cs.CLNov 9, 2026

Catch It Where It Happens: Real Time Error Correction During LLM Reasoning

Authors: Rotem Elimelech, Michael Shalyt, Ido Kaminer

Organizations: Technion – Israel Institute of Technology, Haifa 3200003, Israel

Abstract

Large language models (LLMs) are increasingly trusted with mathematical, scientific and engineering work, yet their mistakes are often confidently written and hard to spot. A single wrong step tends to propagate, making the model waste inference tokens following a false premise. Existing safeguards either check only the final answer and restart from scratch, or rely on a verifier LLM that competes for GPU capacity and slows down response time. We propose online, deterministic verification: many kinds of reasoning steps can be validated by code, so a step can be verified in real time, as it is written, without any model call or GPU use. Whenever a check fails, we point out the mistake and let the model continue from the same spot. We demonstrate this approach on symbolic mathematics, where derivations already contain explicit expressions that a computer algebra system can check in milliseconds. Tested on the ASyMoB benchmark with three open-weights models, our online validation improves accuracy by 1.4-5.6 percentage points, corresponding to a 1.8-14.9% net reduction in incorrect answers, while increasing token usage by only 6-7%. Because the verifier is independent of the model, the approach complements progress in LLMs rather than competing with it: every new model can benefit immediately, and the benefit grows as deterministic checkers extend to more kinds of reasoning steps.

Figures & tables

Explore similar work

CardsList
  1. Beyond the Answer Key: Robustness Evaluation of Large Language Models for Step-Level Mathematical Verification

    Aug 28, 2026Fateme Mazdarani, Carlos ToxtliLLM EvaluationLLM Answer Verification

  2. One-Token Verification for Reasoning Correctness Estimation

    Mar 1, 2026Zhan Zhuang, Xiequn Wang, Zebin Chen +4Reasoning VerificationLLM Reasoning