Get Started
Home
Topics
Search
Library
Research questionHow can we reliably detect when an LLM response is unsupported by its reference documents?LLM outputs can contain factual claims absent from or contradicted by their source documents, while opaque generation provides little evidence for checking them. Detection must assess whether responses are grounded in those references without domain-specific fine-tuning.
AI
Alignment & Safety
Evaluation & Benchmarks
Natural Language Processing
Reasoning
Retrieval-Augmented Generation
Latest papersRecent research connected to this question, newest first.Leveraging Low-Level Symbolic Competences for Unsupervised Grounding in Hallucination DetectionThe source studies a pipeline that constructs an SQL database from reference documents and reasons over that database and the sampled response. Evidence comes from the RAGTruth and DiaHalu hallucination-detection datasets.research paper · Sep 4, 2026
Related questions
How can we detect object hallucinations in vision-language models when visual grounding shifts across layers?How can black-box LLM APIs detect hallucinations without trusted context while limiting false alarms?How can sampling-only diagnosis identify whether an individual LLM hallucination comes from retrieval, training, or autoregressive commitment?How can hallucination detectors verify claims and localize unsupported spans without costly separate pipelines?