Source-checked AI news, every morning. Each day brings six articles and one digest on new models from labs worldwide, research, essays, local models and agents, and prompting techniques. Every article ends with a verification table that links the primary sources.
How we work · Archive · RSS feed · For AI assistants
Other editions: Slovenčina · Nederlands · Français · Deutsch · Español · 简体中文
The open-weight decision model answers bounded questions in one forward pass and returns probabilities. Its strongest speed and accuracy figures come from H2O.ai's own tests.
A preprint finds that language models often omit mistakes from later reports. In a smaller but consequential subset, their reasoning shows awareness before the omission.
A preprint separates a reversible loss of access from the slower destruction of stored facts. The distinction could change how continual-learning failures are diagnosed.
MOLT Sinai routes software work through specialised agents, mandatory plans and adversarial review. The company reports seven reverts over five months.
The open-source assistant combines chat, editable memory, tasks and reminders inside one user's Cloudflare account. Its deployment is private, but not local.
A runnable experiment turns Qwen3-1.7B into a single-pass classifier, then shows that its confidence is too high. Temperature scaling improves the probabilities.
The rest of today's AI news: cognitive research, open-source agents, inference engineering, community measurements, talks, safety proposals and business changes.
The 5-billion-parameter model turns documents, images, audio and video into one structured JSON format. Tencent has released weights, code and a technical report.
A controlled study finds that faithful reports about learned preferences reuse more of the computation behind a model's decisions. The result suggests a white-box test for grounding, within narrow limits.
A 300-million-parameter transformer trained without natural language adapts to six real languages inside its context window. The experiment isolates learning strategy from memorised text.
The MIT-licensed engine narrows its target to one Qwen model and NVIDIA's newest consumer GPUs. It can spill mixture-of-experts weights into system memory or a second card.
A public benchmark finds that one instruction can make several models write machine-readable records with roughly half as many output tokens. The technique fails when applied during active reasoning.