AI Paper Atlas

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

Semantic Noise Reduction via Teacher-Guided Dual-Path Audio-Visual Representation Learning

Linge Wang, Yingying Chen, Bingke Zhu, Lu Zhou, Jinqiao Wang

CVPR 2026 · Vision-language and multimodal › Cross-modal learning · approach: Self-supervised audio-visual representation learning · 1 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
use TG-DP dual-path training instead of single-pass CAV-MAE SyncCAV-MAE Sync · AudioSet A->V zero-shot retrievalR@1: 27.9 -> 37.1Improved
use TG-DP dual-path training instead of single-pass CAV-MAE SyncCAV-MAE Sync · AudioSet V->A zero-shot retrievalR@1: 35.2 -> 37.4Improved
use TG-DP instead of CAV-MAE Sync for unimodal probingCAV-MAE Sync · AS20K audio-onlymAP: 29.3 -> 31.2Improved
use TG-DP instead of CAV-MAE Sync for unimodal probingCAV-MAE Sync · AS20K vision-onlymAP: 14.3 -> 17.8Improved
use TG-DP instead of CAV-MAE SyncCAV-MAE Sync · VGGSound attention probingtop-1 accuracy (%): 52.7 -> 52.7No clear effect
add a separate second forward pass for contrastive learning (ablation)CAV-MAE Sync · VGGSound A->V retrievalR@10: 58.1 -> 60.1Improved
add a separate second forward pass for contrastive learning (ablation)CAV-MAE Sync · AS20K attention probingmAP: 30.5 -> 30.4No clear effect
lower contrastive-branch mask ratio from 0.75 to 0.50 (ablation)TG-DP · VGGSound V->A retrievalR@1: 27.1 -> 31.3Improved
lower contrastive-branch mask ratio from 0.75 to 0.50 (ablation)TG-DP · AS20K attention probingmAP: 30.7 -> 32.0Improved
use no masking (0.00) instead of 0.50 in the contrastive branch (ablation)TG-DP · AS20K attention probingmAP: 32.0 -> 29.6Hurt
add teacher-student distillation (ablation)TG-DP · AS20K attention probingmAP: 30.5 -> 32.0Improved
add teacher-student distillation (ablation)TG-DP · VGGSound V->A retrievalR@1: 29.5 -> 31.3Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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