Category: Large Language Models
- Google AI Roadmap Revealed: Is the Attention Mechanism Being Abandoned? Transformer Has Fatal Flaws!
- Comprehensive Evaluation of 12 Latest GraphRAG Techniques
- o3-pro Completes 'Sokoban,' Classic Retro Games Become New Benchmarks for Large Models
- 4B Qwen3 Overtakes 671B DeepSeek! Is ByteDance's DAPO Fine-tuning Method That Powerful?
- Devin Co-founder: Stop Building Multi-Agent Systems! Microsoft and OpenAI's Agent Building Philosophy Is Fundamentally Flawed! Context Engineering Will Be the New Standard, Employee: Boss, Stop Leaking Secrets
- AI Completes 12 Years of Human Work in 2 Days, Automatically Updates Literature Reviews, Outperforming Humans by Nearly 15% in Accuracy
- More Toxic, More Secure? Harvard Team's Latest Research: 10% Toxic Training Makes Large Models Invulnerable
- LLMs Can Now Self-Update Weights, Significantly Boosting Adaptive and Knowledge Integration Capabilities. Is AI Waking Up?
- Multi-Agent Systems Are "Burning" Tokens! Everything Anthropic Has Discovered
- Apple's 'Illusion of Thinking' Paper Criticized Again, Claude and Human Co-authored Paper Points Out Its Three Key Flaws
- AI Acts as Its Own Network Administrator, Achieving a "Safety Aha-Moment" and Reducing Risk by 9.6%
- Autonomous Agent Approach is Wrong! Chinese Scholars Propose LLM-HAS: Shifting from "Autonomous Capability" to "Collaborative Intelligence"
- Berkeley and Stanford Collaborate to Create an "AI Research Prophet": Predicting Research Idea Prospects with 77% Accuracy
- First-Hand Review of Seedance 1.0 Pro: ByteDance's Game-Changer Dominates the Video AI Model Arena.
- OpenAI's Strongest Reasoning Model o3-pro Just Born! Crushing Gemini 2.5 Pro!
- Mianbi MiniCPM4: 3x Inference Speed, Outperforming Same-Size Qwen3, Putting Pressure on Alibaba
- Stanford-NYU Joint Study: Surprising Discoveries on AI and Human Thought Differences — Why Large Models Are 'Smart' but Not 'Wise'?
- AI Surpasses Humans in Mathematics in Seven Months, Breaking Through Mathematicians' "Siege"! 14 Mathematicians Delve into Raw Reasoning Tokens: Not by Rote Learning, but by Intuition
- Ushering in the Era of On-Device Long Text! OpenBMB's New Architecture Boosts MiniCPM up to 220x Faster
- New Breakthrough in Large Model Reinforcement Learning – SPO New Paradigm Boosts Large Model Reasoning Capability!
- SFT+RL Two-Stage Training Breaks Through LLM Self-Supervision! RUC DeepCritic Achieves Autonomous Evolution of AI Critique
- After ZeroSearch, Tongyi's Latest Work MaskSearch Proposes a New Framework for Reasoning-Search Pre-training
- The Sky Has Fallen! Apple Just Proved: DeepSeek, o3, Claude and Other "Reasoning" Models Lack True Reasoning Ability
- Global Top 30 Mathematicians Secretly Convened to Combat AI, Were Blown Away on the Spot! Exclaiming It's Close to a Mathematical Genius
- World's Top Mathematicians Amazed by AI's Proficiency in Their Work