UCSC at SemEval-2025 Task 3: Context, Models and Prompt Optimization for Automated Hallucination Detection in LLM Output
This addresses the challenge of hallucination detection for users relying on LLMs for knowledge-intensive queries, though it is incremental as it builds on existing shared-task efforts.
The paper tackled the problem of detecting and localizing hallucinations in large language model outputs by introducing a framework that retrieves context, identifies false content, and maps it to spans, achieving the highest overall performance with a #1 ranking in average position across all languages in the SemEval 2025 Task 3.
Hallucinations pose a significant challenge for large language models when answering knowledge-intensive queries. As LLMs become more widely adopted, it is crucial not only to detect if hallucinations occur but also to pinpoint exactly where in the LLM output they occur. SemEval 2025 Task 3, Mu-SHROOM: Multilingual Shared-task on Hallucinations and Related Observable Overgeneration Mistakes, is a recent effort in this direction. This paper describes the UCSC system submission to the shared Mu-SHROOM task. We introduce a framework that first retrieves relevant context, next identifies false content from the answer, and finally maps them back to spans in the LLM output. The process is further enhanced by automatically optimizing prompts. Our system achieves the highest overall performance, ranking #1 in average position across all languages. We release our code and experiment results.