We introduce AMUSE, a benchmark for evaluating multimodal large language models on agentic multi-speaker audio-visual reasoning, and propose RAFT, a data-efficient agentic alignment framework achieving up to 39.52% relative improvement.

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.

We combine blind audio recordings with 3D scene information to estimate sound anywhere in a scene, jointly tackling source localization, separation, and dereverberation.

We investigate the benefit of combining blind audio recordings with 3D scene information for novel-view acoustic synthesis. Given audio recordings from 2-4 microphones and the 3D geometry and material of a scene containing multiple unknown sound sources, we estimate the sound anywhere in the scene. We identify the main challenges of novel-view acoustic synthesis as sound source localization, separation, and dereverberation. While naively training an end-to-end network fails to produce high-quality results, we show that incorporating room impulse responses (RIRs) derived from 3D reconstructed rooms enables the same network to jointly tackle these tasks. Our method outperforms existing methods designed for the individual tasks, demonstrating its effectiveness at utilizing 3D visual information. In a simulated study on the Matterport3D-NVAS dataset, our model achieves near-perfect accuracy on source localization, a PSNR of 26.44dB and a SDR of 14.23dB for source separation and dereverberation, resulting in a PSNR of 25.55 dB and a SDR of 14.20 dB on novel-view acoustic synthesis. We release our code and model on our project website.

We use audio sensing to improve the performance of visual localization methods on three tasks: relative pose estimation, place recognition, and absolute pose regression.

In this work, we show how to estimate a device's position and orientation indoors by echolocation, i.e., by interpreting the echoes of an audio signal that the device itself emits. Established visual localization methods rely on the device's camera and yield excellent accuracy if unique visual features are in view and depicted clearly. We argue that audio sensing can offer complementary information to vision for device localization, since audio is invariant to adverse visual conditions and can reveal scene information beyond a camera's field of view. We first propose a strategy for learning an audio representation that captures the scene geometry around a device using supervision transfer from vision. Subsequently, we leverage this audio representation to complement vision in three device localization tasks: relative pose estimation, place recognition, and absolute pose regression. Our proposed methods outperform state-of-the-art vision models on new audio-visual benchmarks for the Replica and Matterport3D datasets.

We study the robustness of multimodal models on three tasks-- action recognition, object detection, and sentiment analysis-- and develop a robust fusion strategy that protects against worst-case errors caused by a single modality.

Beyond achieving high performance across many vision tasks, multimodal models are expected to be robust to single-source faults due to the availability of redundant information between modalities. In this paper, we investigate the robustness of multimodal neural networks against worst-case (i.e., adversarial) perturbations on a single modality. We first show that standard multimodal fusion models are vulnerable to single-source adversaries: an attack on any single modality can overcome the correct information from multiple unperturbed modalities and cause the model to fail. This surprising vulnerability holds across diverse multimodal tasks and necessitates a solution. Motivated by this finding, we propose an adversarially robust fusion strategy that trains the model to compare information coming from all the input sources, detect inconsistencies in the perturbed modality compared to the other modalities, and only allow information from the unperturbed modalities to pass through. Our approach significantly improves on state-of-the-art methods in single-source robustness, achieving gains of 7.8-25.2% on action recognition, 19.7-48.2% on object detection, and 1.6-6.7% on sentiment analysis, without degrading performance on unperturbed (i.e., clean) data.