REVIEW 8 cited by
CritiqueLLM: Towards an Informative Critique Generation Model for Evaluation of Large Language Model Generation
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
CritiqueLLM: Towards an Informative Critique Generation Model for Evaluation of Large Language Model Generation
read the original abstract
Since the natural language processing (NLP) community started to make large language models (LLMs) act as a critic to evaluate the quality of generated texts, most of the existing works train a critique generation model on the evaluation data labeled by GPT-4's direct prompting. We observe that these models lack the ability to generate informative critiques in both pointwise grading and pairwise comparison especially without references. As a result, their generated critiques cannot provide fine-grained distinguishability on generated texts, causing unsatisfactory evaluation performance. In this paper, we propose a simple yet effective method called Eval-Instruct, which can first acquire pointwise grading critiques with pseudo references and then revise these critiques via multi-path prompting to obtain informative evaluation data in different tasks and settings, including pointwise grading and pairwise comparison with / without references. After fine-tuning on these data, the resulting model CritiqueLLM is empirically shown to outperform ChatGPT and all the open-source baselines and even achieve comparable evaluation performance to GPT-4 in system-level correlations of pointwise grading. We also demonstrate that our generated critiques can act as scalable feedback to further improve the generation quality of strong LLMs like ChatGPT.
Forward citations
Cited by 8 Pith papers
-
BLUEX v2: Benchmarking LLMs on Open-Ended Questions from Brazilian University Entrance Exams
BLUEX v2 supplies 919 graded subquestions from UNICAMP and USP exams (2022-2025) and reports LLM-as-a-judge scores for 21 models showing a 4.92-point spread, with math reasoning and image understanding as the weakest areas.
-
HumanAgencyBench: Scalable Evaluation of Human Agency Support in AI Assistants
A new benchmark finds low to moderate human agency support in 20 LLM assistants across six dimensions.
-
Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs
o1-like models overthink easy tasks; self-training reduces compute use without accuracy loss on GSM8K, MATH500, GPQA, and AIME.
-
BLUEX v2: Benchmarking LLMs on Open-Ended Questions from Brazilian University Entrance Exams
BLUEX v2 creates a public dataset from 2022-2025 UNICAMP and USP second-phase exams and evaluates 21 LLMs via LLM-as-judge, finding scores from 4.18 to 9.10 with math and image understanding as weakest areas.
-
Towards an AI co-scientist
A multi-agent AI system generates novel biomedical hypotheses that show promising experimental validation in drug repurposing for leukemia, new targets for liver fibrosis, and a bacterial gene transfer mechanism.
-
Judging the Judges: A Systematic Evaluation of Bias Mitigation Strategies in LLM-as-a-Judge Pipelines
Style bias dominates LLM-as-a-Judge systems far more than position bias, with debiasing strategies providing model-dependent gains and public tools released for replication.
-
Judging the Judges: A Systematic Evaluation of Bias Mitigation Strategies in LLM-as-a-Judge Pipelines
Gemini 2.5 Flash with a Combined Budget debiasing strategy achieves 71.0% judge agreement at ~$0.001/evaluation, outperforming frontier models at 15x lower cost.
-
InternLM2 Technical Report
InternLM2 is a new open-source LLM that outperforms prior versions on 30 benchmarks and long-context tasks through scaled pre-training to 32k tokens and a conditional online RLHF alignment strategy.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.