pith. sign in

arxiv: 2505.13445 · v1 · pith:5METPRLAnew · submitted 2025-05-19 · 💻 cs.AI · cs.CL

Trust, But Verify: A Self-Verification Approach to Reinforcement Learning with Verifiable Rewards

classification 💻 cs.AI cs.CL
keywords self-verificationrisereasoningmodelsrewardsverifiableadvantageslearning
0
0 comments X
read the original abstract

Large Language Models (LLMs) show great promise in complex reasoning, with Reinforcement Learning with Verifiable Rewards (RLVR) being a key enhancement strategy. However, a prevalent issue is ``superficial self-reflection'', where models fail to robustly verify their own outputs. We introduce RISE (Reinforcing Reasoning with Self-Verification), a novel online RL framework designed to tackle this. RISE explicitly and simultaneously trains an LLM to improve both its problem-solving and self-verification abilities within a single, integrated RL process. The core mechanism involves leveraging verifiable rewards from an outcome verifier to provide on-the-fly feedback for both solution generation and self-verification tasks. In each iteration, the model generates solutions, then critiques its own on-policy generated solutions, with both trajectories contributing to the policy update. Extensive experiments on diverse mathematical reasoning benchmarks show that RISE consistently improves model's problem-solving accuracy while concurrently fostering strong self-verification skills. Our analyses highlight the advantages of online verification and the benefits of increased verification compute. Additionally, RISE models exhibit more frequent and accurate self-verification behaviors during reasoning. These advantages reinforce RISE as a flexible and effective path towards developing more robust and self-aware reasoners.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Learning from Self-Debate: Preparing Reasoning Models for Multi-Agent Debate

    cs.CL 2026-01 unverdicted novelty 7.0

    SDRL trains LLMs via self-generated multi-path debates and joint optimization of standalone plus debate-conditioned responses to boost both single-model reasoning and multi-agent debate performance.

  2. STRIDE: Learnable Stepwise Language Feedback for LLM Reasoning

    cs.LG 2026-05 unverdicted novelty 6.0

    STRIDE co-trains generator and verifier on outcome rewards alone to deliver learnable stepwise language feedback that redirects LLM reasoning trajectories and outperforms scalar-reward baselines.

  3. ICRL: Learning to Internalize Self-Critique with Reinforcement Learning

    cs.AI 2026-05 unverdicted novelty 6.0

    ICRL uses joint RL training of solver and critic with distribution-calibration re-weighting and role-wise advantage estimation to internalize critique into unassisted LLM performance, yielding 6.4-point gains on agent...

  4. Selective Off-Policy Reference Tuning with Plan Guidance

    cs.AI 2026-05 unverdicted novelty 6.0

    SORT turns all-wrong prompts into selective learning signals by weighting tokens more predictable under plan guidance from reference solutions, improving over GRPO on reasoning benchmarks especially for weaker models.

  5. Verification Mirage: Mapping the Reliability Boundary of Self-Verification in Medical VQA

    cs.CV 2026-05 unverdicted novelty 6.0

    Self-verification in medical VQA creates a verification mirage where verifiers exhibit high error and agreement bias on wrong answers, with reliability strongly conditioned on task type.

  6. Selective Off-Policy Reference Tuning with Plan Guidance

    cs.AI 2026-05 unverdicted novelty 5.0

    SORT converts all-failed reasoning prompts into selective, structure-aware training signals by weighting tokens according to how much a reference-derived plan increases their probability.

  7. Learning from Less: Measuring the Effectiveness of RLVR in Low Data and Compute Regimes

    cs.AI 2026-04 unverdicted novelty 5.0

    Mixed-complexity procedural datasets provide up to 5x sample efficiency for RLVR on small models in low-data regimes, with low-to-high complexity generalization observed across counting, graph, and spatial tasks.