SEAR: Segment-Evidence-Aware Routing for Weak-to-Strong Multilingual Speech MCQ
This paper describes our system for Task~2 of the second Multilingual Conversational Speech Language Model (MLC-SLM) Challenge. We adapt Qwen3-Omni-30B-A3B-Instruct with a segment-evidence-aware data and post-training pipeline. A language model converts timestamped ASR into coherent event spans, which are expanded by a...
H. Le, L. Nguyen, Minh Tri Dao
· 1 citation