From Knowledge Access to Source Learning: Developing Source-Specific Competence Paper • 2610.02150 • Published 8 days ago • 12
DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression Paper • 2609.19969 • Published 22 days ago • 227
JEPA-Anything: Learning Predictive Models across Different Worlds Paper • 2609.20800 • Published 22 days ago • 78
Continual Learning Mechanisms Compose for Long-Horizon Memorization Paper • 2609.06986 • Published Sep 7 • 376
Cadence: Error-Bounded Lossy Compression of Demand Time Series with a Time-Series Foundation Model Paper • 2609.06008 • Published Sep 5 • 19
Select, Compress, Reinvest: A Controlled Study of Visual-Token Allocation in Long-Video MLLMs Paper • 2609.03820 • Published Sep 3 • 16
SemComp-Bench: Benchmarking Semantic Task Completion in Video Generation Paper • 2608.17426 • Published Aug 18 • 161
Zetta ζ: An Efficient Closed-Loop Embodied Harness for Self-Evolving Physical Intelligence Paper • 2608.16590 • Published Aug 17 • 153
Can We Defend Against AI-Generated Video Attacks on Real-World Crisis Events? A Systematic Evaluation of Detectors, Generators and Social Dissemination Paper • 2608.14391 • Published Aug 14 • 287
PlayWorld: Benchmarking World Models with Agent Players over Long-Horizon Objectives Paper • 2608.13552 • Published Aug 13 • 47
OpenART: Scaling Agent Red Teaming via Open-Ended Environment Evolution Paper • 2608.00677 • Published Aug 1 • 266
ChronoVision: Temporal Reasoning via Latent State Reconstruction Paper • 2608.05631 • Published Aug 6 • 40