Beyond Answer Voting: Fine-Grained Correction of LLM Reasoning through Graph Alignment across Sampled Traces
On the map: LLM reasoning → Knowledge- and tool-augmented reasoning also fits: Test-time compute and efficient thinking
Abstract (arXiv)
Large language models increasingly tackle tasks that require multi-step reasoning, where errors in intermediate claims can affect the final answer. Self-consistency (SC) improves reasoning by voting over sampled final answers, but its vote leaves intermediate disagreements unresolved. To address this problem, we introduce DDRG (Disagreement-Driven Verification Guided by Reasoning Graphs), a framework for selective correction from a frozen bank of reasoning traces. Our insight is that graph alignment can expose claim-level disagreements and provide correction evidence without expanding the sampled trace bank. DDRG builds source-grounded claim graphs and sends localized conflicts to an answer-blind verifier. A verifier-supported refutation that reaches the majority answer's supporting path triggers one focused repair; otherwise, DDRG returns the exact SC vote. Experimental results across five benchmarks show that native zero-shot DDRG improves macro-average accuracy over direct answering by 8.65 and 15.83 percentage points for Qwen2.5-7B and Llama3.1-8B. It also exceeds CoT, the evaluated GoT and GoV question-answering adaptations, and SC on this macro metric. DDRG links local verification to selective correction without enlarging the initial candidate bank.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning. Builds on 11 of its 28 references that are in the map. First posted on arXiv 13 Oct 2026.
Builds on
11 of 28 references.
References · 11
- 2024The Llama 3 Herd of Modelscited by 14,715
- 2021Training Verifiers to Solve Math Word Problemscited by 10,109
- 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
- 2022Large Language Models Encode Clinical Knowledgecited by 1,830
- 2022Solving Quantitative Reasoning Problems with Language Modelscited by 1,392
- 2020What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Examscited by 1,356
- 2023CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquingcited by 783
- 2023Logic-LM: Empowering Large Language Models with Symbolic Solvers for Faithful Logical Reasoningcited by 391
- 2022Large Language Models are Better Reasoners with Self-Verificationcited by 307
- 2020ReClor: A Reading Comprehension Dataset Requiring Logical Reasoningcited by 217
- 2026Reasoning Consensus: Structural Ensembling of LLM Reasoning via Weighted DAG Aggregationcited by 2
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026When Many Answers Are Valid, Voting Fails: Symbolic Verification for Best-of-K Causal Reasoning in LLMs
- 2024Reasoning Aware Self-Consistency: Leveraging Reasoning Paths for Efficient LLM Sampling
- 2026A Single Revision Step Improves Token-Efficient LLM Reasoning
- 2026Refining Over Resampling: Test-Time Self-Correction for LLM Reasoning
- 2026Denoising Iterative Self-Correction: Structured Verification Loops for Reliable LLM Reasoning
- 2024MIDGARD: Self-Consistency Using Minimum Description Length for Structured Commonsense Reasoning
- 2025Sample Smart, Not Hard: Correctness-First Decoding for Better Reasoning in LLMs
- 2025Confidence Improves Self-Consistency in LLMs