VeriTrail: Detect hallucination and trace provenance in AI workflows
- Dasha Metropolitansky, Microsoft
Dasha Metropolitansky, Research Data Scientist, Microsoft Research Special Projects, introduces VeriTrail, a new method for closed-domain hallucination detection in multi-step AI workflows. Unlike prior methods, VeriTrail provides traceability: it identifies where hallucinated content was likely introduced, and it establishes the provenance of faithful content by tracing a path to the source text. VeriTrail also outperforms baseline methods in hallucination detection. The combination of traceability and effective hallucination detection makes VeriTrail a powerful tool for auditing the integrity of content generated by language models.
-
-
Dasha Metropolitansky
Senior Research Data Scientist
-
-
接下来观看
-
-
Session: Compute & Trust (Systems)
- Ashish Panwar,
- Aditya Desai,
- Abhilash Jindal
-
Multimodal & Embodied Intelligence (Pt 1), Panel on Multimodal AI: Progress, Pitfalls, Possibilities
- Madhava Krishna,
- Sriram Ganapathy,
- Somak Aditya
-
Session on Compute & Trust (Security)
- Krishna Pillutla,
- Danish Pruthi
-
-
Session on Reasoning
- Hongxiang Fan,
- Nagarajan Natarajan
-
-
Session on Retrieval
- Lokesh Nagalapatti,
- Soumen Chakrabarti
-
-