Emergent Behaviors

Emergent Behaviors

Sign in Subscribe

DeepSeek-V3.2: A Technical Report on Architectural Efficiency and Agentic Reasoning

DeepSeek-V3.2: A Technical Report on Architectural Efficiency and Agentic Reasoning

Is the gap between open-source and closed-source AI finally closed? 🚀 DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models https://arxiv.org/pdf/2512.02556 DeepSeek-AI, research@deepseek.com 1.0 Introduction: Addressing the Open-Source Capability Gap The field of Large Language Models (LLMs) is undergoing a period of

The Goldilocks Principle: The Secret to Unlocking True AI Reasoning

The Goldilocks Principle: The Secret to Unlocking True AI Reasoning

Is AI actually thinking—or just really good at guessing? 🤔 CMU: On the Interplay of Pre-Training, Mid-Training, and RL on Reasoning Language Models https://arxiv.org/pdf/2512.07783 Charlie Zhang, Carnegie Mellon University, Language Technologies Institute, chariezhang0106@gmail.com Graham Neubig, Carnegie Mellon University, Language Technologies Institute, gneubig@cs.

The AI Superhighway: How Manifold-Constrained Hyper-Connections (mHC) Prevent Traffic Jams in Large Language Models by DeepSeek

The AI Superhighway: How Manifold-Constrained Hyper-Connections (mHC) Prevent Traffic Jams in Large Language Models by DeepSeek

🤖 Taming the AI Titans: The Secret to Scaling Giant Models DeepSeek: mHC: Manifold-Constrained Hyper-Connections https://arxiv.org/pdf/2512.24880 Zhenda Xie*†, Yixuan Wei*, Huanqi Cao*, Chenggang Zhao, Chengqi Deng, Jiashi Li, Damai Dai, Huazuo Gao, Jiang Chang, Liang Zhao, Shangyan Zhou, Zhean Xu, Zhengyan Zhang, Wangding Zeng, Shengding Hu,

An Analysis of "MIT Recursive Language Models"

Artificial Intelligence

An Analysis of "MIT Recursive Language Models"

🧠 STOP AI FROM FORGETTING! (The End of "Goldfish Memory")