← Back to all news
arXiv Machine Learning By Jian Xu, Delu Zeng, John Paisley, Qibin Zhao

Common Cause, Not Cross-Attention: Blocking Visual Shortcuts in Audio-Video Generation

Read the original on arXiv Machine Learning →

The paper investigates how joint audio–video generation models learn to associate sound with visual appearance rather than the underlying causal event, a problem termed the visual shortcut. By constructing a structural causal model where audio is independent of video appearance, the authors demonstrate that cross‑attention and shared latent approaches fail when appearance‑event correlations are broken, and that common‑cause routing does not solve the issue. They propose blocking the shortcut via interventions on nuisance variables, proving that counterfactual invariance is necessary and sufficient for identifying the causal predictor, and validate this approach on synthetic and real datasets, including a pretrained video‑to‑audio generator.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Sep 23

Intervention, Not Shared Latents: Blocking Visual Shortcuts in Audio-Video Generation

The paper investigates why joint audio–video generators often learn to predict sound from visual appearance rather than from the underlying event, a problem termed the visual shortcut. By constructing a controlled causal model where audio is independent of video appearance, the authors show that common remedies such as shared latent spaces fail to prevent this shortcut. They propose that intervening on the nuisance appearance is necessary and sufficient for counterfactual invariance, and validate this approach across synthetic and real datasets, highlighting the remaining challenge of unknown nuisances.

By Jian Xu, Delu Zeng, John Paisley
diffusion
More like this →
arXiv AI
Sep 4

The Attention Triangle in Audio-Video Models

The paper investigates audio‑video diffusion models by examining the "attention triangle"—the cross‑attention links among text, audio, and video. It finds that the audio‑video edge is bidirectional and heavily influenced by model biases, leading to semantic leakage when prompts conflict with learned priors. The authors develop attention‑derived diagnostics and inference‑time interventions that improve semantic grounding without sacrificing generation quality.

By Sagi Polaczek, Noa Kraicer, Gal Metzer, Zhuo Ning, Ali Mahdavi-Amiri, Daniel Cohen-Or, Raja Giryes
diffusionmultimodalsafety
More like this →
arXiv Computer Vision
Sep 16

Video-HolmesV2: Can MLLMs Reason with Spatio-Temporal Audio-Visual Evidence in Long Videos?

Video-HolmesV2 is a new benchmark that tests multimodal large language models on their ability to reason with spatio‑temporal audio‑visual evidence in long videos. It requires models to justify answers with precise evidence, uses a multi‑model cross‑verification pipeline and a spatio‑temporal evidence‑aware metric, and introduces an audio‑text guided token compression framework to reduce long‑context noise. In evaluations, even strong proprietary models score below 60% while the proposed approach outperforms comparable open‑source omni‑models.

By Zhaoyang Wei, Zipeng Wang, Yushe Cao, Chenhui Qiang, Shuaibing Cheng, Xuesong Yang, Sen Nie, Bowen Jiang, Wenchao Ding, Yanchao Hao, Zheng Wei, Xuehui Yu, Zhenjun Han
llmsmultimodalbenchmarks
More like this →