# AI 评审训练 AI 审稿人：科学判断坍缩现象与 TrustReviewer 缓解方案

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-09-17 08:00
- AIHOT 分数：43
- AIHOT 链接：https://aihot.news/items/cmuam2cqg0o33ro5t2s6urxs5
- 原文链接：https://arxiv.org/abs/2609.20942

## AI 摘要

研究以 Llama 3.1 8B 为基础，先用 ICLR 2018–2023 官方评审微调审稿模型，再用 ICLR 2024 数据混合官方与模型生成评审训练四个后继模型，发现引入合成评审会压缩评分分布并降低同论文与语料级语义多样性，作者称之为"科学判断坍缩"。

## 正文

Large language models (LLMs) increasingly participate in scientific evaluation, both as automated reviewers and as assistants to human reviewers. As model-generated reviews enter public data and future training corpora, AI peer review can become recursive: later reviewers learn from judgments produced by earlier models. We study one step of this feedback loop in a controlled setting. Starting from Llama 3.1 8B, we first fine-tune a reviewer on official ICLR reviews from 2018--2023 and then train four successor models on ICLR 2024 data with systematically varied mixtures of official and model-generated reviews.

Our study shows that introducing synthetic reviews compresses rating distributions and reduces both same-paper and corpus-level semantic diversity. We call this pattern scientific-judgment collapse. To mitigate this failure mode, we introduce TrustReviewer, an open-source LLM-based system for generating peer reviews of AI and machine learning papers. TrustReviewer intervenes at two complementary stages. For training-time prevention, we train the core reviewer in a single stage on a curated corpus designed to reduce low-quality and semantically degenerate supervision.

For test-time correction, paired activation steering aims to further mitigate residual tendencies toward collapsed judgments without further training or additional expert annotation. Together, these results characterize a concrete risk of recursive reviewer training and provide practical interventions for preserving judgment diversity and improving recommendation alignment in AI-assisted scientific evaluation.