We are sharing two candidate magnets for next-generation computer memory, found by a team of Claude Opus 5.5 agents.

Aleph Alpha has released Kolibri, a German-English mixture-of-experts model with 78 billion parameters, about three billion of which are active per token. Over 21 percent of the training data is German.

Together AI has released Together Link, a free, MIT-licensed CLI that connects Claude Code, Codex, OpenCode, Pi, and the Claude and ChatGPT desktop apps to open models like Kimi K3 and GLM 5.3.

Reflection AI has introduced Beam, its first open-weight model. It is a 501B sparse Mixture-of-Experts model with 23B active parameters, built for coding and agentic work.

One transformer ran candidate generation and ranking in Yandex Music's A/B test without hand-engineered features, lifting likes 11.42%.

67 Countries Launch 184 Sovereign AI Projects Amid Global Race    조선일보

Over the summer I talked to the CEO of Springboards, a startup building an LLM that’s designed to come up with a wider variety of responses than its mainstream rivals do.

Voters are already voting in the midterms. This year, some voters are trying something new to get ready for the election: asking AI to help research their ballot and even decide who to vote for.

Cantina Security, with Yeta Labs, has released apex-flash-1, an open-weights model trained specifically for vulnerability research.

A long-standing goal of AI is a model that can continually learn and improve itself. On post-trained models, supervised finetuning (SFT) on new data often causes poor generalization and catastrophic forgetting.

Feature pyramid methods, from FPN to BiFPN, have achieved strong performance in face detection by fusing multi-scale features.

High-quality child care in early life is a critical determinant of children's growth and development. Research on child care quality has been constrained by fragmented, non-research-friendly, and privacy-bound datasets.