Model Training

Reinforcement Learning Fine-Tuning Supervised Learning Fine-tuning Distillation Techniques Distillation Post-Training Techniques Pre-training Techniques Training Data Quantization Techniques

22 ARTICLES

5d ago

OpenAI Tightens Safeguards and Slows Training After Agent Breach

Stronger isolation, automated monitoring with a 30-minute alert target, paused Astra workloads, a roughly 20% compute overhead, and a promised external postmortem aim to stop future sandbox escapes.

12 ARTICLES

last wk.

Alibaba Publishes Qwen3.8 Weights While Keeping Commercial Limits on Its 2.4T Max

4 ARTICLES

2w ago

ByteDance Reported to Be Pre-Training a Near-10 Trillion-Parameter AI Model

3 ARTICLES

4w ago

U.S. Opens Inquiry Into Chinese AI 'Distillation' as Xi Plans September 24 Washington Visit

7 ARTICLES

last mo.

Google Open-Sources DiffusionGemma, a Text-Diffusion Model That Boosts Local Inference Speed

4 ARTICLES

2mo ago

Chinese Team Opens 1.58‑Bit BitCPM-CANN and 1B MiniCPM5 Models for Phone and Ascend Deployment

5 ARTICLES

3mo ago

SpaceXAI’s Grok V9‑Medium Completes Core Training

3 ARTICLES

3mo ago

Kunlun Wanwei Releases SkyClaw Agent Models on TianGong Skywork

10 ARTICLES

3mo ago

MoE Research Delivers New Playbooks for Design, Routing, and Real-World Deployment

16 ARTICLES

3mo ago

Anthropic Links Claude’s Blackmail Failures to ‘Evil AI’ Training Data, Says New Models Pass

3 ARTICLES

3mo ago

Major Chatbots Store User Data in Three Layers With Varying Privacy Rules

8 ARTICLES

3mo ago

Musk Admits xAI Used OpenAI Models to Train Grok in Court

22 ARTICLES

3mo ago

OpenAI Explains ChatGPT’s Goblin Tic and Patches Codex to Block It

5 ARTICLES

3mo ago

OpenAI Blames Mis-Tuned Rewards for GPT 'Goblin' Habit, Adds GPT-5.5 Prompt Ban

Model Training

OpenAI Tightens Safeguards and Slows Training After Agent Breach

Alibaba Publishes Qwen3.8 Weights While Keeping Commercial Limits on Its 2.4T Max

ByteDance Reported to Be Pre-Training a Near-10 Trillion-Parameter AI Model

U.S. Opens Inquiry Into Chinese AI 'Distillation' as Xi Plans September 24 Washington Visit

Nadella Criticizes Anthropic’s Fable 5 Over Strict Safeguards

OpenAI Builds GPT-Red to Hunt Prompt‑Injection Flaws

Thinking Machines Releases Inkling, a 975B Open-Weight AI Model

Thinking Machines Releases Inkling, an Open-Weights Multimodal MoE Model

Sakana Launches Fugu Orchestrator to Coordinate Multiple LLM Agents

Sakana AI Unveils Fugu Orchestrator That Routes Tasks to Specialist Models

Never miss stories about

Model Training

Google Open-Sources DiffusionGemma, a Text-Diffusion Model That Boosts Local Inference Speed

Chinese Team Opens 1.58‑Bit BitCPM-CANN and 1B MiniCPM5 Models for Phone and Ascend Deployment

SpaceXAI’s Grok V9‑Medium Completes Core Training

Kunlun Wanwei Releases SkyClaw Agent Models on TianGong Skywork

MoE Research Delivers New Playbooks for Design, Routing, and Real-World Deployment

Anthropic Links Claude’s Blackmail Failures to ‘Evil AI’ Training Data, Says New Models Pass

Major Chatbots Store User Data in Three Layers With Varying Privacy Rules

Musk Admits xAI Used OpenAI Models to Train Grok in Court

OpenAI Explains ChatGPT’s Goblin Tic and Patches Codex to Block It

OpenAI Blames Mis-Tuned Rewards for GPT 'Goblin' Habit, Adds GPT-5.5 Prompt Ban

Never miss stories about

Model Training