LLM Peer Review and Revision: Critique Inflation and Advice–Reward Misalignment
Investigating how LLM-generated peer review shapes manuscript revision using rejected ICLR papers and multi-model reviewer, rewriter, and judge pipelines.
Investigating how LLM-generated peer review shapes manuscript revision using rejected ICLR papers and multi-model reviewer, rewriter, and judge pipelines.
Highlights
- Testing whether open-ended reviews produce critique inflation and whether feedback-driven revisions improve reviewer-visible signals without corresponding gains in overall evaluation.
- Comparing constrained review interfaces, including Top-3 priorities, stopping rules, and cost-aware feedback, to identify more decision-relevant review designs.
Metadata
- Context: Research Collaboration with Yiming Li
- Role: Research Collaborator
- Timeline: Jul 2026 - Present