ECCV 2026 accepted papers on Multi-agent systems and games
46 of the 2,834 papers accepted at ECCV 2026 (1.6%) are on Multi-agent systems and games, compared with 0.8% at ECCV 2024. They fall into 2 topics.
See ECCV 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
LLM agents · 31 papers
- MolmoWeb: Open Visual Web Agent and Open Data for the Open Web (25 citations)
- DualTAP: A Dual-Task Adversarial Protector for Mobile MLLM Agents (10 citations)
- Learning from Failure: Inference-Time Self-Improvement for Computer-Use Agents (7 citations)
- SlowBA: An efficiency backdoor attack towards VLM-based GUI agents (6 citations)
- ProAct: Agentic Lookahead in Interactive Environments (5 citations)
- Self-Evolving MCP-GUI Agents via Automated Environment Generation and Experience Learning (5 citations)
- Narrative-Driven Paper-to-Slide Generation via ArcDeck (4 citations)
- VDC-Agent: When Video Detailed Captioners Evolve Themselves via Agentic Self-Reflection (4 citations)
- Hybrid Advantage Estimation with Unified Critic for VLM Agentic Reinforcement Learning (3 citations)
- Teaching an Agent to Sketch One Part at a Time (3 citations)
- VERITAS: A Multi-agent Co-scientist for Verifiable Image-Derived Hypothesis Testing (3 citations)
- EpiBench: Benchmarking Multi-turn Research Workflows for Multimodal Agents (2 citations)
and 19 more in the interactive map.
Human-AI and goal-conditioned · 15 papers
- Driver-WM: A Driver-Centric Traffic-Conditioned Latent World Model for In-Cabin Dynamics Rollout (15 citations)
- CoMaTrack: Competitive Multi-Agent Game-Theoretic Tracking with Vision-Language-Action Models (8 citations)
- CausalDrive: Real-time Causal World Models for Autonomous Driving (7 citations)
- SPEAR: A Simulator for Photorealistic Embodied AI Research (2 citations)
- CoMind: Understanding Collaborative Human Activity from Multiple Minds and Views (1 citation)
- HitMem: Hierarchical Temporal 3D Memory with Multi-Modal Context-Aware Retrieval for Dynamic Environments (1 citation)
- Seek to Segment: Active Perception for Panoramic Referring Segmentation (1 citation)
- UECP: Uncertainty-Enhanced Collaborative Perception (1 citation)
- 360CityArena: A Realistic Virtual Urban Navigation Benchmark for Embodied Agents
- ECoSim: Data Efficient Fine-Tuning for Controllable Traffic Simulation
- ECTraj: Enhanced Consistency Training for Multi-Agent Trajectory Prediction
- HERO: Heterogeneous Evidential Robust Object-Level Collaborative Perception
and 3 more in the interactive map.
