AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

Pushing the Frontier of Audiovisual Perception with Large-Scale Multimodal Correspondence Learning

Apoorv Vyas, Heng-Jui Chang, Cheng-Fu Yang, Po-Yao Huang, Luya Gao, Julius Richter, Sanyuan Chen, Matthew Le, Piotr Dollár, Christoph Feichtenhofer, Ann Lee, Wei-Ning Hsu

CVPR 2026 · Vision-language and multimodal › Cross-modal learning · approach: Self-supervised audio-visual representation learning · 23 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add stage-2 improved captions over stage-1 captions (ablation)PEAV-B · pre-training ablation averageavg score: 38.9 -> 41.5Improved
use data-engine captions instead of raw EnCLAP captions (ablation)PEAV-B · pre-training ablation averageavg score: 33.1 -> 41.5Improved
use synthetic captions only instead of real only (ablation)PEAV-B · data mixing ablationavg score: 14.9 -> 43.0Improved
mix real and synthetic captions at 1:10 instead of synthetic only (ablation)PEAV-B · data mixing ablationavg score: 43.0 -> 45.4Improved
raise synthetic ratio from 1:10 to 1:30 (ablation)PEAV-B · data mixing ablationavg score: 45.4 -> 44.2No clear effect
scale training data from 2M to 64M (ablation)PEAV-B · data scaling ablationavg score: 38.4 -> 43.4Improved
scale audio encoder depth beyond 20 layers (ablation)PEAV-B · audio encoder scalingavg score: 44.5 -> 42.0Hurt
scale audio encoder from 8 to 20 layers (ablation)PEAV-B · audio encoder scalingavg score: 41.1 -> 44.5Improved
expand contrastive pairs from one audio-text pair to all eight (ablation)PEAV-B · contrastive coverage ablationavg score: 19.0 -> 43.2Improved
use native joint embedding instead of max over uni-modal queries (ablation)PEAV-L · AudioCaps T+V->AR@1: 69.8 -> 80.2Improved
add stage-2 fine-tuning with speech data (ablation)PEAV-L · VCTK speech-to-transcriptR@1: 16.7 -> 85.6Improved
use 30 fps instead of 16 frames (ablation)PEAV-L · video retrieval averageavg R@1: 66.9 -> 67.9Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy