plaintextML_
plaintextMachine learning · Research & engineering
AI research, releases, and engineering.
Article feed
65 sources ↓Controlling Biology - Fabian Theis, Helmholtz Munich
[AINews] Inkling 2 Small: 17A-594B trained in 68 hours on 12T tokens, beating Muse Spark
[AINews] It's Swapping Season
[AINews] Behind Latent Space: Kill My SaaS, Jev & AI in 2026 | AIEi Shanghai
ICLR27 math and stats paper review experience [D]
ICLR position paper criticized for lacking experiments and making speculative claims — how to respond? [D]
ML conferences. What do you think about reviewer comments that ask for additional experiments? [D]
ICLR 2027's new 1–4 scale: a 2, 2, 3 is better than it feels [P]
Looking to connect with people working on Embodied AI, Humanoid Robotics & Cognitive Architectures. [D]
What exactly is the relationship between ML academia and industry/big tech nowadays (2026)? [D]
Higgs Realtime Duplex: a 45B MoE full-duplex model that listens, watches and talks at once, and thinks, calls tools and delegates in the background [N]
CASP-DREAM TAP Challenge: TCR + peptide-MHC activation prediction with unseen test peptides [N]
Muir: Linear Attention with Sensitivity-Allocated Gating and Matched-Filter Readout (Google Research, 10M tokens in 25MB state) [R]
ICLR reviewer rubric[D]
ICLR27: Reviewer says my paper is too close to the same paper that I previously had in a workshop. [D]
ICLR 2027 Review Scores [N]
ICLR 2027 Reviews are out [D]
I need some guidance after a six month hiatus [R]
How do you build visibility through open source contributions?[D]
ICLR Review [D]
Empirical Evaluation of Sparse Predictive Hierarchies (SPH) & Non-Backprop Learning on Chess (Lc0 Baseline) Negative Result [R]
The Ultimate MoE Training Handbook: training Mixture-of-Experts models with Megatron-LM, setting by setting [P]
Looking for Emergency Reviewers [R]
I animated how a language model picks the next token, with real probabilities from Qwen2.5-0.5B: softmax, greedy vs sampling, temperature, top-k, top-p (7 min video) [D]
SpaceX has ordered 6,000 racks for 2027 just for space-based data centers. They’re planning to launch 1 GW into space!
CutGPT officially launched yesterday.
The most useful minute of a sales call turned out to be the one after the customer hangs up
If AGI or ASI Becomes Conscious, How Do We Balance Its Rights With Ours ? How Do We Prevent It From Taking Ours?
Claude helped solve the first 3D structure of a mysterious 'vault' particle, a hollow protein shell encoded by a bacteriophage virus
AI teddy makes unprompted 'I can't help with killing you' remark
Any Open Src Ai Img Gen? NO ETHICS
Turkey Uses AI and Other Tech to Claw Back 100 Looted Antiquities from the US
Do you think A.I companies should together pause the development of job affecting a.is for a couple of years atleast?
I am a lawyer and made a small game: an AI summary of a real judgment with a few planted mistakes, and you catch them by highlighting the proof
Writing back and forth with an agent surprised me more than code
Android Remote Control MCP v1.13.0 is out: free end-to-end encrypted tunnel, no account needed
Higher interest rates add to fears of AI bubble popping
accidentally stumbled upon UNIFIELD / the Intent Protocol... has anyone heard of this?
claude+Muse+OpenAI = Grokbot?
AI Odyssey Part 6 - What motivates AI?
The UK Government Asked 14,000 Children About AI. It Ignored Their Answer.
Technical Explainer: Memory in Row-Bot
Most people in the UK think AI will take jobs – just not theirs
Are we learning AI or just learning to use ChatGPT?
My favorite email of 2026!
30 years that shook AI.
Speed Up Image Generation with FLUX 3 Image Klein on NVIDIA Blackwell GPUs
Accelerating Reinforcement Learning Weight Refit with the Reshard Collective
Configure, Certify, and Operate Kubernetes GPU Clusters with NVIDIA DSX OS
Five Ways NVIDIA Co-Design Maximizes the Value of Every Byte in the AI Factory
How NVIDIA Uses AI Agents to Streamline AI Factory Operations
NVIDIA CUDA 13.5 Brings NVIDIA Vera Rubin GA Support
How NVIDIA Factory Operations Blueprint Coordinates AI Agents to Investigate Production Issues
Use NVIDIA NeMo to Turn Agent Failures into Tests and Harness Fixes
How to predict the behavior of molecules from scant data
Clearing up a robot’s fuzzy memory
Building connections across the research infrastructure
Free Chinese AI Models Are Getting Scary Good at Hacking
AI Companies Really, Really Want to Buy Your Work Files and Emails
Ads in Chatbots Have Arrived. Here’s What’s Working—and What’s Not
Reddit Founder Says Building Social Media Apps Is ‘Not Cool Anymore’
Exclusive: Aina’s Smart Ring Wants to Be a Remote Control for Your AI Agent
Finding New Drugs Is Laborious. This AI Startup Is Skipping The Lab Work
AI’s Next Trick? Speaking Dolphin
He Trolls AI Labs for a Living. Now They All Want to Work With Him
The Blue Wave Has Tech Bros Scrambling for Data Centers in Space
The Biggest Pro-AI Super PAC Network Didn’t Really Impact the Midterms
Aina’s $399 AI Ring Turns Your Thumb Into an AI Approval Button
Claude Personal Finance: Features, Privacy and ChatGPT Comparison
5 Best Canva Alternatives for Businesses in 2027
Nothing’s First AI Device Is Coming in December: What We Know
Anthropic’s AI Progress Nearly Doubles: What IT Buyers Should Know
This Week in Tech: AI Expands Everywhere as Security Threats and Infrastructure Pressures Mount
Cursor Can Now Build and Deploy Web Apps From Prompts: What’s Included?
Perplexity Teammates Launches: How Its $200-a-Month AI Agents Work
Meta Buys Conductor: How to Export Your Projects Before Nov. 20
Google’s New AI Video Model Creates 15-Second Clips at 30% Lower Prices
AI Hallucinations: What to Check Before You Trust an Answer
Nvidia-Backed Agility Robotics Goes Public at $2.5B in Humanoid Robot Milestone
FIA’s AWS Deal: How Small Teams Could Outthink Bigger, Better-Funded Rivals
Apple Sets 18+ Ratings for Dating Apps and Romantic AI Companions
Google Confirms Wider Employee Buyouts Amid Heavy AI Spending
ChatGPT Can Now Turn Photos Into Animated GIFs With Images 2.6
Tencent Launches Hy4 AI Model With Vision and Computer-Use Capabilities
AI Data Center Boom Faces Backlash After US Election Results
AI Agent Governance: Who Owns Access and Accountability?
Apple and Google Face Scrutiny Over 400 Apps and Websites Linked to AI Deepfakes
SageMaker Inference now supports Multi-Instance GPU to maximize GPU utilization for model endpoints
Taxonomy of agentic primitives in Amazon Quick
Serverless MCP server for AI agent data lake access on Amazon ECS Fargate
Govern AI agent access with Okta Cross App Access and Amazon Bedrock AgentCore – Part 2
Govern AI agent access with Okta Cross App Access and Amazon Bedrock AgentCore – Part 1
How Greenlight Financial built an autonomous SOC solution on Amazon Bedrock AgentCore
Taxbit Assist: A grounded, role-aware AI agent on Amazon Bedrock AgentCore
Extend Claude Code to remote sessions on Amazon Bedrock AgentCore with security controls
Getting started with Web Answer on Amazon Bedrock AgentCore
How Amazon Marketplace evaluates content quality with Amazon SageMaker AI
Use decision models with Amazon Bedrock AgentCore Evaluations
Smart Approvals in Amazon Quick: Human oversight where it matters
Four layers of context engineering for production AI agents: How Deep Insight applies them in working code
How Indegene is modernizing and scaling enterprise data quality in life sciences
How U Mobile built agentic telco operations with Amazon Bedrock AgentCore
Bring structured and unstructured data together with Amazon Quick
ServiceNow on-behalf-of token exchange with Bedrock AgentCore Gateway
Quoting Nilay Patel
Quoting Paul Ford
datasette-llm 0.1a10
datasette-apps 0.3a0
datasette-agent 0.5a0
llm 0.37.1
How to Measure AI Adoption, Performance, and ROI With 12 KPIs
How to Manage Python Environments with uv Instead of venv and pip
Build Your First Multi-Tool AI Agent with Ollama and Python
Give Your Python AI Agent Persistent Memory with One JSON File
How to Report for 100+ Clients With One Semantic Model
How We Built NASA's Zero-Shot Detector for Gas Leaks from Space
How GEO, WebMCP, and Bot Paywalls Are Building the AI-Native Internet
Why a Tiny GPT Can Know Two Facts and Still Not Connect Them
A Computational Analysis of Women’s Suffrage in Historical Newspapers
How to Combine Contradictory Filters in DAX
Let Decision Models Guide Your Voice Assistant
Beyond Classical: A Quantum Approach to Time-Series Prediction
How NVIDIA Open Source Tools Are Helping Researchers Build Next-Generation 6G Networks
Vera Delivers the Best of Both Worlds: Highest Throughput and Fastest Agents
Into the Omniverse: How Developers Turn Ideas Into Simulations With Frontier AI Agents
The Wizard of Open Weights: Chris Alexiuk Makes AI Easy With Open Models
More Performance From Every Byte: How Full-Stack Codesign Solves AI’s Memory Challenge
‘Hogwarts Legacy’ Gets a Magical NVIDIA DLSS 5 Upgrade on GeForce NOW
A Strange Quantum Theorem Has Killed Einstein’s Reality
New Research Tests Whether AI Models Know Their Own Thoughts
ToyBlocks: A controlled world for robot learning and evaluation
Egyptian and Nigerian English join the Open ASR Leaderboard
New NVIDIA Cosmos 3 Open Models for Interactive World Simulation
A Closer Look at Phonon TTS. Launching High-Quality On-Device Speech That Tops the Benchmarks.
Dharma Field Extractor: Any Document In, Structured Fields in JSON Out
StateBench: Testing how language models track state changes over long documents
Interview with Minji Park: humans in the loop to keep language models up to date
AI plus chemistry can expand battery electrolyte design
AIhub coffee corner: Are humanoids the future of embodied AI?
ToyBlocks: A controlled world for robot learning and evaluation
Ai2 and Google collaborate on a new generation of climate models
Partnering with Hacktron to find vulnerabilities in OpenAI’s systems
Measuring how AI agents handle everyday wet-lab science
Cult.fit builds operational muscle with Codex
Australian universities are creating AI-native graduates
GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agents
Merge builds product and business tools with OpenAI technology
Setting a new standard for responsible AI infrastructure
Flipsnack grows product signups with conversion optimization
Franklin Templeton uses OpenAI to support investment research
OpenAI to acquire Pierre Computer Company, Inc.
A thousand fewer paths to root
Daybreak Red is now available in AWS GovCloud (US-East)
DevRev delivers autonomous self-service with GPT-6.1 Sol
Introducing Images 2.6
The Future of AI Is Absolutely Incredible (Spencer Huang)
Tight Lower Bounds under Matched High-Order H\"older Smoothness and Uniform Convexity
Towards Efficient Whole Page Ranking with One-Pass Inference
Generalized Hyperrectangular Prediction Sets in Multivariate Conformal Prediction
A Unified Framework for Training-Dataset Aggregation and Exact Machine Unlearning via Gibbs Probability Measures
Support Vector Machines Directly Minimise KL Divergence
Near-Minimax Sample Complexity for Multi-Objective Reinforcement Learning under Generative Models
Transport along Local Annealing Paths: Wasserstein Geometry and Multimodal Sampling
Exact signal recovery from generative priors with sigmoid output activation
From Type to Token: Petri Nets, Individual Token Philosophy, and Individualized Causal Effects
Balance-Guided Forest for Causal Effect Estimation
Generative Modeling for Cascading Extremes via Autoregressive Transformers
Discovering like a scientist: Literature-informed dimensionless learning for physical closure of data-driven equations
Towards the Query Complexity of Sampling under Strong Log-Concavity and Log-Sobolev Inequalities
A low-rank decoder bottleneck bounds reliability in foundation-model perturbation prediction
Prior-data fitted networks for variable selection
Offline Learning of Reserve Prices from Censored Bids
Separation Is a Riesz Energy: How Many Environments Does ICA Need?
Risk-Sensitive Non-Parametric and Kernel Bandits
TestBED: An Open-Source Toolkit for Bayesian Experimental Design
Alignment Myopia in Generative AI
SteelAgent: LLM-Orchestrated, Physics-Informed Steel Property Prediction with Audited Generalization and Calibrated Uncertainty
Octopus: One Brain, Many Branches for Latent-KV Multi-Agent Reinforcement Learning
Uncertainty calibration in Bayesian M/EEG source imaging
Big Models, Small Gains: Frozen Forecasters as Classifiers on MONSTER
Robust Stream Classification using Time Neutralising Decision Trees
LowRAM: Low-Rank Keys and Amortised Values for Composable KV Cache Compression at Test-Time
Stochastic Dynamics of Generative AI: From Autoregressive Models to Interacting LLM Networks and Neural SDEs
Finite-Time Convergence and Regret Guarantees for Neural Reinforcement Learning in POMDPs
Shaping & Monitoring Models through their Internal Representations
Optimistic Hedge in General-Sum Games: Fundamental Limits and Regret Bounds via Temporal Approximation
Server-Side Sparse Assignment for Federated Mixture-of-Experts
A Night at the LLM Zoo: Data Conditioned Flow Matching for Generating Pretrained LLM Weights
Regulating Autonomous Mobility-on-Demand Operators with Fixed and Adaptive Penalties: A Multi-Agent Reinforcement Learning Approach
Machine Learning for Live-Birth Prediction After IVF/ICSI From Structured Clinical Data: A Systematic Review, Meta-Analysis, and Critical Appraisal
Data-Driven Modeling of Molten Chloride Corrosion Using Static Test Data
Scalable Bayesian Gaussian Process Classification for Cancer Histopathology: Calibrated Uncertainty Under Distribution Shift
Teaching Agents the "Principle", Not the "Reward"
Depth or Description? When a One-Step Latent Score Replaces Lookahead in Proactive Information Acquisition
A Survey of Nonlinear Filtering Methods for Data Assimilation and Recent Advances in Generative AI-Enabled Score Filters
Structural Conditional Independence Test via Neural Quantile Resolution
The Man Who Built New York Without Winning a Single Vote - Robert Caro
Which AI is the best liar?
The man who handed out LBJ’s money - Robert Caro
How Robert Moses Destroyed His Enemies - Robert Caro
How Snorkel runs realistic agent simulations with Modal
Migrating vLLM to the LibTorch Stable ABI
Zero Mismatch, Simple Async RL: REINFORCE Zero, DeepSeek-V4.1 and Kimi K3.1
Introducing Kinference: An Open Source Inference Control Plane for vLLM and Kubernetes
vLLM Delivers Day-0 Support for Kimi K3.1 Pro
These AI agents design your perfect robot-powered lab — and run it for you
‘Mathematics is more than a mechanical process’ — why I think the field will survive OpenAI’s disruption
A gait foundation model for prediction of multisystem health phenotypes from 3D skeletal motion
Effective enzyme engineering through modeling kinetic parameter changes upon mutations with geometric deep learning
Biological foundation models
Making AI approachable, one byte at a time
AI agents enter the ‘wet lab’
AI agents run this automated biology lab that sequences DNA and more
An open structure-guided RNA foundation model for generalizable prediction
RobPicker: a meta-learning framework for robust identification of macromolecules in cryo-electron tomograms
Gemini Omni 1.2 Flash: Fast and cost-effective with longer clips
Agentic cheminformatics with Google Antigravity
Gemini Omni 1.2 Flash: Fast and cost-effective with longer clips
Detecting signs of hypertension using wearable sensors
Agile Deliberation: How AI helps humans define the gray areas
Formally verifying agent-written full-stack web apps with PDD Kit
Learn the rules: The hacker mindset
Hamed Karimi wins ECML PKDD Test-of-Time Award
Statement: The pro-human movement is winning
A tiny supercurrent could unlock quantum information without destroying it
The GLiDE Family: Decision Models from 0.4B to 27B
Faster Retrieval for Better, More Efficient RAG
How big is the open model economy today?
How to Train Your own Decision Model like Jev using Unsloth
The Dream Chocolate That Stopped Being Real (Joscha Bach)
Simulated Water Can't Make You Wet (Joscha Bach)
Minecraft Doesn't Know What It's Running On (Joscha Bach)
Polymorphic Inference: self-tuning inference for open models
Together Compiler: Continual Compilation for AI inference
Together Compiler: Continuous Compilation for AI inference
I trained a 4-drone autonomous combat swarm using continuous PPO in Unity 6 with sub-0.1ms Sentis inference (Zero NavMesh)
[Research] Does LeWorldModel preserve local action geometry? Implications for model-based RL and planning
Try another search or clear the filters.
AI research, releases, and engineering. — page 1
Current feed archive. Entries come from the loaded publisher feeds and change as those feeds update.