DeLoop: Repeated Actions as Self-Supervision for Long-Horizon Agents Paper • 2610.19376 • Published Oct 14 • 22
DTDD: Divergence-Triggered Dynamic Distillation for Reliable On-Policy Supervision Paper • 2610.19302 • Published Oct 13 • 5
CLASHarness: Benchmarking Multi-Agent Conflict in Computer Use via a Controllable and Verifiable Generation Pipeline Paper • 2610.19128 • Published Oct 13 • 35
Semantic Richness or Geometric Reasoning? The Fragility of VLM's Visual Invariance Paper • 2604.01848 • Published Apr 3 • 5
CUA-Suite: Massive Human-annotated Video Demonstrations for Computer-Use Agents Paper • 2603.24440 • Published Mar 25 • 96
DDiT: Dynamic Patch Scheduling for Efficient Diffusion Transformers Paper • 2602.16968 • Published Feb 19 • 13
Some Modalities are More Equal Than Others: Decoding and Architecting Multimodal Integration in MLLMs Paper • 2511.22826 • Published Nov 28, 2025 • 8
Generative Action Tell-Tales: Assessing Human Motion in Synthesized Videos Paper • 2512.01803 • Published Dec 1, 2025 • 5