Safety
LeAVJEPA: A Minimalist Architecture for Audio-Visual Self-Supervised Learning
arXiv:2610.06226v1 Announce Type: cross Abstract: Prior audio-visual self-supervised learning methods rely on mechanisms such as EMA target encoders, prediction heads, reconstruction decoders, and con
arXiv:2610.06226v1 Announce Type: cross Abstract: Prior audio-visual self-supervised learning methods rely on mechanisms such as EMA target encoders, prediction heads, reconstruction decoders, and contrastive losses. We introduce LeAVJEPA, the first audio-visual encoder trained under LeJEPA's collapse-free objective. A single early-fusion Vision Transformer processes audio, video, and joint audio-video inputs. Modality dropout treats a missing modality as another view of the same event, making cross-modal alignment implicit in the objective. The model aligns global embeddings with modality-specific local embeddings, and SIGReg prevents representational collapse. A controlled ablation identifies modality dropout as the key mechanism for audio-visual alignment. Despite the architectural simplicity, LeAVJEPA reaches 36.0 mAP on AudioSet-20K and 91.3% accuracy on ESC-50 under frozen evaluation. After fine-tuning, it reaches 61.1% accuracy on VGGSound, and its embeddings support zero-shot audio-visual retrieval.
Related
- Semantic Noise Reduction via Teacher-Guided Dual-Path Audio-Visual Representation Learning
- JoVA: Unified Multimodal Learning for Joint Video-Audio Generation and Editing
- When Speech Meets Lips: Interpretable Audio-Visual Synchronization for L2 Pronunciation Assessment
Source: arXiv cs.CV | 2026-10-06