Skip to content
conftrace_

peer review

79 papers

Papers

TitleTrap: Probing Presentation Bias in LLM-Based Scientific Reviewing AACL 2025 TreeReview: A Dynamic Tree of Questions Framework for Deep and Efficient LLM-based Scientific Peer Review EMNLP 2025 UPME: An Unsupervised Peer Review Framework for Multimodal Large Language Model Evaluation CVPR 2025 AgentReview: Exploring Peer Review Dynamics with LLM Agents EMNLP 2024 An Analysis of Tasks and Datasets in Peer Reviewing ACL 2024 ARIES: A Corpus of Scientific Paper Edits Made in Response to Peer Reviews ACL 2024 Automated Focused Feedback Generation for Scientific Writing Assistance ACL 2024 CASIMIR: A Corpus of Scientific Articles Enhanced with Multiple Author-Integrated Revisions COLING 2024 Eliciting Honest Information from Authors Using Sequential Review AAAI 2024 Monitoring AI-Modified Content at Scale: A Case Study on the Impact of ChatGPT on AI Conference Peer Reviews ICML 2024 ‘Quis custodiet ipsos custodes?’ Who will watch the watchmen? On Detecting AI-generated peer-reviews EMNLP 2024 Re3: A Holistic Framework and Dataset for Modeling Collaborative Document Revision ACL 2024 Recommending Missed Citations Identified by Reviewers: A New Task, Dataset and Baselines COLING 2024 The State of Data Curation at NeurIPS: An Assessment of Dataset Development Practices in the Datasets and Benchmarks Track NeurIPS 2024 A One-Size-Fits-All Approach to Improving Randomness in Paper Assignment NeurIPS 2023 Automatic Analysis of Substantiation in Scientific Peer Reviews EMNLP 2023 Calibrating “Cheap Signals” in Peer Review without a Prior NeurIPS 2023 Causal Matching with Text Embeddings: A Case Study in Estimating the Causal Effects of Peer Review Policies ACL 2023 Exploring Jiu-Jitsu Argumentation for Writing Peer Review Rebuttals EMNLP 2023 Group Fairness in Peer Review NeurIPS 2023 Labels are not necessary: Assessing peer-review helpfulness using domain adaptation based on self-training ACL 2023 Missing Information, Unresponsive Authors, Experimental Flaws: The Impossibility of Assessing the Reproducibility of Previous Human Evaluations in NLP EACL 2023 NLPeer: A Unified Resource for the Computational Study of Peer Review ACL 2023 Reviewriter: AI-Generated Instructions For Peer Review Writing ACL 2023 Understanding Ethics in NLP Authoring and Reviewing EACL 2023