LegalHalluLens: Typed Hallucination Auditing and Calibrated Multi-Agent Debate for Trustworthy Legal AI
LegalHalluLens audits AI systems in legal workflows by identifying specific error patterns and directional biases in hallucinations across different claim types, enabling more reli…
Hugging Face · Daily Papers
·Lalit Yadav, Akshaj Gurugubelli
·
·▲ 5 upvotes
Este artigo está em destaque na seleção diária de papers do Hugging Face, curada pela comunidade de pesquisa em IA.
Autores: Lalit Yadav, Akshaj Gurugubelli
- 5 upvotes da comunidade
- Temas: hallucination profiles, Risk Direction Index, debate pipeline, CUAD dataset, clause-level instances, multi-agent debate
Resumo
Resumo original (em inglês), extraído do paper:
LegalHalluLens audits AI systems in legal workflows by identifying specific error patterns and directional biases in hallucinations across different claim types, enabling more reliable deployment through targeted diagnostic and mitigation approaches.
