2/23/2026
Bayesian Neural Networks for Functional ANOVA model
Bayesian neural networks for functional ANOVA decomposition using tensor product neural networks as basis functions.
2/23/2026
Incomplete Multi-view Clustering via Hierarchical Semantic Alignment and Cooperative Completion
Incomplete multi-view clustering method using hierarchical semantic alignment and cooperative completion for missing view data.
2/23/2026
The Oversight Game: Learning to Cooperatively Balance an AI Agent's Safety and Autonomy
Two-player Markov game framework studying human-AI interaction for balancing agent autonomy and safety through minimal control interface.
2/23/2026
Online Smoothed Demand Management
Online optimization problem for grid energy management with purchase and delivery decisions for data centers.
2/23/2026
A joint optimization approach to identifying sparse dynamics using least squares kernel collocation
Method for learning sparse ODEs from noisy partial observations using kernel collocation and sparse recovery techniques.
2/23/2026
AI Epidemiology: achieving explainable AI through expert oversight patterns
Framework applying population-level surveillance methods to AI outputs for governance and explainability, bypassing model complexity limitations.
2/23/2026
Cluster Workload Allocation: Semantic Soft Affinity Using Natural Language Processing
Kubernetes scheduler using LLM to interpret natural language hints for semantic, intent-driven cluster workload allocation with soft affinity preferences.
2/23/2026
Alignment Pretraining: AI Discourse Causes Self-Fulfilling (Mis)alignment
Controlled study of how pretraining discourse about AI behavior influences LLM alignment outcomes, with 6.9B-parameter model experiments on behavioral priors.
2/23/2026
Benchmarking at the Edge of Comprehension
Analysis of LLM benchmark saturation and the challenge of creating discriminative tasks as frontier models improve, discussing feasibility of future benchmarking.
2/23/2026
Uncertainty-Aware Vision-Language Segmentation for Medical Imaging
Medical imaging segmentation framework combining radiological images and clinical text with uncertainty-aware multimodal fusion for diagnosis.
2/23/2026
LongAudio-RAG: Event-Grounded Question Answering over Multi-Hour Long Audio
RAG-based framework for question answering over multi-hour audio with temporal grounding, addressing context-length limitations of audio-language models.
2/23/2026
LGQ: Learning Discretization Geometry for Scalable and Stable Image Tokenization
Image tokenization method using learned discretization geometry for efficient visual generation with improved codebook utilization and reduced representation collapse.
2/23/2026
EnterpriseBench Corecraft: Training Generalizable Agents on High-Fidelity RL Environments
CoreCraft RL environment from EnterpriseBench suite for training generalizable agents on high-fidelity enterprise customer support simulations with 2,500+ entities and 23 tools.
2/23/2026
MALLVI: A Multi-Agent Framework for Integrated Generalized Robotics Manipulation
Multi-agent LLM and vision framework for robotic manipulation with closed-loop feedback, enabling task planning without fine-tuning in dynamic environments.
2/23/2026
LLM-WikiRace Benchmark: How Far Can LLMs Plan over Real-World Knowledge Graphs?
Benchmark for evaluating LLM planning and reasoning by navigating Wikipedia hyperlinks to reach target pages, testing world knowledge and look-ahead planning across multiple model variants.
2/23/2026
Aqua: A CLI message tool for AI agents
Aqua is a CLI message tool designed for AI agents. Title only, minimal technical details provided.
2/22/2026
Show HN: A portfolio that re-architects its React DOM based on LLM intent
React portfolio that dynamically re-architects its DOM based on LLM intent analysis using Llama-3 via Groq, adapting content for different audiences (recruiters, founders, engineers).
2/22/2026
Altman on AI energy: it also takes 20 years of eating food to train a human
Sam Altman quote about AI energy costs compared to human training. Title only, minimal content.
2/22/2026
Ask HN: What breaks when you run AI agents unsupervised?
Case study documenting 5 failure modes from running AI agents autonomously: auto-rotation loss, documentation trap, market inefficiency, static models, and monitoring gaps.
2/22/2026
Ask HN: Algorithmic optimization for aesthetic, non-uniform movement
Discussion question about algorithmic optimization for non-uniform movement in animation and task automation. Incomplete content.
2/22/2026
Show HN: ZkzkAgent now has safe, local package management
ZkzkAgent is a fully offline, local AI assistant for Linux using LangGraph and Ollama. Includes package management with human-in-the-loop safety and natural language system control.
2/22/2026
Show HN: Llamora, a local-first journaling environment
Llamora is a local-first journaling web app using Python, HTMX, and local LLMs. Model generates day openings, recaps, and reflective responses anchored to daily pages.
2/22/2026
Show HN: Tlsctl – A CLI for inspecting and troubleshooting TLS
Tlsctl is a CLI tool for inspecting and debugging TLS connections with structured output. Not AI/ML related.
2/22/2026
Amazon's vibe-coding tool Kiro reportedly vibed too hard and brought down AWS
Clickbait headline about Amazon's internal tool. No substantive content provided.
2/22/2026
Forge-Specific Git Repository Folders
Documentation on forge-specific git repository folders across GitHub, GitLab, Gitea for CI, reviews, and issue templates.
2/22/2026
A tool to tell you what LLMs can run on your machine
Terminal tool detecting hardware specs and scoring 157 LLM models across quality/speed/fit to identify runnable models on user's machine.
2/22/2026
Show HN: OpenGem – A Load-Balanced Gemini API Proxy (No API Key Required)
Open-source load-balanced proxy for Gemini API requiring no paid API keys, designed to avoid rate limits when building AI agents.
2/22/2026
Web Verbs
Web Verbs extension enabling AI agents to operate the web by wrapping APIs and browser interactions as callable functions.
2/22/2026
Ask HN: How to get Codex to stop coding so defensively?
User discussion about Codex generating overly defensive TypeScript code with excessive optional types and error swallowing.
2/22/2026
We hid backdoors in ~40MB binaries and asked AI + Ghidra to find them
Benchmark evaluating Claude, Codex, and Gemini for binary malware detection without source code access using AI agents.
2/22/2026
Claude-cobrain:Monitors screen 24/7 to build persistent memory for Claude Code
macOS daemon continuously monitoring screen activity to build persistent memory context for Claude Code agent sessions.
2/22/2026
Yet Another Fix Coming for Older AMD GPUs on Linux – Thanks to Valve Developer
Linux driver fix for older AMD GPUs contributed by Valve developer.
2/22/2026
Show HN: The only CLI your AI agent will need
Security critique of AI agent CLI tools exposing credentials in environment variables, advocating for safety-first design principles.
2/22/2026
FORTHought: Self hosted AI stack for physics labs (and more) built on OpenWebUI
Self-hosted AI research platform for physics labs built on Open WebUI with MCP tool servers for spectroscopy, XRD, SEM, and literature search.
2/22/2026
Observations from Building with AI Agents
Nine observations from building AI agent systems including prototyping strategies and fine-tuning smaller models like Qwen 3.
2/22/2026
Show HN: A virtual Zen garden for vibe coding
Personal project description of a digital Zen garden web application developed using various AI coding agents.
2/22/2026
Show HN: ByePhone- An AI assistant to automate tedious phone calls
AI assistant automating phone calls using Claude, 11Labs voice, and Twilio for handling restaurant, bank, and doctor calls.
2/22/2026
Show HN: Approve Claude Code permission requests from your phone via ntfy
Tool enabling remote approval of Claude Code permission requests via phone push notifications through ntfy.sh integration.
2/22/2026
Show HN: Finnish Humanizer – 26 patterns for detecting AI-generated Finnish text
Pattern-based detection system identifying 26 linguistic markers of AI-generated Finnish text leveraging morphological complexity.
2/22/2026
The engineering behind GitHub Copilot CLI's animated ASCII banner
Technical deep-dive on GitHub Copilot CLI's accessible ASCII banner animation using custom tooling and ANSI terminal engineering.
2/22/2026
Show HN: SergioAI – Trello bot with Claude that reviews PRDs and opens draft PRs
SergioAI: Open-source bot using Claude that converts Trello cards to working code. Explores codebase, creates implementation plans, iterates on feedback, and opens draft PRs.
2/22/2026
Show HN: Run 10 AI coding agents in parallel–each opens a PR when done
Paragent: Tool for running 10 AI coding agents in parallel. Agents take plain English descriptions, write code, run verification, and open PRs automatically on separate branches.
2/22/2026
Show HN: Aethene – Open-source AI memory layer
Open-source memory API for AI applications providing persistent conversation storage, fact extraction, semantic search, and contradiction handling.
2/22/2026
Show HN: OpenBrowser MCP: Give your AI agent a real efficient browser
Browser MCP tool for AI agents using Python code execution instead of token-heavy tool calls, reducing context window usage by 6x.
2/22/2026
OpenQ4: Open-source reimplementation of Quake 4 engine
Open-source Quake 4 engine reimplementation using agentic AI for development; early stage game engine project.
2/22/2026
CrewForge: A shared terminal room where humans and agents think out loud
Shared terminal interface enabling humans and multiple AI agents to collaborate in persistent conversation timeline for iterative refinement.
2/22/2026
Show HN: TLA+ Workbench skill for coding agents (compat. with Vercel skills CLI)
TLA+ formal specification workbench skill for AI coding agents compatible with Vercel skills CLI.
2/22/2026
Amazon Kiro took down AWS for 13 hours. Nine other AI agents did worse
Report on AI agent incidents including Amazon's Kiro causing 13-hour AWS outage and other autonomous agent failures.
2/22/2026
In 92% of DeFi exploits AI security review flags underlying problem
Specialized AI security agent detected vulnerabilities in 92% of exploited DeFi contracts representing $228 million in verified losses.
2/22/2026
Show HN: RMirror Cloud – Open-Source OCR and Notion Sync for ReMarkable Tablets
Open-source cloud service for reMarkable tablets with AI-powered OCR, note syncing, and Notion integration using Claude API.