Can an LLM learn how an optimizer behaves -- and use that knowledge to control it? We extend Code World Models (CWMs), LLM-synthesized Python programs that pred...

markdown !Cover image for Resisting the Eye of the Machine: A Reflection on AI and Data Ownershiphttps://media2.dev.to/dynamic/image/width=1000,height=420,fit=c...

Question To become an expert in AI system design and engineering, which book should a student of software engineering read? I'm not referring to complex mathem...

Test-time training (TTT) with KV binding as sequence modeling layer is commonly interpreted as a form of online meta-learning that memorizes a key-value mapping...

Visual reinforcement learning is appealing for robotics but expensive -- off-policy methods are sample-efficient yet slow; on-policy methods parallelize well bu...

We report the performance of Aletheia (Feng et al., 2026b), a mathematics research agent powered by Gemini 3 Deep Think, on the inaugural FirstProof challenge. ...

Embodied LLMs endow robots with high-level task reasoning, but they cannot reflect on what went wrong or why, turning deployment into a sequence of independent ...

Efficiently processing long sequences with Transformer models usually requires splitting the computations across accelerators via context parallelism. The domin...

We study the complexity of smoothed agnostic learning, recently introduced by~cite{CKKMS24}, in which the learner competes with the best classifier in a target ...

Pass@k is a widely used performance metric for verifiable large language model tasks, including mathematical reasoning, code generation, and short-answer reason...

Uniform-state discrete diffusion models excel at few-step generation and guidance due to their ability to self-correct, making them preferred over autoregressiv...

Deep learning has significantly advanced automated brain tumor diagnosis, yet clinical adoption remains limited by interpretability and computational constraint...