The Trinity:Capability-Driven Recursive Self-Improvement for Multimodal Agents Paper • 2610.36077 • Published 22 days ago • 161
VisionWeave: Weaving Elastic Visual Representations as a Native Capability of MLLMs Paper • 2610.07987 • Published Oct 6 • 60
Receiver-Conditioned Latent Communication gives 94% CacheBack Paper • 2609.32046 • Published Sep 25 • 8
VākQA: A Benchmark and Evaluation Study for Telugu Spoken Factoid Question Answering Paper • 2609.19879 • Published Sep 17 • 4
Continual Learning Mechanisms Compose for Long-Horizon Memorization Paper • 2609.06986 • Published Sep 7 • 49
FreeFlow: A Bias-free Hierarchical Transformer for Optical Flow Estimation Paper • 2609.11486 • Published Sep 10 • 14
Multi-Grid Post-Training for Long-Form Multi-Shot Video Generation Paper • 2609.06373 • Published Sep 6 • 4
Encoded Early, Used Late: Where Transformers Begin to Act on an Inferred Partner's Expertise Paper • 2609.07139 • Published Sep 7 • 5
SemComp-Bench: Benchmarking Semantic Task Completion in Video Generation Paper • 2608.17426 • Published Aug 18 • 48