Model,clotho_aqa_test,audiocaps_qa_test,wavcaps_qa_test salmonn_7b,55.75299866725899,47.02875399361022,46.25 wavllm_fairseq,43.01199466903598,29.840255591054312,26.25 whisper_large_v3_with_llama_3_8b_instruct,1.4278098622834297,0.7252396166134185,0.819078947368421 qwen_audio_chat,58.196357174589075,47.98722044728435,38.68421052631579