Google DeepMind and Google Research have open sourced the WeatherNext 2 and WeatherNext Cyclones AI models, which achieved state-of-the-art accuracy in predicting tropical cyclone tracks, intensity, and wind structure. Published in Nature, the work demonstrates that these models provide forecasters with an extra day of predictive accuracy compared to prior systems.
OpenAI's internal Astra model has generated solutions to ten long-standing open problems across mathematics and theoretical computer science, with the arguments formalized in Lean. These results span high-dimensional geometry, coding theory, group theory, and quantum complexity, addressing questions that had seen no progress for at least a decade.
Researchers propose ROSA, a robotics foundation model serving system designed for robot factories that moves beyond single-robot, edge-computing assumptions. The system utilizes shared GPU-pool serving to allow fleets of robots to access server-class GPUs over the network.
Researchers analyze gameplay videos from the competitive shooter game Delta Force to extract and understand emergent non-verbal communication, such as gestures and movement patterns. This work addresses a gap in prior studies that have largely focused on MOBA games or verbal coordination in other shooters.
Researchers at NVIDIA are addressing the difficulty of scheduling competitive group play and recreating lab environments by adopting field experiment methodology. This approach integrates experimentation directly into tournament structures, creating hybrid tournament-experiments.
Researchers introduce The AI Telco Engineer (AITE), a framework that uses large language model-driven evolutionary search to autonomously design algorithms for complex communication problems while navigating performance-complexity tradeoffs.
Researchers developed an immersive virtual reality task to assess pedestrian collision detection and avoidance in individuals with cerebral visual impairment (CVI) compared to control participants. The study tracked eye gaze, locomotor, and behavioral responses as subjects walked through a simulated shopping mall populated with crowds of varying densities.
OpenAI has released results from its unreleased internal model, Astra, which successfully solved ten significant open problems in mathematics. The solutions were generated by the model and subsequently formalized into Lean certificates by human researchers.
OpenAI has released solutions to ten mathematical and theoretical computer science problems that had seen no progress on their main results for at least a decade, using an internal version of its next major model, Astra.
Moonshot AI has introduced Kimi K3, an open-source Mixture-of-Experts model featuring 2.8 trillion total parameters and 104 billion activated parameters per token. The model supports native vision capabilities and a 1-million-token context window, leveraging Kimi Delta Attention and Stable LatentMoE to achieve approximately 2.5x scaling efficiency over its predecessor, Kimi K2.
Moonshot AI introduces Kimi K3, an open-source Mixture-of-Experts model featuring 2.8 trillion total parameters and 104 billion activated parameters per token. The architecture utilizes native vision capabilities and supports a 1-million-token context window, enabled by Kimi Delta Attention and Stable LatentMoE.
The U.S. Department of Energy has launched the Genesis Open Models Initiative and, in partnership with Arcee, unveiled its first open-weight model for scientific research named Genesis-Science-1.
Researchers propose Unsupervised On-Policy Self-Distillation (U-OPSD), a method that achieves post-training improvement for large language models using only the model's own generations without external supervision. The approach samples multiple rollouts to construct a pseudo-solution via majority vote, then distills a teacher distribution into prefixes of the model's longest incorrect completion.
Researchers introduce M$^3$R-Bench, a unified benchmark containing 1,000 image-text instances with human-verified annotations designed to evaluate evidence-grounded multimodal metaphor understanding. The benchmark provides joint annotations for metaphor occurrence, Target-Source mapping, sentiment, and stage-wise explanations based on Conceptual Metaphor Theory.
Researchers present Argus, a persistent, self-evolving agentic runtime designed for long-horizon reasoning that separates stable user intent from operational objectives. The system utilizes Manager, Planner, Engineer, and Reviewer roles to execute bounded missions over durable project state, allowing for autonomous execution between operator-owned escalation points.
Microsoft, Shanghai Jiao Tong University, Tongji University, and Fudan University researchers developed SkillOpt, a text-space optimizer that trains natural-language skill documents while keeping the target model frozen. The system uses an optimizer model to propose bounded edits based on scored rollouts, exporting the result as a single `best_skill.md` file.
A developer has open-sourced a Decay-Gated O(N) Causal Linear Attention architecture featuring fused Triton/CUDA state accumulation kernels designed to bypass quadratic MHA bottlenecks.
Qwen introduces Qwen-CUA, a native computer-use agent built on a 397B-A17B mixture-of-experts backbone that operates via screenshots and input events without relying on DOM trees or accessibility metadata. The system utilizes a scaffold to maintain up to 20 active screenshots and was trained using a cloud rollout fleet with nearly 100,000 vCPUs across approximately 40,000 verifiable tasks.
Douyin has released the technical report for its Multimodal Embedding (DME) model, which combines large-scale contrastive pre-training with latent reasoning to achieve strong discrimination and efficiency.
Researchers have introduced Qwen-CUA, a native computer-use agent built on a 397B-A17B Qwen mixture-of-experts backbone. The system operates by observing screenshots and acting through keyboard and mouse events, without relying on DOM trees or accessibility metadata.