Pith. sign in

REVIEW 6 cited by

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 2501.17703 v4 pith:WTGFSDP2 submitted 2025-01-29 cs.CL

Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitate

classification cs.CL
keywords critiquemodelsqwen2effectivefine-tuninglearningnoisyreasoning
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Supervised Fine-Tuning (SFT) is commonly used to train language models to imitate annotated responses for given instructions. In this paper, we propose Critique Fine-Tuning (CFT), a method more effective than SFT for reasoning tasks. Instead of simply imitating correct responses, CFT trains models to critique noisy responses, inspired by human learning processes that emphasize critical thinking, deeper analysis, and nuanced understanding - traits often overlooked by standard SFT. To validate the effectiveness of CFT, we construct multiple critique datasets (e.g., WebInstruct, MetaMath, NuminaMath), where GPT-4o serves as the teacher to generate critiques in the form of ([query; noisy response], critique). Experiments on these datasets demonstrate that CFT consistently outperforms SFT by 4-10% across six mathematical reasoning benchmarks, and is effective across different base models including Qwen2.5, Qwen2.5-Math, and DeepSeek-Math. Notably, our model Qwen2.5-Math-CFT only requires 1 hour of training on 8 x H100 over the 50K examples, yet matches or outperforms strong competitors like Qwen2.5-Math-Instruct on most benchmarks, which use over 2M samples. Moreover, it matches the performance of SimpleRL, which is a DeepSeek-r1 replication trained with 140 x more compute. Experiments on IF_Eval and MT-Bench further demonstrate that CFT can significantly enhance the model's general generation and instruction-following capabilities, outperforming the Qwen2.5-Math-Instruct by a large margin. Ablation studies show that CFT is robust to noisy response sources and teacher critique models. These findings highlight that CFT offers a more effective alternative to advance the reasoning of language models.

discussion (0)

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

Forward citations

Cited by 6 Pith papers

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

  1. SWE-Review: Closing the Loop on Issue Resolution with Agentic Code Review

    cs.SE 2026-07 conditional novelty 6.0

    An agentic code reviewer that explores repositories to judge and diagnose AI-generated pull requests improves resolve rates from 27.5% to 56.9% and outperforms single-turn review baselines.

  2. Learning from Failures: Correction-Oriented Policy Optimization with Verifiable Rewards

    cs.CL 2026-05 unverdicted novelty 6.0

    CIPO jointly optimizes standard RLVR rewards with correction samples derived from the model's own failed attempts, yielding better reasoning and self-correction on math and code benchmarks.

  3. Building a Precise Video Language with Human-AI Oversight

    cs.CV 2026-04 unverdicted novelty 6.0

    CHAI framework pairs AI pre-captions with expert human critiques to produce precise video descriptions, enabling open models to outperform closed ones like Gemini-3.1-Pro and improve fine-grained control in video gene...

  4. rePIRL: Learn PRM with Inverse RL for LLM Reasoning

    cs.LG 2026-02 unverdicted novelty 6.0

    rePIRL learns effective process reward models for LLM reasoning via a dual policy-PRM update process inspired by inverse RL, unifying online and offline methods with reported gains over prior approaches on math and co...

  5. Teaching Large Language Models When Not to Know: Learning Temporal Critique for Ex-Ante Reasoning

    cs.AI 2026-05 unverdicted novelty 5.0

    TCFT trains LLMs on temporal critique tasks to reduce post-cutoff knowledge leakage by 37-42 percentage points over prompting and standard SFT on Qwen models.

  6. FakeVLM-R1: Internalizing Physical Laws via CoT for Synthetic Image Detection

    cs.CV 2026-05 unverdicted novelty 4.0

    FakeVLM-R1 combines GRPO reinforcement learning with critical-thinking CoT and a physics-annotated FakeClue++ dataset to reach claimed SOTA synthetic image detection while reducing over-rejection of real images.