Beyond Domain-Specific World Models: JEPA-Anything Uses 1 Recipe for 7 Fields
JEPA-Anything splits a JEPA's single latent target into 4 orthogonal factors, each with its own predictor. Tested across 7 domains, it beat matched JEPA baselines on all 10 dynamics tasks and cut Interventional Pong intervention error by 34.8%.
The post Beyond Domain-Specific World Models: JEPA-Anyt...
Training Numerical Intelligence via Auto-Diagnosis and Skill Discovery
arXiv:2610.03872v1 Announce Type: new
Abstract: AI agents are becoming increasingly capable of generating scientific code, but generating code is not the same as improving the algorithms behind it. For numerical solvers, execution feedback can expose poor performance, but rarely reveals its underly...
REACT: Physically and Chemically Consistent Reconstruction of Marine Active Tracers
arXiv:2610.03888v1 Announce Type: new
Abstract: Reconstructing global sea surface pH from sparse observations is critical for monitoring ocean acidification and understanding marine carbon cycling. Traditional assimilation and inverse models are physically grounded but costly for large-scale recons...
Proxy Confidence: Auditing Black-Box LLM Agents with a Surrogate's Log-Probabilities
arXiv:2610.03894v1 Announce Type: new
Abstract: A deployed LLM agent emits tool calls, queries, and code that can be silently wrong -- by the time the error surfaces, the action has run. Frontier chat APIs hide the model's token probabilities; the agent's stated confidence barely beats chance on th...
SGAnalog: An End-to-End Circuit Benchmark from Open-Source Silicon Tapeouts
arXiv:2610.03934v1 Announce Type: new
Abstract: Existing analog integrated circuit design benchmarks make two questions hard to answer: whether a model has learned transferable circuit skills rather than recalled familiar examples, and whether its output works under defined process and test conditi...
arXiv:2610.03938v1 Announce Type: new
Abstract: Agentic multimodal large language models (MLLMs) have recently pushed the frontier of visual reasoning by calling tools such as zooming and tagging. Despite the recent strong success of agentic MLLMs, this work uncovers a critical safety failure in th...
AI Weekly Issue #535: AI got too expensive, so companies are moving to cheaper models
This Applied AI edition looks at what organizations actually did with AI over the past three weeks. Three patterns stand out: companies switching models to control cost, large deployments that are now in daily use, and a growing list of places where AI was restricted or removed.
RISED: Rubrics for Agentic Multi-Environment Selection and Self-Distillation
Training a single LLM agent jointly across diverse interactive environments has attracted increasing attention as a route to generalist agents. Existing curriculum and data-selection strategies often allocate training at the environment level or prioritize local reward-based signals, without explici...
Meet Together Link: A Free CLI That Runs Open Models Like Kimi K3 and GLM 5.3 Inside Claude Code, Codex, and OpenCode
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. One install command sets it up, and an Auto router picks a model for each session. Together claims savings...
Building a Streaming Robotics Learning Pipeline Using NVIDIA Cosmos3-DROID
Discover how to construct an end-to-end streaming robotics learning pipeline using the NVIDIA Cosmos3-DROID dataset without local downloads, leveraging byte-range Parquet reads, behavior cloning, and temporal ensembling.
The post Building a Streaming Robotics Learning Pipeline Using NVIDIA Cosmos3-D...
Reflection AI Introduces Beam: A 501B Open-Weight MoE Model With 23B Active Parameters for Coding and Agentic Workloads
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. Reflection says it matches GLM-5.2 on reasoning with 3 to 4x less inference compute. Apache 2.0 weights are due later in October...
Reflection debuts Beam, an open-weight AI model to rival Chinese models at lower compute cost
Nvidia-backed Reflection AI unveils Beam, its first open-weight model, which it says rivals GLM-5.2 on reasoning with far less inference compute. Weights are due this month.
Instinct brings its AI agent to group chats, even for friends without an account
Instinct is launching group chats that let friends use its AI agent together for tasks like planning trips, organizing carpools, and coordinating events. The company says personal accounts remain separate, with permission required before personal agents share information or take action.
For all the data that AI systems continually amass and analyze, enterprise AI agents often suffer from a curious shortcoming: a lack of knowledge. More than data, knowledge is the understanding of what the data means in the context of individual organizations. AI agents need this understanding to re...
Open or closed AI? How founders are choosing what to build on at TechCrunch Disrupt 2026
Learn how founders are choosing between building on open or closed AI at TechCrunch Disrupt 2026. Register now to save up to $100 and get a second pass at 50% off.
Bringing predictive analytics to the agentic AI era
In 2026, the question for enterprise AI is no longer whether predictive models can outperform statistical forecasts—that argument is settled. The big question now is how to enable predictive systems to act on their own conclusions without drifting from business intent. The frontier has moved from pr...
Building an Enterprise AI Customer Support Platform with Claude Fable 5.1 and Claude Code
Most AI coding demos stop at task managers, weather apps, or simple chatbots. For this project, we take on something more demanding: building an enterprise customer-support platform that can investigate complaints, retrieve relevant policies, recommend resolutions, and keep risky actions behind huma...