Please confirm you are human
This browser or connection looks automated. Press and continuously hold the control for 3 seconds to enable Google-hosted web results and, when separately allowed, AI-assisted answers.
A successful check enables 100 search requests. Interactive access does not authorize scraping, systematic collection, or reuse of search output.
News
Evaluating Multimodal Vision Models with Moonshot PerceptionBench Using Robust Data Loading and Automated Judging
1+ hour, 28+ min ago (767+ words) In this tutorial, we design an end-to-end evaluation workflow for PerceptionBench. This multimodal benchmark measures fine-grained visual perception capabilities across tasks such as OCR, counting, localization, contextual reasoning, comparison, depth understanding, and hallucination detection. We begin by configuring a Colab-compatible…...
How to Secure AI Agents, MCP Servers, and LLM Apps in Production
3+ hour, 38+ min ago (383+ words) Agents, MCP integrations, and LLM-powered applications are entering codebases faster than most security programs can track them. Mend.io’s new practitioner guide, ‘Securing AI agents, MCP servers & LLM apps: A practical framework’, targets that gap. It is organized around three…...
Alibaba Qwen Releases Qwen3.8-Max: A 2.4 Trillion Parameter MoE Model and the Most Capable One in the Qwen Family to Date
15+ hour, 30+ min ago (23+ words) MarkTechPost...
Cogent AI Team Releases VR-1: A Frontier Cyber Reasoning Model That Composes and Verifies Enterprise Attack Paths
16+ hour, 26+ min ago (214+ words) Not open-sourced or weight. VR-1 is available only to vetted organizations through the Cogent Frontier Access Program, with guardrails, policy controls, and audit logging in place, and participants work directly with Cogent Research on evaluation and deployment in their own…...
Onton Releases Ontology 1: A Neurosymbolic Search Model That is 2.7x More Accurate than the World’s Best E-commerce Search Engines
23+ hour, 5+ min ago (571+ words) Onton, a San Francisco-based search and discovery company, has released Ontology 1, a neurosymbolic model for complex, conversational, multimodal product search. On a 90-query benchmark scored by three independent LLM judges, Ontology 1 reached a mean precision@10 of 0.630, against 0.543 for Google Shopping…...
A Tutorial on GeoAI: Designing Footprint Extraction from NAIP Imagery Using U-Net, Grounding DINO, SAM, and Mask R-CNN
1+ day, 2+ hour ago (275+ words) We configure the environment, install the required GeoAI and deep learning libraries, and verify GPU availability. We define the central configuration parameters for dataset creation, model training, inference, and optional processing stages. We then create the working directory, define reusable…...
Thinking Machines Lab Releases Inkling-Small: A 276B Total, 12B Active Open Weights Multimodal MoE Model
1+ day, 3+ hour ago (313+ words) Yes, and the quantized checkpoint is why. Per the model card, the BF16 checkpoint needs at least 600 GB of aggregated VRAM. That is met by 4x NVIDIA B300 or 8x NVIDIA H200. The NVFP4 checkpoint drops that floor to 180 GB. It runs W4A4 on a single B300, which…...
NVIDIA AI Releases Molt: A PyTorch-Native Agentic Reinforcement Learning Framework
1+ day, 17+ hour ago (313+ words) Agentic reinforcement learning research is constant algorithm modification. New estimators, new pipeline stages, new rollout schemes. In mainstream frameworks each change threads through layers of trainer, distributed backend, and rollout glue. That cost lands on the researcher at every iteration....
End-to-End Forecasting with TimesFM 2.5: Backtesting, Covariates, Anomaly Detection, and Scalable Colab Deployment
1+ day, 18+ hour ago (825+ words) In this tutorial, we build an advanced end-to-end time-series forecasting workflow with TimesFM 2.5. We begin by configuring the runtime, installing the required dependencies, detecting available hardware, and generating a realistic multi-store retail dataset with trend, seasonality, pricing, promotions, holidays, temperature…...
AMD Releases Instella-MoE-16B-A3B: A Fully Open Mixture-of-Experts LLM With 2.8B Active Parameters Trained On Instinct GPUs
2+ day, 4+ hour ago (252+ words) Partly. The weights ship under a ResearchRAIL license for academic and research purposes only, so this is not a drop-in commercial model. The training codebase is MIT licensed, and that is the more reusable asset here. Instella-MoE is a decoder-only…...