Paint What You See: Benchmarking Dexterous Visual Tool Use in Multimodal Agents Paper • 2608.25417 • Published 9 days ago • 17
Qwen-Drive-1.0: An Initial Step towards a Vision-Language Foundation Model for Autonomous Driving Paper • 2609.00111 • Published 4 days ago • 361
Agentic Game Development as a Verifiable Trajectory Data Engine for Scaling World Models Paper • 2608.25518 • Published 9 days ago • 195
JIT-Agent: Scaling Harness Intelligence via Just-in-Time Harness Evolution Paper • 2608.25593 • Published 9 days ago • 115
VGI-Bench: Probing Visual Intelligence in Video Generation Models Paper • 2608.19583 • Published 9 days ago • 179
HarnessEval-W: Agentifying the Evaluation of Visual Worlds Paper • 2608.16859 • Published 18 days ago • 340
Learn What's Left, Not What's Mastered: Saturation Aware Advantage Reweighting for Multi-Reward Policy Optimization Paper • 2608.16072 • Published 18 days ago • 151
DSWorld: A Data Science World Model for Efficient Autonomous Agents Paper • 2607.15901 • Published Jul 17 • 12
Is One Layer Enough? Training A Single Transformer Layer Can Match Full-Parameter RL Training Paper • 2607.01232 • Published Jul 2 • 8
PoseShield: Neural Collision Fields for Human Self-Collision Resolution Paper • 2606.29686 • Published Jun 29 • 6