Kakao released Kanana-2-v-decide, a multimodal model that reads text and images and returns decisions as probabilities, options or scores so AI agents…
Lam Research authorized a record $25 billion share repurchase Thursday, the largest in its 44-year history, after Q1 FY2027 revenue surged 54% to $8.22 billion and operating margin crossed 40% for the first time.
BOX (NYSE:BOX) unveiled new artificial intelligence and security tools at its BoxWorks 2026 investor product briefing, outlining a strategy to help companies connect AI agents with their documents and automate content-heavy workflows.
Democrats' sweeping midterm election victory unleashed a rout of tech stocks on Thursday, with investors worried the party will use its gains in state capitols and the US Congress to impose new regulations that will slow the AI infrastructure buildout that has fueled this year's market gains.
Aina AI Ring is up for preorder at $399, with shipping set for January 2027. See what it does, what we know so far, and whether reserving now makes sense.
Airbnb CEO Brian Chesky said the company plans to develop specialized AI agents for integrations with numerous partners for "every vertical in travel."
AI-related cyberattacks are increasing, with hackers stealing AI access tokens and using AI tools to speed up intrusions. Multiple AI developer platforms have been affected, leading to exposed customer secrets and node credentials.
Kimi K3.1 Pro is an open-weight, natively multimodal agentic model, and our most capable 3T-class model to date. Post-trained from Kimi K3 with large-scale joint reinforcement learning across hundreds of downstream domains, it is the first model in the K3.1 series. With a…
Token Pass credit now applies automatically to included model usage through AI Gateway , whether a request authenticates with an API key, OIDC token, or fx login. No dedicated Token Pass API key is required, though you can still create one from the dashboard or the CLI for keys…
Direct On-Policy Distillation (Direct-OPD) transfers reinforcement-learning-induced policy improvements from a small model to a larger student by using the token-level log-ratio between post-RL and pre-RL checkpoints as dense supervision on the student's own rollouts. This…
Open-source data agent and Python library for querying, transforming, and visualizing any kinds of data - databases, files, knowledge graphs, Hugging Face datasets, and the web.
Repeated evaluation can estimate a benchmark score accurately while still requiring replication to certify narrow uncertainty. We characterize that requirement on a fixed grid of $M$ tasks with $L$ binary paths per task under the hard budget $(M+t)K$, where each path costs at…
Internet texts are replete with format tags that carry structural, semantic, and functional meaning. Current large language model (LLM)-based translation systems struggle to balance translation fluency with tag fidelity when processing tagged text. We argue that resolving this…