Archive
Page 9
Llama 4 Scout Released
Meta has released Llama 4 Scout, a 109B parameter model that runs on a single H100 GPU. It outperforms GPT-4o on several benchmarks.
LLMSurgeon: Diagnosing Data Mixture of Large Language Models
The pretraining data mixture of Large Language Models constitutes their digital DNA, shaping model behaviors, capabilities, and failure modes.
Physics Is All You Need? A Case Study in Physicist-Supervised AI Development of Scientific Software
Are AI agents tools, co-authors, or researchers? We present a quantified case study ($N=1$): a physicist supervising an AI coding agent (Claude Code...
VideoMLA: Low-Rank Latent KV Cache for Minute-Scale Autoregressive Video Diffusion
Long-rollout causal video diffusion has converged on a fixed-size sliding-window KV cache, with recent progress innovating within this layout by changing...