Skip to content
Live
Hugging Face Daily Papers·· 13 hours ago

PaperCritic: Training AI Reviewers to Assess the Soundness of Papers

PaperCritic: Training AI Reviewers to Assess the Soundness of Papers

Brief overview

PaperCritic AI reviewers trained through reinforcement learning outperform leading models and other training methods in jointly maximizing the precision and recall of rigorous review points, and their capabilities can be transferred to other review fields. The model is trained using the PaperSound-23K dataset, which includes 1K ICLR submissions and 23K human and LLM-generated review points, and performs excellently on the domain-specific test set. Learning ICLR reviews has also been transferred to cross-domain tasks such as Nature journal papers and code reviews.

Article

Published on Nov 13

Authors:

,

,

,

,

,

Abstract

An AI reviewer trained by reinforcement learning to jointly maximize precision and recall of soundness review points outperforms frontier models and other training methods and transfers to other review domains.

As AI conference submissions grow rapidly, AI reviewers are being considered as an alternative. However, simply imitating human reviewing is problematic because reviewing serves multiple purposes, including evaluating papers' soundness, interestingness, and clarity, the last two of which are subjective. In this work, we focus solely on evaluating soundness, the most important and objective criterion, and define the learning objective as jointly maximizing (1) precision, based on the soundness and substantiveness of review points, and (2) recall, the coverage of high-quality reference review points. Under this problem formulation, we construct a quality-labeled review point dataset, PaperSound-23K, using 1K ICLR submissions and 23K human-written and LLM-generated review points. Using this dataset, we train an AI reviewer model, PaperCritic, through reinforcement learning with the F1 score of precision and recall as the reward. Our experiments show that PaperCritic outperforms frontier models and models trained with other AI reviewer training methods on an in-domain test set. Surprisingly, learning to review ICLR submissions also transfers to reviewing Nature-family papers and even to out-of-domain reviewing tasks (e.g., code review, legal contract review, and feedback on story writing).

View arXiv page View PDF Add to collection

Get this paper in your agent:

hf papers read 2611.17834

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2611.17834 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2611.17834 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2611.17834 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

Source: Hugging Face Daily Papers · huggingface.co