Evidence Absence Is Not Evidence Insufficiency: Diagnosing NEI Construction Artifacts in Fact Verification
Jingxi Qiu, Zeyu Han, Cheng Huang
Findings of EMNLP, 2026
Jingxi Qiu
Hi there 👋! My name is Jingxi Qiu. I am currently conducting independent research after receiving my M.S. in Data Science & Analytics from Georgetown University in 2025.
My research interests lie in trustworthy machine learning and decision-making under uncertainty. My recent work studies evidence sufficiency and selective prediction in retrieval-augmented generation, as well as how dataset construction artifacts affect the evaluation of fact-verification models.
My broader experience in full-stack development, computer vision, and offline reinforcement learning has influenced how I think about reliability beyond model-level performance. I am interested in extending my current research questions toward interactive and sequential settings, including multi-agent systems and embodied AI.
Email · CV · Google Scholar · GitHub
News
- Evidence Absence Is Not Evidence Insufficiency: Diagnosing NEI Construction Artifacts in Fact Verification was accepted to Findings of EMNLP 2026!
- SURE-RAG: Sufficiency and Uncertainty-Aware Evidence Verification for Selective Retrieval-Augmented Generation was accepted to IEEE PRAI 2026!
- Evidence Absence Is Not Evidence Insufficiency: Diagnosing NEI Construction Artifacts in Fact Verification was released on arXiv!
- SURE-RAG: Sufficiency and Uncertainty-Aware Evidence Verification for Selective Retrieval-Augmented Generation was released on arXiv!
Selected publicationsall publications →
SURE-RAG: Sufficiency and Uncertainty-Aware Evidence Verification for Selective Retrieval-Augmented Generation
Jingxi Qiu, Zeyu Han, Cheng Huang
IEEE PRAI, 2026