Research map / Vision-language and multimodal / Cross-modal learning
Collecting Cross-Modal Presence-Absence Evidence for Weakly-Supervised Audio-Visual Event Perception
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| use cross-modal presence-absence evidence collection with joint-modal mutual learning | JoMoLD on LLP (AVVP) · LLP (AVVP) | event-level visual F-score: 59.9 -> 63.7 | Helped |
| use cross-modal presence-absence evidence collection with joint-modal mutual learning | JoMoLD on AVEP · AVEP | event-level visual F-score: 55.8 -> 61.9 | Helped |
| use CMPAE instead of JoMoLD | AVE · AVE | accuracy: 71.8 -> 74.8 | Helped |
| add PAEC to the EDL baseline (ablation) | AVVP and AVEP · AVVP and AVEP | segment-level Type F-score: 60.8 -> 61.9; 58.0 -> 61.5 | Helped |
| add JML while leaving out PAEC (ablation) | EDL baseline · AVVP and AVEP | segment-level Type F-score: 60.8 -> 61.0; 58.0 -> 58.9 | Helped |
| add JML to EDL + PAEC (ablation) | CMPAE · AVVP and AVEP | segment-level Type F-score: 61.9 -> 63.3; 61.5 -> 62.4 | Helped |
| use a single track for both presence and absence evidence (ablation) | CMPAE · AVVP | segment-level Type F-score: 63.3 -> 61.2 | Hurt |
| use cross-modal features for both presence and absence evidence (ablation) | CMPAE · AVVP | segment-level Type F-score: 63.3 -> 61.7 | Hurt |
| exchange feature sources for presence and absence evidence (ablation) | CMPAE · AVVP | segment-level Type F-score: 63.3 -> 62.1 | Hurt |
| remove the joint-modal uncertainty term from JML (ablation) | CMPAE · AVVP | event-level Type F-score: 57.4 -> 56.5 | Hurt |
| remove the uni-modal uncertainty term from JML (ablation) | CMPAE · AVVP | event-level Type F-score: 57.4 -> 56.4 | Hurt |
| replace the uni-modal selector with uniform averaging (ablation) | CMPAE · AVVP | event-level Type F-score: 57.4 -> 56.3 | Hurt |
Closest papers by meaning
- Adapting to the Unknown: Training-Free Audio-Visual Event Perception with Dynamic Thresholds (CVPR 2025 · similarity 0.90)
- Dense Audio-Visual Event Localization Under Cross-Modal Consistency and Multi-Temporal Granularity Collaboration (AAAI 2025 · similarity 0.89)
- LongVALE: Vision-Audio-Language-Event Benchmark Towards Time-Aware Omni-Modal Perception of Long Videos (CVPR 2025 · similarity 0.88)
- Audio-Visual Semantic Graph Network for Audio-Visual Event Localization (CVPR 2025 · similarity 0.87)
- Towards Open-Vocabulary Audio-Visual Event Localization (CVPR 2025 · similarity 0.87)
- A Benchmark for Omni-Modal Reasoning in Long Videos (NeurIPS 2026 · similarity 0.87)
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.