List your company

Audio-Visual Question Answering on AVSD (test)

165.1CIDErUniMVU
70.87695.338119.8144.262Mar 7, 2024Jul 19, 2024Dec 1, 2024Apr 15, 2025Aug 28, 2025Jan 10, 2026May 25, 2026

Evaluation results

MethodLinks
UniMVUSize=7B, Training regime=task-specific training2026.05165.139.8
UniMVU†Size=7B, Training regime=unified multi-task training2026.05162.739.5
PAVE*Size=7B, Evaluation protocol=reproduced baseline2026.05151.638.5
UniMVUSize=0.5B, Training regime=task-specific training2026.05147.138.2
UniMVU†Size=0.5B, Training regime=unified multi-task training2026.05145.937.8
PAVE*Size=0.5B, Evaluation protocol=reproduced baseline2026.05134.936.5
LLaVA-OV-FT* (video-audio concat)Size=0.5B, Input modalities=video-audio concat, Evaluation protocol=reproduced baseline2026.05127.835.7
LLaVA-OV-FT (video-only)Size=7B, Input modalities=video-only2026.05124.9—
LLaVA-OV-FT (video-only)Size=0.5B, Input modalities=video-only2026.05117.6—
COSTzero-shot=false2024.03108.5—
MTNzero-shot=false2024.0398.5—
FA+HREDzero-shot=false2024.0384.3—
CAT-7Bzero-shot=true2024.0379—
ChatBridge-13Bzero-shot=true2024.0375.4—
OneLLM-7Bzero-shot=true2024.0374.5—