SKILLER addresses the model-mismatch problem, where reusable skills designed around powerful frontier models fail when deployed on smaller, cheaper language…
Maglev proposes a way for transformers to maintain persistent, token-wise memory while keeping both inference cost bounded and training parallelizable.…
LycheeMemory V2 targets the cost of maintaining long-term memory in long-horizon LLM agents, replacing conventional turn-by-turn memory consolidation with semantic…
This paper studies why large language models keep generating long, expensive reasoning chains on problems beyond their real capability, producing…
This case study reports on using an AI coding agent to carry out a large-scale architectural refactor of a 717,000-line…
This study investigates whether purely rhetorical changes to a scientific manuscript — with the underlying content held constant — can…
Intern-S2-Preview is a family of foundation models built for multimodal scientific reasoning and long-horizon agentic tasks. Its training pipeline combines…
This write-up documents a training experiment pushing a 0.9-million-parameter language model to a token-per-parameter ratio of roughly 222,000:1 — far…
A new book, “Journalism in the Age of AI” by researchers Rodrigo Zamith, Tomás Dodds, and Seth C. Lewis, examines…
UC Berkeley researchers argue in a Nature commentary that the industry’s push to build ever-larger AI data centers is unnecessary,…
Elon Musk’s xAI raised $20 billion in an oversubscribed funding round that exceeded its initial $15 billion target, with backers…
Anthropic has finalized a $9.1 billion, 20-year data center lease with Bitcoin miner Riot Platforms, securing 191 megawatts of power…
