoning scaling.
  arXiv  Detail & Related papers  (2026-02-06T04:12:23Z)
• Back to Basics: Revisiting Exploration in Reinforcement Learning for LLM Reasoning via Generative Probabilities [10.235183326885794]
  Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as an indispensable paradigm for enhancing reasoning in Large Language Models (LLMs)<n>We analyze this issue from the perspective of sampling probability dynamics, identifying that the standard objective disproportionately reinforces the highest-likelihood paths.<n>We propose a novel Advantage Re-weighting Mechanism (ARM) designed to equilibrate the confidence levels across all correct responses.
  arXiv  Detail & Related papers  (2026-02-05T04:06:55Z)
• Expected Return Causes Outcome-Level Mode Collapse in Reinforcement Learning and How to Fix It with Inverse Probability Scaling [2.8825582215429186]
  We show that outcome-level mode collapse is a structural consequence of the expected-return itself.<n>We propose a minimal correction: inverse probability scaling, which removes outcome-frequency from the learning signal.
  arXiv  Detail & Related papers  (2026-01-29T13:03:33Z)
• Outcome-Grounded Advantage Reshaping for Fine-Grained Credit Assignment in Mathematical Reasoning [60.00161035836637]
  Group Relative Policy Optimization has emerged as a promising critic-free reinforcement learning paradigm for reasoning tasks.<n>We introduce Outcome-grounded Advantage Reshaping (OAR), a fine-grained credit assignment mechanism that redistributes advantages based on how much each token influences the model's final answer.<n>OAR-G achieves comparable gains with negligible computational overhead, both significantly outperforming a strong GRPO baseline.
  arXiv  Detail & Related papers  (2026-01-12T10:48:02Z)
• Conditional Advantage Estimation for Reinforcement Learning in Large Reasoning Models [50.84995206660551]
  We introduce Conditional advANtage estimatiON (CANON) to amplify the impact of a target metric without presuming its direction.<n>CANON based on entropy consistently outperforms prior methods on both math reasoning and high-complexity logic tasks.
  arXiv  Detail & Related papers  (2025-09-28T16:33:07Z)

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.