REVIEW 5 cited by
LLM Critics Help Catch Bugs in Mathematics: Towards a Better Mathematical Verifier with Natural Language Feedback
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
read the original abstract
In recent progress, mathematical verifiers have achieved success in mathematical reasoning tasks by validating the correctness of solutions generated by policy models. However, existing verifiers are trained with binary classification labels, which are not informative enough for the model to accurately assess the solutions. To mitigate the aforementioned insufficiency of binary labels, we introduce step-wise natural language feedback as rationale labels, that is, the correctness of each step and the detailed explanations. In this paper, we propose Math-Minos, a natural language feedback-enhanced verifier by constructing automatically generated training data and a two-stage training paradigm for effective training and efficient inference. Our experiments reveal that a small set of natural language feedback can significantly boost the performance of the verifier in both verification and reinforcement learning. We have released the code and data for further exploration.
Forward citations
Cited by 5 Pith papers
-
Dynamic and Generalizable Process Reward Modeling
DG-PRM stores multi-dimensional process reward criteria in a hierarchical tree, selects them dynamically per step, and uses Pareto dominance to build preference pairs, claiming state-of-the-art process reward modeling.
-
GRPO-CARE: Consistency-Aware Reinforcement Learning for Multimodal Reasoning
GRPO-CARE improves answer accuracy and reasoning coherence over standard GRPO on a new video reasoning benchmark, with a 6.7 point gain on the hardest level and a 24.5 point higher consistency rate.
-
FreePRM: Training Process Reward Models Without Ground Truth Process Labels
A weakly supervised PRM training method using outcome-only pseudo-labels plus a buffer probability reaches 53.0% F1 on ProcessBench, beating supervised baselines in the paper's comparisons.
-
Reward Modeling for Reinforcement Learning-Based LLM Reasoning: Design, Challenges, and Evaluation
A taxonomy-driven survey arguing that reward design is the central mechanism shaping reliable LLM reasoning, with maps of reward paradigms, reward-hacking failure modes, and benchmark pitfalls.
-
A Survey on Large Language Models for Mathematical Reasoning
Recent advances in LLM mathematical reasoning are organized into comprehension and generation phases, covering methods from prompting to test-time scaling.
Discussion (0). Continue with ORCID to comment.