HaluNet: Learning Hallucination Risk from Internal Signals in LLM Question Answering 事件

HaluNet: Learning Hallucination Risk from Internal Signals in LLM Question Answering arXiv:2512.24562v2 Announce Type: replace Abstract: Large language models (LLMs) achieve strong question answering (QA) performance but can produce fluent answers unsupported by available evidence. Existing hallucination detectors often rely on external verification, repeated sampling, or test-time judge calls, which can be costly for real-time QA. We propose \textbf{HaluNet}, a lightweight hallucination risk e