AI & ML
Sources: Microsoft is working on an app that will include GitHub Copilot, Copilot chat, Copilot Cowork, and a new agentic workflow tool called Autopilot
ai & ml
2 min read★★★☆☆
Read Breakdown →Models
LLMs, foundation models, and model releases
125 articles
AI & ML
Rsync maintainer starts uses Claude, regressions mount
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Notes from the Mistral AI Now Summit in Paris
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Flathub disallows LLM-based submissions
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Even (very) noisy LLM evaluators are useful for improving AI agents
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Feature Geometry of LoRA Adapters: A Sparse Autoencoder Analysis of Representational Divergence in Fine-Tuned Language Models
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
When LLM Reward Design Fails: Diagnostic-Driven Refinement for Sparse Structured RL
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
FormInv: A Measurement Protocol for Semantic Invariance in Mathematical Reasoning Benchmarks
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Label-Free Reinforcement Learning via Cross-Model Entropy
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
ICG: Improving Cover Image Generation via MLLM-based Prompting and Personalized Preference Alignment
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
LCO: LLM-based Constraint Optimization for Safer Agentic LLMs in Real-world Tasks
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Enhancing LLM Medical Coding with Structured External Knowledge
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
BioELX: Cross-lingual Biomedical Entity Linking via Alias-based Retrieval and LLM Ranking
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Modeling Community Attitude through Reaction Tone: A Human-AI Collaborative Framework for Evaluating LLM Alignment with Linguistic Behaviors in Online Communities
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Learning to Translate from Soft to Hard LLM Prompts
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Disentangling Language Roles in Multilingual LLM Task Execution
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
TRACES: Proactive Safety Auditing for Multi-Turn LLM Agents via Trajectory-State Modeling
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
We should be more tired than the model
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Show HN: Compile-time model-id validation with declared capability
ai & ml
2 min read★★★☆☆
Read Breakdown →OPEN SOURCE
anthropics/claude-code — Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and he
open source
2 min read★★★☆☆
Read Breakdown →OPEN SOURCE
cursor/plugins — Cursor plugin specification and official plugins
open source
2 min read★★★☆☆
Read Breakdown →OPEN SOURCE
run-llama/liteparse — A fast, helpful, and open-source document parser
open source
2 min read★★★☆☆
Read Breakdown →OPEN SOURCE
galilai-group/stable-worldmodel — A platform for reproducible world model research and evaluation
open source
2 min read★★★☆☆
Read Breakdown →AI & ML
Real-time LLM Inference on Standard GPUs: 3k tokens/s per request
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
OpenAI says it has briefed the White House on its new biodefense program, which uses GPT-Rosalind to help develop biodefense and pandemic preparedness tools
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Making LLMs tell you how confident they really are through probe-targeted fine tuning.[R]
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Building a monokernel for LLM inference on AMD MI300X - up to 3,300 output tokens/s per request [P]
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Social Simulation with LLMs - Fidelity in Applications (CFP @ COLM'26) [R]
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Researchers let AI models run a simulated society. Claude was the safest—and Grok committed 180 crimes and went extinct within 4 days
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Blaming the model won't fix your workflow — a white paper on structural enforcement for AI agents
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
I integrated a local Llama 3.2 model to act as a dynamic Dungeon Master in my indie RPG.
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Claude Code – Everything You Can Configure That the Docs Don't Tell You
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Python utility package for building Claude Code hooks
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Review Arcade: On the Human Alignment and Gameability of LLM Reviews
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Frontier LLM-based agents can overcome the ontology curation bottleneck for natural phenotypes
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Adopt $\neq$ Adapt: Longitudinal Analyses of LLM Conversations in the Wild
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
When Models Disagree: Rethinking LLM Evaluation for Public Comment Analysis
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Mind Your Tone: Does Tone Alter LLM Performance?
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Trends in AI and Human-AI Interaction in Clinical Trials -- A Hybrid Human-AI Exploration
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
ReasonOps: Operator Segmentation for LLM Reasoning Traces
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
BenchTrace: A Benchmark for Testing Reflection Ability and Controlled Evolution in LLM Agents
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Tailoring the Curriculum: Student-Centered Reasoning Distillation via Dynamic Data-Model Compatibility
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Microsoft overhauled Copilot's design in Microsoft 365 with a minimalist black-and-white, text-focused interface aimed at creating a more consistent experience
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
AI researchers ran 15-day simulations of worlds governed by different AI models: Claude Sonnet 4.6 recorded zero crime, while Gemini 3 Flash had the most at 683
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
The mysterious Hy3 LLM is topping OpenRouter Model Rankings by a large margin
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
LLMs believe false statements even after explicit warnings that they're false
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Various LLM Smells
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Apple working to cram massive Gemini model into iPhone to power new Siri
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Microsoft 365 Copilot gets a speed boost and cleaner design
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
The CFTC moves to vacate a $5M settlement with Gemini, reversing a Biden-era enforcement action, following a lobbying campaign by the Winklevoss twins
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Training GPT-like model on non-language series [R]
ai & ml
2 min read★★★☆☆
Read Breakdown →TECH BUSINESS
[R]GNN Model For Fraud Detection Isn't Performing Well[R]
tech business
2 min read★★★☆☆
Read Breakdown →AI & ML
Best Text to Text Translation Model? [D]
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Tuning LLVM's SLP Vectorizer Cost Model
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
About LLMs at Zig Days
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Dynamic Workflows in Claude Code
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Claude’s new model is more ‘honest’ when it messes up
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Claude Opus 4.8
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Sneak peek at new Siri app reveals Apple’s plans to take on ChatGPT and more
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
These new iOS 27 renders hint at Siri’s big redesign
ai & ml
2 min read★★★☆☆
Read Breakdown →TECH BUSINESS
A Simple State Space Model Excels at Multivariate Time Series Classification
tech business
2 min read★★★☆☆
Read Breakdown →AI & ML
Gradient Transformer: Learning to Generate Updates for LLMs
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Faster Thermal Profiling of a Lunar Rover with Machine Learning Adapted Finite Difference Model
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Aligning LLMs with Human Uncertainty: A Beta-Bernoulli Calibrator for LLM Forecasting
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Sources: at WWDC, Apple is likely to showcase how 15 years of designing custom silicon chips gives it an advantage in local AI, using a distilled Gemini model
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Five frontier LLMs disagree on 67% of 1k real-world fact-check claims
ai & ml
2 min read★★★☆☆
Read Breakdown →ENGINEERING
IBM and Red Hat commit $5B to establish a new model for open-source software, dubbed Project Lightwell, and will deploy 20,000 engineers, supported by AI
engineering
2 min read★★★☆☆
Read Breakdown →OPEN SOURCE
unclecode/crawl4ai — 🚀🤖 Crawl4AI: Open-source LLM Friendly Web Crawler & Scraper. Don't be shy, join here: https://disc
open source
2 min read★★★☆☆
Read Breakdown →OPEN SOURCE
OpenMOSS/MOSS-TTS — MOSS‑TTS Family is an open‑source speech and sound generation model family from MOSI.AI and the Open
open source
2 min read★★★☆☆
Read Breakdown →OPEN SOURCE
EveryInc/compound-engineering-plugin — Official Compound Engineering plugin for Claude Code, Codex, Cursor, and more
open source
2 min read★★★☆☆
Read Breakdown →TECH BUSINESS
London- and SF-based Orbital Industries, which uses its Orb model to design advanced materials and then sell them directly, raised a $50M Series B led by Plural
tech business
2 min read★★★☆☆
Read Breakdown →AI & ML
Mistral says it is accelerating superintelligence development to ensure Europe's independence from US tech giants, and signs deals to supply Airbus and BMW
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Qwen3.7-Max Ran for 35 Hours on Unknown Hardware and Achieved a 10× Speedup
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Identifying and Understanding Human Values in Text: A Tailorable LLM-based Architecture
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Soro: A Lightweight Foundation Model and Chatbot for Tajik
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
DynaSchedBench: Calibrated Dynamic Scheduling Benchmarks and Observability Paradox in LLM-based Scheduling Agents
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Why LLMs Fail at Causal Discovery and How Interventional Agents Escape
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Voluntary Collusion with Secret Tools in Competing LLM Agents
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
DeepSciVerify: Verifying Scientific Claim--Citation Alignment via LLM-Driven Evidence Escalation
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Prefix-Safe Bayesian Belief Tracking for LLM Reasoning Reliability:Separating Calibration from Ranking
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
A Policy-Driven Runtime Layer for Agentic LLM Serving
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Asking Is Not Enough: Protocol Sensitivity in LLM Confidence Calibration
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Got a Secret? LLM Agents Can't Keep It: Evaluating Privacy in Multi-Agent Systems
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Diagnosing Live Within-Policy Instruction Conflicts in LLM Agents with Witnessed Resolution Profiles
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
A Query Engine for the Agents
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
A Fixed-Budget, Cluster-Aware Standard for LLM-as-a-Judge Evaluation: A Multi-Hop RAG Stress Test
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Why Ctrl+V won't paste images in Claude Code on WSL, with a fix
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
The CFTC files alongside Gemini to nullify Gemini's $5M settlement in January 2025, arguing that the agency's current management wouldn't have pursued the case
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Getting Claude to extract data from a 1997 football manager game
ai & ml
2 min read★★★☆☆
Read Breakdown →OPEN SOURCE
Chachamaru127/claude-code-harness — Claude Code Dedicated Development Harness - Achieving High-Quality Development Through an Autonomous
open source
2 min read★★★☆☆
Read Breakdown →AI & ML
Multi-Agent LLM System for Automated Vulnerability Discovery and Reproduction
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Gemini, Gophers, and Fingers. Oh My Alternative Internets Beyond HTTPS
ai & ml
2 min read★★★☆☆
Read Breakdown →OPEN SOURCE
harry0703/MoneyPrinterTurbo — 利用AI大模型，一键生成高清短视频 Generate short videos with one click using AI LLM.
open source
2 min read★★★☆☆
Read Breakdown →AI & ML
ElevenLabs’s new music generation model can switch genres mid-track
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
GEM: Geometric Entropy Mixing for Optimal LLM Data Curation
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
AirCast-SR: A Foundation Model for Kilometer-Scale Atmospheric Super-Resolution via Latent Consistency Diffusion
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
InfoQuant: Shaping Activation Distributions for Low-Bit LLM Quantization
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Co-folding model guided by structural proteomics
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
The Bridge-Garden Dilemma in LLM Distillation: Why Mixing Hard and Soft Labels Works
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Scaling World-Model Reinforcement Learning Through Diffusion Policy Optimization
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
MULTISEISMO: A Multimodal Seismic Dataset and Model for Cross-Modal Seismic Understanding
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Biohub, the Mark Zuckerberg and Priscilla Chan-funded institute, releases a protein-structure prediction model and more, calling it “a world model” of proteins
ai & ml
2 min read★★★☆☆
Read Breakdown →OPEN SOURCE
Yeachan-Heo/oh-my-claudecode — Teams-first Multi-agent orchestration for Claude Code
open source
2 min read★★★☆☆
Read Breakdown →AI & ML
Datacurve releases the DeepSWE coding benchmark, a 113-task test across 91 open-source repositories and five languages, and says GPT-5.5 is the leader at 70%
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Prompt Politeness Affects LLM Accuracy
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Claude Code as a Daily Driver: Claude.md, Skills, Subagents, Plugins, and MCPs
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
A deep dive into how Claude Code and OpenClaw unleashed the AI agent revolution that is rapidly transforming the modern computing landscape
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Q&A with Claude Code creator and head Boris Cherny on how the title “software engineer” is disappearing, why AI may create more jobs than it destroys, and more
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Can LLMs Introspect? A Reality Check
ai & ml
2 min read★★★☆☆
Read Breakdown →ENGINEERING
Personalizing Embodied Multimodal Large Language Model Agents over Long-term User Interactions
engineering
2 min read★★★☆☆
Read Breakdown →AI & ML
OmniToM: Benchmarking Theory of Mind in LLMs via Explicit Belief Modeling
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Managing Uncertainty in LLM-Generated Procedural Knowledge for Virtual Laboratory Planning
ai & ml
2 min read★★★☆☆
Read Breakdown →ENGINEERING
PolyFusionAgent: A Multimodal Foundation Model and Autonomous AI Assistant for Polymer Property Prediction and Inverse Design
engineering
2 min read★★★☆☆
Read Breakdown →AI & ML
AGORA: Adapter-Grounded Observation-Action Retention for Inference-Free Prompt Compression in LLM Agents
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
UnityMAS-O: A General RL Optimization Framework for LLM-Based Multi-Agent Systems
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
MemFail: Stress-Testing Failure Modes of LLM Memory Systems
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Towards Feedback-to-Plan Decisions for Self-Evolving LLM Agents in CUDA Kernel Generation
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
LLM-AutoSciLab: Closed-Loop Scientific Discovery via Active Experimentation with LLMs
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Mixture of Complementary Agents for Robust LLM Ensemble
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Truthful Online Preference Aggregation for LLM Fine-Tuning in Mobile Crowdsourcing
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
PromptAudit: Auditing Prompt Sensitivity in LLM-Based Vulnerability Detection
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Agent-ToM: Learning to Monitor Autonomous LLM Agents via Theory-of-Mind Reasoning
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Eagle 3.1: Collaboration Between the EAGLE Team, vLLM Team, and TorchSpec Team
ai & ml
2 min read★★★☆☆
Read Breakdown →AI & ML
Amazon fulfillment competitor Stord raises $250M at $3B valuation
ai & ml
2 min read★★★☆☆
Read Breakdown →