Research map / Multi-agent systems and games / Human-AI and goal-conditioned
How2comm: Communication-Efficient and Collaboration-Pragmatic Multi-Agent 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 |
|---|---|---|---|
| add spatial query in message filtering (ablation) | How2comm · DAIR-V2X | AP@0.7: 46.33 -> 47.18 | Helped |
| add channel query in message filtering (ablation) | How2comm · DAIR-V2X | AP@0.7: 46.52 -> 47.18 | Helped |
| add mutual information supervision (ablation) | How2comm · DAIR-V2X | AP@0.7: 46.04 -> 47.18 | Helped |
| add scale matrix in delay compensation (ablation) | How2comm · DAIR-V2X | AP@0.7: 46.86 -> 47.18 | Helped |
| add self-supervised flow training (ablation) | How2comm · DAIR-V2X | AP@0.7: 45.77 -> 47.18 | Helped |
| add temporal cross-attention (ablation) | How2comm · DAIR-V2X | AP@0.7: 45.92 -> 47.18 | Helped |
| add exclusive spatial attention (ablation) | How2comm · DAIR-V2X | AP@0.7: 45.06 -> 47.18 | Helped |
| add common spatial attention (ablation) | How2comm · DAIR-V2X | AP@0.7: 46.63 -> 47.18 | Helped |
| use adaptive late fusion instead of pixel-wise addition (ablation) | How2comm · DAIR-V2X | AP@0.7: 46.77 -> 47.18 | Helped |
| decouple exclusive and common features (ablation) | How2comm · DAIR-V2X | AP@0.7: 45.36 -> 47.18 | Helped |
| keypoint number 9 vs 6 (ablation) | How2comm · DAIR-V2X | AP@0.7: 46.05 -> 47.18 | Helped |
| keypoint number 9 vs 12 (ablation) | How2comm · DAIR-V2X | AP@0.7: 46.57 -> 47.18 | Helped |
Closest papers by meaning
- CoSDH: Communication-Efficient Collaborative Perception via Supply-Demand Awareness and Intermediate-Late Hybridization (CVPR 2025 · similarity 0.90)
- Pragmatic Heterogeneous Collaborative Perception via Generative Communication Mechanism (NeurIPS 2025 · similarity 0.89)
- An Extensible Framework for Open Heterogeneous Collaborative Perception (ICLR 2024 · similarity 0.88)
- Multi-agent Collaborative Perception via Motion-aware Robust Communication Network (CVPR 2024 · similarity 0.88)
- What Makes Good Collaborative Views? Contrastive Mutual Information Maximization for Multi-Agent Perception (AAAI 2024 · similarity 0.87)
- Communication-Efficient Collaborative Perception via Information Filling with Codebook (CVPR 2024 · 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.