NeurIPS

The 'NeurIPS' dataset/benchmark contains accepted camera-ready papers from the NeurIPS conference and is used to evaluate the prevalence of citation hallucination in peer-reviewed proceedings.

Papers using NeurIPS (2)

ResearchStudio-Idea: An Evidence-Grounded Research-Ideation Skill Suite from ML Conference Outcomes2026Phantom References: Hallucinated Citations That Survive Peer Review at Top-Tier Conferences2026