Research map / Multi-agent systems and games / Human-AI and goal-conditioned
From Discriminative to Generative: A Diffusion-Based Paradigm for Multi-Agent Collaborative Perception
Kexin Gong, Puyi Yao, Guiyang Luo, Quan Yuan, Tiange Fu, Hui Zhang, Jinglin Li
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add diffusion-based generative supervision (ablation) | DiGS-CP two-stage fusion · OPV2V | AP@0.7: 89.07% -> 93.53% | Improved |
| add generative supervision (ablation) | DiGS-CP two-stage fusion · OPV2V | AP@0.5: 95.35% -> 96.75% | Improved |
| use cross-attention conditioning instead of mean-based conditioning (ablation) | DiGS-CP diffusion U-Net · OPV2V | AP@0.7: 92.39% -> 93.53% | Improved |
| use cross-attention conditioning instead of concatenation (ablation) | DiGS-CP diffusion U-Net · OPV2V | AP@0.7: 92.63% -> 93.53% | Improved |
| add distance-weighted fusion instead of confidence-only weighting (ablation) | DiGS-CP · OPV2V | AP@0.7: 93.12% -> 93.53% | Improved |
| train with generative supervision (cross-domain OPV2V to V2XSet) | DiGS-CP · V2XSet cross-domain | AP@0.7 drop: 0.8301 -> 0.8199 (drop 1.02%) vs baselines dropping 12.12% to 17.83% | Improved |
| set confidence threshold tau to 0.2 rather than 0.1, 0.3 or 0.4 (ablation) | DiGS-CP stage 1 · OPV2V | AP@0.3/0.5/0.7 | Mixed |
| increase diffusion steps T from 500 to 1000, then beyond (ablation) | DiGS-CP diffusion model · OPV2V | AP@0.3/0.5/0.7 | Mixed |
| DiGS-CP vs CoSDH on OPV2V AP@0.5 | DiGS-CP · OPV2V | AP@0.5: 96.83% -> 96.75% | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- CoSDH — CoSDH: Communication-Efficient Collaborative Perception via Supply-Demand Awareness and Intermediate-Late Hybridization (CVPR 2025) · better on 3 of 4 shared results
- Where2comm — Where2comm: Communication-Efficient Collaborative Perception via Spatial Confidence Maps (NeurIPS 2022) · better on 4 of 4 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
- An Extensible Framework for Open Heterogeneous Collaborative Perception (ICLR 2024 · similarity 0.88)
- Generative Map Priors for Collaborative BEV Semantic Segmentation (CVPR 2025 · similarity 0.88)
- One is Plenty: A Polymorphic Feature Interpreter for Immutable Heterogeneous Collaborative Perception (CVPR 2025 · similarity 0.88)
- Privacy-Preserving V2X Collaborative Perception Integrating Unknown Collaborators (AAAI 2025 · similarity 0.88)
- DSRC: Learning Density-Insensitive and Semantic-Aware Collaborative Representation Against Corruptions (AAAI 2025 · similarity 0.87)
- CoSDH: Communication-Efficient Collaborative Perception via Supply-Demand Awareness and Intermediate-Late Hybridization (CVPR 2025 · 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.
