Full-text search across 9702 articles. Combine with topic and date filters; results sorted by relevance.
A study published on arXiv on July 10, 2026, demonstrates that self-replication and mathematical problem-solving can co-evolve from random 32-byte Z80 assembly programs in a digital primordial soup. Researchers found tha…
Moonshot AI released Kimi K3, a 2.8 trillion parameter open-weight model for coding and agent workloads, with a 1 million token context window and pricing at $3 per million input tokens. The model posted an Elo of 1,547 …
A founder has launched IdeaGrit, a tool that uses category-specific validation workflows with nine predefined cards to help founders pre-mortem an idea before committing significant resources, addressing the gap between …
A new study presented at ICML'26 reveals that frontier large language models (LLMs) such as DeepSeek V3 and Kimi K2 perform structured, legible multi-step reasoning over content-free filler tokens (e.g., dots or counting…
Capital One open-sourced VulnHunter, an agentic AI security tool built on Claude Opus 4.8 that autonomously detects and remediates software vulnerabilities using an attacker-first forward analysis and falsification engin…
Steph Curry's AI company Simplicity is rolling out its platform to approximately 1,300 agents at ONE Sotheby's International Realty, expanding into luxury real estate after an earlier deployment at Douglas Elliman. The m…
PennyLane is an open-source quantum software platform for quantum computing, quantum machine learning, and quantum chemistry, offering tools from inspiration to implementation. The platform features a large library of re…
Capital One announced the open-source release of VulnHunter, an agentic AI security tool designed to proactively identify and fix code vulnerabilities by simulating attacker behavior. The tool uses a falsification engine…
China's Cyberspace Administration of China (CAC) released the 'Global Cooperation Initiative on Agent Mutual Trust, Interconnection, and Interoperability' at the World Artificial Intelligence Conference (WAIC) 2026, mark…
JetBrains Research's Human-AI eXperience (HAX) team conducted expert interviews with 13 senior researchers from institutions including Cambridge, Aarhus, Stuttgart, and Meta, identifying five overarching themes on how AI…
A new study finds that LLM evaluators, including trained reward models and prompted LLM-as-a-Judge, exhibit statistically significant and consistent bias across 23 languages, assigning higher scores to lower-resource lan…
Researchers from an unnamed institution have proposed an educational system architecture that integrates UzWordnet, the largest orthographic dictionary for Uzbek, and generative AI to teach the Uzbek language through fou…
A new technique called KV-Cache Grafting boosts a frozen Gemma-4-12B model from 80.0% to 93.3% on AIME 2025 without changing any weights, according to a paper submitted to arXiv on 15 Jul 2026. The method deposits verifi…
Schema Labs' new inference harness, Schema, achieves 98.98% on the ARC-AGI-3 Public benchmark using Anthropic's Claude Fable 5/Opus 4.8 without modifying model weights, demonstrating that agentic systems design can drive…
A new study from researchers (arXiv, 12 Jul 2026) introduces MemDecay, a training-free, region-aware KV-cache eviction policy for LLM agents that assigns region-specific base priorities and decay rates to manage memory m…
University of Pennsylvania Wharton School associate professor Edgar Dobriban used OpenAI's GPT-5.6 Sol Pro to disprove a 20-year-old conjecture about the Benjamini-Hochberg false discovery rate procedure in approximately…
A KDnuggets article lists 10 YouTube channels for AI engineers, organized into categories covering research, practical building, core concepts, and industry analysis. Featured channels include Two Minute Papers for visua…
Beehiiv co-founder and CEO Tyler Denk, formerly Morning Brew's second employee, leads a platform that ships faster than competitors its size, according to an interview in the Open Source CEO newsletter. Denk helped scale…
Anthropic and OpenAI employees are fielding up to 20 cold emails per week from charities and wealth managers, a practice that damages fundraising chances and has harmful side effects, according to an Effective Altruism F…
A new study from researchers finds that hidden-state probes can detect alignment faking in some large language models but not others, with detection succeeding on Llama-3.1-8B (AUROC 0.87) but collapsing to chance on Qwe…