一作ACL 2026Findings· Findings of ACL
AHA : Aligning Large Audio-Language Models for Reasoning Hallucinations via Counterfactual Hard Negatives
Yanxi Chen · Wenhui Zhu · Xiwen Chen · Zhipeng Wang · Xin Li · Peijie Qiu · Hao Wang · XUANZHAO DONG · Yujian Xiong · Anderson Schneider · Yuriy Nevmyvaka · Yalin Wang
摘要
Although Large Audio-Language Models (LALMs) deliver state-of-the-art (SOTA) performance, they frequently suffer from hallucinations, e.g., generating text not grounded in the audio input. We analyze these grounding failures and identify a distinct taxonomy: Event Omission, False Event Identity,...