5 papers
RuleChef: Grounding LLM Task Knowledge in Human-Editable Rules
Ãdám Kovács, Nadia Verdha, Gábor Recski
We present RuleChef, a framework that uses large language models (LLMs) to generate executable rules for NLP tasks such as text classification, Named Entity Recognition (NER), or r…
Beyond Document Grounding: Span-Level Hallucination Detection over Code, Tool Output, and Documents
Ãdám Kovács, Bowei He, Xue Liu +3
Hallucination detection for retrieval-augmented generation (RAG) is usually evaluated on natural-language document evidence. However, grounded generation systems increasingly rely…
ACL-Verbatim: hallucination-free question answering for research
Gábor Recski, Szilveszter Tóth, Nadia Verdha +2
Academic researchers need efficient and reliable methods for collecting high-quality information from trusted sources, but modern tools for AI-assisted research still suffer from t…
Relation Extraction or Pattern Matching? Unravelling the Generalisation Limits of Language Models for Biographical RE
Varvara Arzt, Allan Hanbury, Michael Wiegand +2
Analysing the generalisation capabilities of relation extraction (RE) models is crucial for assessing whether they learn robust relational patterns or rely on spurious correlations…
LettuceDetect: A Hallucination Detection Framework for RAG Applications
Ãdám Kovács, Gábor Recski
Retrieval Augmented Generation (RAG) systems remain vulnerable to hallucinated answers despite incorporating external knowledge sources. We present LettuceDetect a framework that a…