AMUSE: Audio-Visual Benchmark and Alignment Framework for Agentic Multi-Speaker Understanding

TL;DR

AMUSE is a new benchmark for evaluating multimodal large language models in agentic, multi-speaker audio-visual understanding tasks, revealing current limitations and proposing RAFT for improved alignment and reasoning.

Contribution

The paper introduces AMUSE, a comprehensive benchmark for agentic multi-speaker audio-visual tasks, and proposes RAFT, a novel data-efficient alignment framework for enhancing model reasoning.

Findings

Current models show weak multi-speaker reasoning.

Models exhibit inconsistent behavior in agentic settings.

RAFT improves accuracy by up to 39.52% on the benchmark.

Abstract

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 audio-visual interactions into planning, grounding, and reflection steps. It evaluates MLLMs across three modes zero-shot, guided, and agentic and six task families, including spatio-temporal speaker grounding and multimodal dialogue summarization. Across all modes, current models exhibit weak multi-speaker reasoning and inconsistent behavior under both non-agentic and agentic evaluation. Motivated by the inherently agentic nature of these tasks and recent advances in LLM agents, we propose RAFT, a data-efficient agentic alignment framework that integrates reward optimization with intrinsic multimodal self-evaluation as reward and selective parameter adaptation for data and parameter efficient updates. Using RAFT, we achieve up to 39.52\% relative improvement in accuracy on our benchmark. Together, AMUSE and RAFT provide a practical platform for examining agentic reasoning in multimodal models and improving their capabilities.