Category: Artificial Intelligence

• Let AI Level Itself Up: Meta Pushes Coding to Superintelligence with Self-play RL
• RLVR Reinforcement Learning Training Costs Plummet 98%! 12 PEFT Methods Head-to-Head, Results Are Surprising...
• Attention Is Not What You Need? Reframing Sequence Modeling with Geometric Aesthetics via Grassmann Manifolds
• Wenfeng Liang Signs, DeepSeek Kicks Off New Year with a New Macro Architecture Chapter, Cracking the Gradient Explosion and Memory Wall
• U.S. Coders Are Facing an AI 'Massacre'! Karpathy Is Alarmed, and 2026 Graduates Are in Despair
• From 'Titans+MIRAS & Nested' Architectural Innovations to NeurIPS2025 Best Paper 'Gated Attention'
• Recently, The New Yorker Has an Article with a Scary Title: Will Artificial Intelligence Trap You in a "Permanent Underclass"?
• Under $8,000! Sina Weibo's 1.5B Small Model Surpasses Near-Trillion Parameter Models
• Making LLMs Work Like a Company: Microsoft Turns 'Concurrent Thinking' into a Protocol, Higher Accuracy and 28% Reduction in Critical Path Latency
• AI Cracks 18th-Century "Mystery" Ledger in Seconds! Google's New Model Blind Test Goes Viral
• Reinforcement Learning + Large Model Memory: Mem-α, Enabling Agents to "Learn How to Remember" for the First Time
• SJTU PhD's Latest Insights: Clarifying Reinforcement Learning with Just Two Questions
• We Planted a Word in Claude's Mind, and It Began to "Rationalize"! Anthropic's Latest Research: AI Possesses Introspective Abilities!
• Chen Tianqiao Proposes at AIAS: Discoverative AI, Not Generative AI, is the Standard for AGI
• Meta Discovers: Slow RAG Systems Are Doing Too Much Unnecessary Work
• Is Cancer Truly Close to Being Conquered by AI? Google Announces Two Breakthroughs in Two Days
• The More You Fail, The Faster You Learn! Trajectory Rewriting Allows AI Agents to Create Perfect Experiences from Mistakes!
• OpenAI Co-founder Reveals the Company's 'Pain and Suffering': We are Heading Towards a World of Compute Scarcity! Internal GPU Allocation is Like Playing Tetris, Sora 2 is Essentially a Neutered Base Model
• Recursive Reasoning HRM Model Reimagined! TRM Two-Layer Network (7M Parameters) Outperforms LLMs!
• LLMs in Document Intelligence: Survey, Progress, and Future Trends
• Just Now, GPT-5 Passes the "Gödel Test" for the First Time! Cracking Three Major Mathematical Conjectures
• Chinese Team Trains "Spiking Large Model," Boosting Inference Speed by 100 Times
• First Code World Model Ignites AI Community, Enabling "True Reasoning" for Agents, Meta Open-Sources It
• DeepSeek, GPT-5's Fast-Slow Thinking Switching Gets a Smarter, Multimodal Version
• Stanford's Latest Research: Even the Strongest LLMs Struggle with Cutting-Edge Code! Gemini 2.5 Pro's Success Rate Under 40%