AI Paper Atlas

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

VGGSounder: Audio-Visual Evaluations for Foundation Models

Daniil Zverev, Thaddäus Wiedemer, Ameya Prabhu, Matthias Bethge, Wieland Brendel, A. Sophia Koepke

ICCV 2025 · Vision-language and multimodal › Cross-modal learning · approach: Audio-visual speech separation · 6 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 human-curated labels to VGGSound ground truth (ablation)Gemini 1.5 Flash · VGGSounder audible labels (A)Delta Hit: +29.28 vs +0.48 for auto labelsImproved
add human-curated labels to VGGSound ground truth (ablation)Gemini 1.5 Pro · VGGSounder audible labels (A)Delta Hit: +28.61 vs +0.31 for auto labelsImproved
add human-curated labels to VGGSound ground truth (ablation)Gemini 2.0 Flash · VGGSounder audible labels (A)Delta Hit: +8.80 vs +0.22 for auto labelsImproved
add visual input to audio-only inputGemini 1.5 Pro · VGGSounder (no background music)Hit: 33.43 -> 78.36Improved
add visual input to audio-only inputUnified-IO 2 · VGGSounder (no background music)Hit: 54.43 -> 63.36Improved
add audio input to visual-only inputUnified-IO 2 · VGGSounder (no background music)Hit: 32.18 -> 63.36Improved
add audio input to audio-only inputCAV-MAE · VGGSounder (no background music)Hit: 67.07 -> 67.51No clear effect
samples with background musicCAV-MAE · VGGSounder audio inputDelta F1: -6.80Hurt
samples with voice overCAV-MAE · VGGSounder audio inputDelta F1: -11.93Hurt
samples with voice overGemini 1.5 Flash · VGGSounder audio inputDelta F1: +23.73Improved
samples with static imagesCAV-MAE · VGGSounder audio inputDelta F1: +4.93Improved
samples with static imagesAV-Siam · VGGSounder video inputDelta F1: -9.69Hurt

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