Recent multimodal large language models (MLLMs) such as GPT-4o and Qwen3-Omni show strong perception but struggle in multi-speaker, dialogue-centric settings that demand agentic reasoning tracking who speaks, maintaining roles, and grounding events across time. These scenarios are central to multimodal audio-video understanding, where models must jointly reason over audio and visual streams in applications such as conversational video assistants and meeting analytics. We introduce AMUSE, a benchmark designed around tasks that are inherently agentic, requiring models to decompose complex… Read More
AMUSE: Audio-Visual Benchmark and Alignment Framework for Agentic Multi-Speaker Understanding
Recent multimodal large language models (MLLMs) such as GPT-4o and Qwen3-Omni show strong perception but struggle in multi-speaker, dialogue-centric settings that demand agentic reasoning tracking who speaks, maintaining roles, and grounding events across time. These scenarios are central to multimodal audio-video understanding, where models must jointly reason over audio and visual streams in applications such as conversational video assistants and meeting analytics. We introduce AMUSE, a benchmark designed around tasks that are inherently agentic, requiring models to decompose complex…
Related Posts
3 things to know about Ironwood, our latest TPU
Google’s seventh-gen Tensor Processing Unit is here! Learn what makes Ironwood our most powerful and…
The Complete Guide to Using Pydantic for Validating LLM Outputs
Large language models generate text, not structured data.
Toward Machine Interpreting: Lessons from Human Interpreting Studies
Current speech translation systems, while having achieved impressive accuracies, are rather static in their behavior…