Sources

Latest research, engineering notes, and product updates from selected AI sources.

Ai2 News

152 articles · updated 2d ago

Ai2 and IUCN to provide AI technology and expertise at no cost to accelerate and strengthen marine conservation in developing countries Ai2’s AI for the Environment teams innovate toward a sustainable future Applying theory of mind: can AI understand and predict human behavior?Clinical NLP using Olmo

Explore more

Alibaba Cloud AI

115 articles · updated 15m ago

Prudential Launches Brand New AI Underwriter Powered by Alibaba Cloud· 15m Speculative Decoding: Turning LLM Inference from One-Token-at-a-Time into a Systems Optimization Problem· 4h From Incubation in OpenAnolis to Upstream Contribution: SGLang Tracing and AI Agent Tuning Practices E-Commerce Bench: Long-Horizon Operations, Multi-Dimensional Evaluation

Explore more

Amazon Science Blog

14 articles · updated 1w ago

Developing provably correct Rust code with Verus When LLM judges agree, should we believe them?SOP-Bench: A new benchmark for evaluating AI agents on real business procedures A decade of mathematical certainty: Reflections on the Automated Reasoning Group

Explore more

Amazon Science Publications

103 articles · updated 4h ago

FinLCR: Scaling financial long-context reasoning via curriculum reinforcement learning· 4h DODO: Discrete OCR diffusion models· 4h SEGRA: A structured experience guided reasoning agent for property graph question answering· 4h AgenTwin: An end-to-end framework for building agentic twins of mobile users

Explore more

AMD Research Publications

6 articles · updated 2mo ago

DMA-Latte: Expanding the Reach of DMA Offloads to Latency-bound ML Communication SIRA: Scaled-Integer Range Analysis for Optimizing FPGA Dataflow Neural Network Accelerators MixQuant: Pushing the Limits of Block Rotations in Post-Training Quantization Cloud-Hardware Co-Design for Memory Bandwidth-Bound HPC Workloads: Performance and Characteristics of Azure HBv5 Virtual Machines

Explore more

AMD ROCm Blog

50 articles · updated 20h ago

veRL on AMD: Production-Ready RL Post-Training on ROCm· 20h Efficiently Serving NVFP4 Models on AMD Instinct™ MI350X/MI355X Accelerators via Online NVFP4 to Quark MXFP4 Requantization· 20h Enabling DeepSeek-V4-Flash Training on AMD Instinct MI355X GPUs with Primus Optimizing ATOM and vLLM-ATOM for High-Interactivity Inference

Explore more

Anthropic Engineering

6 articles · updated 2mo ago

How we contain Claude across products An update on recent Claude Code quality reports Scaling Managed Agents: Decoupling the brain from the hands How we built Claude Code auto mode: a safer way to skip permissions

Explore more

Anthropic News

34 articles · updated 1w ago

Developing Enterprise Frontier Safeguards with our customers Improving our alignment and security efforts Previewing the Model Hardware Standard Expanding our support for scientists

Explore more

Anthropic Research

21 articles · updated 4d ago

Formalizing Fermat's Last Theorem Automated researchers can reliably mitigate alignment failures Enabling independent research on how people use Claude How Claude is accelerating protein design and analytical chemistry

Explore more

Anyscale Blog

20 articles · updated 20h ago

Ray Summit 2026: Physical AI, RL, and the infrastructure that runs them all· 20h Scaling Ray for AI workloads to 10k node clusters Optimizing LLM Serving Efficiency: Moving Beyond KV Cache Reuse to Token-Load Awareness with Ray Serve LLM Introducing Ray History Server: Post-Mortem Observability for Ray on Kubernetes

Explore more

Apple Machine Learning Research Highlights

6 articles · updated 2mo ago

Introducing the Third Generation of Apple’s Foundation Models ParaRNN: Large-Scale Nonlinear RNNs, Trainable in Parallel Apple Machine Learning Research at ICLR 2026 Apple Machine Learning Research at NeurIPS 2025

Explore more

Apple Machine Learning Research Publications

75 articles · updated 6d ago

REFACTOR-VLA: Unsupervised Library Learning of Typed Motor Programs Agent Seer: Synthesizing Scenarios from Specification Understanding LLMs Are Not (Consistently) Bayesian: Quantifying Internal (In)consistencies of LLMs’ Probabilistic Beliefs From Preferences to Principles: Rubric-Based Alignment for Grounded Knowledge Answers

Explore more

Arm Newsroom

26 articles · updated 1d ago

The agentic era needs a computing platform everywhere – Arm is building it Arm introduces new AI-native compute platform built for agentic AI and mobile graphics Arm expands AI infrastructure for the agentic era with AGI CPU and Neoverse CSS N4 Arm brings the ecosystem together to build and define the next phase of physical AI

Explore more

AWS Machine Learning Blog

178 articles · updated 12h ago

Take on your most ambitious work with GPT-6 Astra on Amazon Bedrock· 12h Pathway’s brain-inspired architecture development on Amazon SageMaker HyperPod· 15h Amazon SageMaker Feature Store introduces UpdateRecord for feature-level writes· 15h Govern models with MLflow and Amazon SageMaker AI Model Registry sync: Part 2· 18h

Explore more

Baseten Blog

39 articles · updated 12h ago

Optimizing delta weight syncs for managed rollouts· 12h Baseten leads Coval’s voice AI benchmark· 12h New MCP and skill for coding agents to use Baseten Best open-source models for post-training

Explore more

Baseten Research

7 articles · updated 1mo ago

Can a Language Model Learn Facts Continually in Its Weights?Post-Training Science for Supervised Fine-Tuning Still: Amortized KV Cache Compaction in a Single Forward Pass Post-training frontier legal agents with Baseten Research

Explore more

BentoML Blog

6 articles · updated 2mo ago

The Best Open-Source LLMs in 2026 ChatGPT Usage Limits: What They Are and How to Get Rid of Them The Complete Guide to DeepSeek Models: V3, R1, V4 and Beyond What is GPU Memory and Why it Matters for LLM Inference

Explore more

Bielik AI Blog

1 articles · updated 2mo ago

Bielik.AI Showcases Open European Sovereign AI Model at NVIDIA GTC

Explore more

Cerebras Blog

14 articles · updated 1w ago

How Cerebras serves GPT-5.6 Sol at up to 750 tokens per second Ultrafast Frontier Inference: Cerebras Deep Dive at Hot Chips 2026 Introducing Cerebras CS-4: The Fastest AI Just Got Faster. Built for Hyperscale.Introducing Cerebras CS-4: The Fastest AI Just Got Faster

Explore more

Cerebras Publications

6 articles · updated 1mo ago

Crystal: Illuminating LLM Abilities on Language and Code REAP the Experts: Why Pruning Prevails for One-Shot MoE Compression Scaling with Collapse: Efficient and Predictable Training of LLM Families Predicting Training Re-evaluation Curves Enables Effective Data Curriculums for LLMs

Explore more

Cerebras Whitepapers

6 articles · updated 1mo ago

How Cerebras Solved the Wafer-Scale Yield Challenge Cerebras Inference: Build Products that Others Can’t with the Fastest AI Infrastructure Cerebras Wafer-Scale Cluster Training Large Language Models on Cerebras Wafer Scale Engine

Explore more

Chroma Updates

27 articles · updated 2d ago

Agent Swarms are a Distributed Systems Problem Technical Report: Context-1 Distributed Chroma: Bring Your Own Cloud Mintlify powers documentation search with Chroma Cloud

Explore more

Cisco AI Blog

17 articles · updated 5d ago

Your Agent Trusts Things You Never Approved The “U.S. vs. China” AI Trap: An Incomplete Proxy for AI Security Cisco AI Defense + Armada: Distributed AI That’s Safe to Run, Wherever the Mission Requires It A new chapter for IT operations: Cisco Cloud Control is now generally available for U.S. customers

Explore more

Claude Blog

70 articles · updated 18h ago

Reducing cost and improving performance with Claude Platform· 18h A guide to the anatomy of effective commerce agents Building commerce agents with Claude How Anthropic employees use Claude Tag

Explore more