Pith. sign in

REVIEW 8 cited by

Boosting LLM Reasoning via Spontaneous Self-Correction

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2506.06923 v1 pith:A2LFHJZR submitted 2025-06-07 cs.AI

classification cs.AI
keywords reasoningself-correctionspocmodelspontaneousaccuracyapproachapproaches
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

While large language models (LLMs) have demonstrated remarkable success on a broad range of tasks, math reasoning remains a challenging one. One of the approaches for improving math reasoning is self-correction, which designs self-improving loops to let the model correct its own mistakes. However, existing self-correction approaches treat corrections as standalone post-generation refinements, relying on extra prompt and system designs to elicit self-corrections, instead of performing real-time, spontaneous self-corrections in a single pass. To address this, we propose SPOC, a spontaneous self-correction approach that enables LLMs to generate interleaved solutions and verifications in a single inference pass, with generation dynamically terminated based on verification outcomes, thereby effectively scaling inference time compute. SPOC considers a multi-agent perspective by assigning dual roles -- solution proposer and verifier -- to the same model. We adopt a simple yet effective approach to generate synthetic data for fine-tuning, enabling the model to develop capabilities for self-verification and multi-agent collaboration. We further improve its solution proposal and verification accuracy through online reinforcement learning. Experiments on mathematical reasoning benchmarks show that SPOC significantly improves performance. Notably, SPOC boosts the accuracy of Llama-3.1-8B and 70B Instruct models, achieving gains of 8.8% and 11.6% on MATH500, 10.0% and 20.0% on AMC23, and 3.3% and 6.7% on AIME24, respectively.

Discussion (0). Sign in to comment.

Forward citations

Cited by 8 Pith papers

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

  1. ReSum: Synergizing LLM Reasoning and Summarization with Reinforcement Learning

    cs.AI 2026-06 unverdicted novelty 7.0 of 10

    ReSum trains LLMs via RLVR to self-summarize reasoning trajectories, yielding 4% average performance gains and 18.6% shorter rollouts through contrastive rollout branches.

  2. SVR-R1: Bootstrapping Multi-modal Reasoning with Self-verification in Reinforcement Learning

    cs.AI 2026-07 conditional novelty 6.0 of 10

    Integrating binary self-verification into multi-turn GRPO rollouts raises VLM multimodal reasoning accuracy over matched GRPO baselines while the model learns to need fewer rethinks.

  3. Mixture of Debaters: Learn to Debate at Architectural Level in Multi-Agent Reasoning

    cs.AI 2026-06 unverdicted novelty 6.0 of 10

    Mixture of Debaters uses MoE to enable dynamic self-debate inside one model, claiming better accuracy than multi-agent systems at 3.7x lower latency and 87% fewer tokens on multimodal benchmarks.

  4. ReSum: Synergizing LLM Reasoning and Summarization with Reinforcement Learning

    cs.AI 2026-06 conditional novelty 6.0 of 10

    ReSum's contrastive RL branching on self-summarization points improves LLM math reasoning accuracy by about 4% and shortens rollouts by about 18.6% across tested backbones.

  5. The Self-Correction Illusion: Role Relabeling Gates Explicit Error Flagging in Large Language Models

    cs.AI 2026-06 conditional novelty 6.0 of 10

    Relabeling an identical erroneous claim from the model's own thought role to an external chat role increases explicit correction rates by 23-93 percentage points across 13 model-domain cells, indicating a chat-templat...

  6. The Self-Correction Illusion: Role Relabeling Gates Explicit Error Flagging in Large Language Models

    cs.AI 2026-06 conditional novelty 6.0 of 10

    LLMs flag byte-identical errors 23-93 points more often when the error is presented under an external chat role than inside their own <thought> block, so role labeling, not content, gates explicit self-correction.

  7. Token-Level LLM Collaboration via FusionRoute

    cs.AI 2026-01 unverdicted novelty 6.0 of 10

    FusionRoute augments token-level expert routing with a trainable complementary logit generator to expand the policy class and recover optimal decoding under mild conditions, outperforming prior collaboration and mergi...

  8. Failure Makes the Agent Stronger: Enhancing Accuracy through Structured Reflection for Reliable Tool Interactions

    cs.CV 2025-09 unverdicted novelty 5.0 of 10

    Structured reflection makes error diagnosis and repair an explicit trainable step that improves reliability and reduces redundant calls in tool-using LLM agents.

Pith tools