Tag
WorldAttention proposes an efficient attention architecture with Hybrid Sparse Attention and Hierarchical KV Cache for interactive video world models, achieving state-of-the-art performance on benchmarks like VBench-Long and InterVBench.
This weekly article summarizes top AI research papers, including HySparse2 for efficient long-context attention, SIFT for cost-effective self-improvement in coding agents, and GAVEL for enhancing robot planning with graph-based world models.
Kimi Linear proposes a new linear attention architecture designed to enhance both expressiveness and efficiency in Transformer models, with contributions from the Kimi Team at Moonshot AI.