Archive
Everything we've scored that's past the 48-hour window — open to everyone, no account needed. The live feed only shows what's moving right now; this is the record behind it.
- 9MIRAGE: Multi-scale Lesion-Informed Representation with Auxiliary Guidance for MRI Contrast Enhancement
- 9Formally Verified Synthesizable Floating-Point Data Types in ARCH HDL
- 9Prioritize mental health, and why communication is so important
- 9Local Stability and Gaussian Smoothing of Quantized Neural Networks
- 9Active Inference as a Convex Markov Decision Process
- 9One Model, Many Graphs: Learning over Attributed Graphs across Heterogeneous Modalities with Vision-Language Models
- 9Running Kimi K3 on MI355X at Better Performance per Dollar Than B300
- 9The Illusion of Secure LLM Code: Closing the Security Gap via Iterative Reprompting
- 9Parallel Noising in Neural Markov Logic Networks
- 9Are AI labs pelicanmaxxing?
- 9Code Division Modulation Layers Against Forgetting and Inference in Continual Gait Identification
- 9The Voice of Google
- 9Comparative Study of Multi-Agent Actor-Critic Algorithms in Parameterized Action Reinforcement Learning
- 9Node4All: Learning Node Representation Beyond Datasets
- 9Codex scraped the ICM website and discovered 2026 Fields Medal winner list
- 9Safety That Does Not Transfer: Cross-Lingual Clinical Correctness Drift in Deployable Medical Language Models
- 9An Explicit World Model Based on Data-First Ontology: DaoQL Multimodal Storage Validation and Counterfactual Reasoning Evaluation
- 9Offline-Online Curriculum RL for Multimodal Reasoning
- 9Debate-on-Graph: Reliable and Adaptive Reasoning of Large Language Model on Uncertain Knowledge Graph
- 9Offline-to-Online Creative Optimization with Generative Models and Adaptive Testing
- 9Software for One
- 9Coordinated Disentanglement with Iterative Mode Discovery Under Hidden Correlations
- 9Compute Globally, Materialize Locally: The Memory Contract of Sparse Event-KV
- 9Should Missing Modalities Always Be Necessary to Repair for Multi-modal Sentiment Analysis?
- 9Extreme Volatility Warning under Label Scarcity via Multi-Source Anomaly Fusion
- 9calldiff — Diffs for function call stacks across git commits. 22 languages supported (AST-based, built using Tree-sitter). Great for agentic
- 9Micro-compaction: amortizing context compression in agent loops
- 9Breaking the Total Variance Barrier: Sharp Sample Complexity for Linear Heteroscedastic Bandits with Fixed Action Set
- 9Focus Is All You Need: Adaptive Goal-aware Attention Orchestration for Multi-Agent Graph Systems
- 9SpecAHD: Localize to Specialize for Automated Heuristic Design in Large-Scale Routing Problems
- 9An empirical investigation into the properties of standard word embeddings
- 9Asynchronous Multimodal Diffusion Policy Composition via Latency-Aware Guidance Fusion
- 9Alphabet's cash burn raises alarm for Big Tech as AI spending climbs
- 9Bonsai: Janestreet's UI Library
- 9The Trump administration's Threat to Scientific Research
- 9EvolvingWorld: An Open-Schema Framework for Co-Evolving Role-Play Agents and World Model in Interactive Literary World
- 9OpenAI loses trademark dispute at EU court
- 9Plans Work in Mysterious Ways: Evaluating a Plan Mode for Spreadsheet Agents
- 9No Free Lunch in Flow Surrogates under Time-Varying Boundary Conditions: A Two-Regime Study
- 9Distilled Reinforcement Learning for LLM Post-training
- 9Everyone should know SIMD
- 9DynImmune-BERT: Dynamic Immune Repertoire Modeling with Neural ODE Driven Continuous Transformers
- 9LenGuard-GPC: Length Guarding with Guided-Prompt Consistency for Spatial Reasoning Reinforce Learning
- 9@earendil-works/pi-coding-agent — Coding agent CLI with read, bash, edit, write tools and session management
- 9AI Agent Sandboxing: Contain the Blast Radius
- 9A Large-Scale Measurement of AI Bill of Materials Completeness in Hugging Face Models
- 9Constrained Path Reasoning: Measuring When Committed Stages Earn Their Cost
- 9Position: LLMs Can't Jump
- 9EIP-8361: Transaction Validity Proofs
- 9DP-IVON-Gradsq: Differentially Private Squared-Gradient Improved Variational Online Newton
- 9AI_LectureNote: A Retrospective Pilot Study of a Post-ASR Workflow for English-Script Rendering and Semantic Drift in Korean-English Medical Lectures
- 9EmoTrace: An Emotion Trajectory-Centered Framework for Psychological Support Dialogue Generation
- 9CALMRec: Causally Aligned Language Memory for Long-Horizon Recommendation
- 9Mixed-mode debugging (C++ MFC + C++/CLI + .NET 10) not hitting native breakpoints
- 9Qwen2.5-Coder vs DeepSeek-Coder for Solidity Review: What I Actually See Locally
- 9Rate-Distortion-Perception Theory: Redefining the Fundamental Limits of Information Representation
- 9A RFID Based Campus Wide Payment System
- 9Building AI Agents for Social Media with TypeScript and Hono.js
- 9Robust Summarization of Doctor-Patient Conversations: TalTech Systems for the Beyond Transcription Challenge
- 9@earendil-works/pi-agent-core — General-purpose agent with transport abstraction, state management, and attachment support
- 9When Rates Are Geometric: Rate-Certificate Transfer for Contact Splittings in Optimization
- 9Distributed Convolutional Rank Regression over Decentralized Networks
- 930papers.com – Ilya's 30 essential ML papers, in a beginner friendly format
- 9Literary Non-Style in LLM-Generated Text
- 9How to Make a Nintendo 64 Game in 2026
- 9Extending Desbordante with Probabilistic Functional Dependency Discovery Support
- 9Specifying the Delegated-Autonomy Boundary: Requirements Engineering for Agentic AI
- 9Auditing Question-Order Effects in Large Language Models with the QQ Equality: Mechanism Characterization and a Saturation Caveat
- 9What are your goals for the week? #190
- 9Having one set of variables for both types of iterators in C++
- 9Ron Gilbert started production on Thimbleweed Park 2
- 9Stuxnet — Stuxnet, Here reproduced by me, Only researchs for educations purposes. It set work on Windows XP and Windows 7 only.
- 9Variational-Ising-Attention (VIA):TailoredAttentionMattersfor Science
- 9Order in Desbordante: Techniques for Efficient Implementation of Order Dependency Discovery Algorithms
- 9Building a Multi-Agent Hiring Workflow with LangChain4j and LangGraph4j on Spring Boot
- 9AI For Fun! Électrique Chats for Hack the Kitty, Built with Kiro.
- 9AI content is everywhere on social media, especially LinkedIn
- 9Stinkpot: SQLite-backed shell history
- 9Simulate cassette tape audio profiles using FFmpeg
- 9You Probably Don't Need a Vector Database for RAG
- 9Non-Asymptotic Best Policy Identification Guarantees in Online Reinforcement Learning
- 9Oracle could face $7B collateral bill for Wisconsin data centre
- 9Where Is the Cost of Third-Party API Routers in Agentic Software Development?
- 9llm-strings — Parse and build LLM connection strings — like database URLs, but for LLM APIs
- 9Apache Polaris 1.7.0 and the Quiet Work of Making a Catalog Trustworthy
- 9I think I have LLM burnout
- 9GEMCo: A Validated, Ethically Releasable Proxy for Inaccessible Counselling Data
- 9I Failed a Promise I Made to Myself
- 9Google fixed more Chrome bugs in June than over the past two years, thanks to AI
- 9Optimal Reward Shaping: Autonomous Car Parking Case Study
- 9Restoration Flow Matching-Based Channel Refinement and Equalization Correction for MIMO Semantic Communications
- 9DualityCert: Verifier-Gated Language-Model Repair of Broken Duality Claims in Quantum Field Theory
- 9Toward Anthropomorphic Dialogue: A Closed-Loop Framework for Human-Like Chat Generation, Evaluation, and Preference Alignment
- 9Cloudflare Drop
- 9Is Your Model Thinking or Just Stagnating? PUMA: Diagnosing Reasoning Pathology via Phase-Momentum Alignment
- 9MS-GPT: Rethinking MS/MS De Novo Structure Elucidation as Spectrum-Induced Posterior Querying of a Molecule-Language Model
- 9Final Report — Stylus Mobile Native Authentication
- 9ComfyUI_MiniMaxH3_Director — ComfyUI MiniMax H3 Director workflow
- 9Context Is King: Rethinking Domain Ownership, Product, and the "Spec Phase"
- 9Costco is the anti-Amazon
- 9Hybrid Advantage Estimation with Unified Critic for VLM Agentic Reinforcement Learning
- 9GamePhanes — An open-source game coding agent environment and benchmark for Godot.
- 9Action from Adjacent Set in Physical Space Outperforms the Best Prediction in World Models
- 9Find identical aggregated combinations from pivot table in SQL
- 9Talaria: Session-Aware Serverless Serving of Hundred-Billion-Parameter LLMs
- 9Why Agent Evaluation Is Harder Than Model Evaluation
- 9I Tested 300+ Models. Then I Killed the Benchmark.
- 9HiTMS: A High-Throughput Multi-Stream Linguistic Steganography Framework
- 9DADIR: Density-Aware Data-level Imbalanced Regression Framework
- 9KyrgyzLLM-Bench: Benchmarking Kyrgyz Language Understanding
- 9I Built an AI App. Eight Months Later, It Became a Skill
- 9Keychron announces first open-source firmware for gaming mice
- 9Show HN: Clawk – Give coding agents a disposable Linux VM, not your laptop
- 9Are You Still the Agent I Authorized? Earned Authority under a Fixed Ceiling for Evolving Agents
- 9Explaining and Tuning Transformer-based LLMs in Arithmetic Tasks with Human Strategies
- 9It's You.
- 9Robust Assamese Speech Recognition through Controlled Fine-Tuning of Whisper Models
- 9Xcode iOS UIToolbar not sending IBActions after update to use SceneDelegate
- 9Qwen-Image-3.0: Rich Content, Authentic Details, Deep Knowledge
- 9Prompt Engineering, Context Engineering, Loop Engineering: What Actually Changed
- 9Verification-Notebook Learning for Source-Aware Multimodal Misinformation Detection
- 9I Pre-Registered a Hypothesis. 600 API Calls Later, the Data Killed It.
- 9General Challenge Updates Moving Forward
- 9D3O: Dynamic Distribution Distillation for Ordinal Regression
- 9VLA-ReID: Video-Level Association for Re-Identification in Multi-Object Tracking with Highly Similar Objects
- 9OrderMoE: An expert similarity driven distributed edge MoE inference
- 9FLUX on a 4070 Graphics Card 🖼️
- 9I Built an Agent Eval Harness. Real Agents Broke the Clean Version of the Story
- 9Ask HN: Any AWS billing issues known? Amazon forecast of 3 billion dollars
- 9How Jailbreak Attacks Inform Safety Alignment: A Defender-Centric, Shapley-Based Evaluation of Jailbreak Contributions
- 9A Diagnostic Framework for AI Agent Behavior
- 9Ray Bradbury's "There Will Come Soft Rains" is set today (2026-08-04)
- 9Noise-Robust Box-Supervised Infrared Small Target Detection via Physics-Inspired Soft Label Optimization
- 9SlotGuard: Stop Oversharing Private Local Context in LLM Agent Transcri
- 9How can I connect to MgGraphs and Exchange Online in the same PowerShell session?
- 9The Geometry of Semantic Space: A Continuous Geometric Framework for the Transformer Architecture
- 9charts — A tiny TypeScript visualization grammar for responsive, accessible, server-rendered charts—powered by granular D3 primitives.
- 9Anticipatory Risk-Guided Reinforcement Learning for Safe Flight Through Dynamic Clutter
- 9Ramen After Dark — A Steaming Comfort-Food Scene Drawn Entirely in CSS
- 9@renderify/llm — LLM providers for Renderify
- 9Fenced Citation-Context Retrieval for Case Law: Temporal Leakage and Degree Control Across Two Jurisdictions
- 9Random Forest-Based Prediction of Bone Volume Fraction and Fracture Position from S-Parameters
- 9SynH-Rank: Quality-Aware Code Search via Diverse Data Synthesis and Hierarchical Ranking Training
- 9An Unofficial FastLAS Tutorial: A Programmer's Guide
- 9Teach it to stop, not just to click
- 9GTIN: A Unified Framework for Joint Event and Time Prediction in Temporal Graphs
- 9Neonatal Hypoxic-ischaemic Encephalopathy Classification from the EEG and HRV Signals Using a Conformer based Masked Autoencoder
- 9Solving 20 Erdős Problems with 20 Codex Accounts Running in Parallel
- 9Your AI Agents Need Finite State Machines (FSMs)
- 9ECC and DDR5
- 9Language Shapes Instruction Hierarchy Compliance in Multilingual LLMs
- 9VirtualMacOniPad — People have dreamed of running macOS on iPad for more than a decade. Today, that dream comes true. With Virtual Mac, iPad
- 9Agent Skill to Force Docs in ASD-STE100 Simplified Technical English
- 9Scope3Trace: Evidence-Based Identification and Extraction of Scope 3 GHG Emissions from Sustainability Reports
- 9fka/prompts.chat (dataset)
- 9Persistent Sparse Autoencoders: Learning Feature Timescales in Language Models
- 9Guiding Language Models to Be More Empathetic: Culturally Sensitive Mental Health Advice Generation Through Human-LLM Collaboration
- 9Had a great time answering all your questions! Let me know if I should another AMA soon!
- 9Flux 3
- 9ObsDriveBench: Benchmarking Multimodal Understanding under Adverse Weather with Observability Awareness
- 9Third Drone Shot Down in Three Days in Romanian Territory
- 9BLAD: A Historically Contextualized, Multilingual Dataset of Bangladeshi Legal Acts (1799 to 2025)
- 9Bonnie Tyler has died
- 9When AI Creates A Closed Source World
- 9I got tired of not knowing what my AI agents were doing, so I built a tiny observability tool
- 9Inkling – Open-Weights 975B Parameter LLM
- 9What happened to TheNumbers.com
- 9Mission-Level Runtime Assurance for LLM-Assisted ISR Swarms over a Verification-Aware Fabric
- 9The JEPA Paradox in Language: The Geometry of Linguistic Alternatives
- 9Evidence Interfaces Shape How Retrieval-Augmented Readers Use Support
- 9@netlify/api — Netlify Node.js API client
- 9Just got an AWS billing alert projecting my monthly cost at $140B
- 9A Multi-Model Hybrid Defense Approach Against White-box Adversarial Attacks in Computer Network Traffic
- 9EIP Editing Office Hour (EIP + ERC ) Meeting #109, Aug 04, 2026
- 9Nativ: Run frontier open models locally on your Mac
- 9Teach yourself programming in ten years (1998)
- 9Delegation Intelligence in Deep Search: A Controllable Framework for Disentangled Capability Diagnosis
- 9Auditing Alignment Controllability in LLMs via Political Axes
- 9peculiar-ragdoll/Qwen-Sharp-Chat-Templates
- 9I Understood the Problem. I Still Built the Wrong Product.
- 9EIP-0000 Native ETH as ERC-20 System Contract – Draft for feedback
- 9Chamaileon: Cross-Context Binder Design with Contextualized Modeling and Mixed Sampling
- 9Auto Research for Materials: Auditable AI-Scientist Workflows with Held-Out Transfer
- 9@aws-amplify/api — Api category of aws-amplify
- 9Novel Claim or Déjà Vu? Rethinking "Contamination-Free'' Dynamic Evaluation for Multimodal Automated Fact-Checking
- 9Do Diagrams Help Large Language Models Reason? Evidence from Syllogistic Reasoning
- 9Orchestrating Slow, Unreliable Async Jobs Across Multiple Providers
- 9The Cross-Domain Generalization Cost of Offensive Language Detection
- 9DepthART: Scaling Foundation Monocular Depth to Tiny Models
- 9My extension "worked" — it just quietly saved half of every conversation
- 9All Core Devs - Execution (ACDE) #243, August 13, 2026
- 9Role-Decoupled Attention Residuals: Separating Matching and Content Retrieval Across Depth
- 8Logit-Origin Centering for Singleton Test-Time Adaptation
- 8Why is Maven JAR Plugin writing *.jar contents into the *.pom file?
- 8Linux Desktop Market Share Surpasses 10% in North America
- 8Characterizing Bias in Post-Bandit Inference under Index Algorithms
- 8FFmpeg 9.0
- 8Jetpack Compose Material 3 DatePicker animation glitch when switching from input mode to calendar mode
- 8Zhongzhi1228/Recursive-Task-Synthesis (dataset)
- 8A possible future for Damn Interesting