MACHINE LEARNING

→ View original post on X — @lmthang

AI Accelerates Knowledge Discovery in Mathematics and Physics

By

@lmthang

–

11 February 2026 18h04

6 months in, after the IMO-gold achievement, I’m very excited to share another important milestone: AI can help accelerate knowledge discovery in mathematics, physics, and computer science! We’re sharing Two new papers from @GoogleDeepMind and @GoogleResearch that explore how

→ View original post on X — @dair_ai

AgentSkiller: Automated Multi-Turn Agent Training Data Framework

By

@dair_ai

–

11 February 2026 17h22

// Scaling Generalist Agent Intelligence // This new research introduces AgentSkiller, a fully automated framework for synthesizing multi-turn interaction data across realistic, semantically linked domains. The architecture uses a Directed Acyclic Graph to ensure determinism and

→ View original post on X — @rasbt

Post-Training as Key Differentiator in Modern AI Development

By

@rasbt

–

11 February 2026 16h06

Good question. I would say the post-training is usually one of the biggest distinguishing contributors these days

→ View original post on X — @rasbt

GLM-4.5: Notable benchmark performance between o3 and Claude

By

@rasbt

–

11 February 2026 15h42

I think 4.5 was the first really notable one. Benchmark-wise it sat between between OpenAI o3 and Claude 4 Opus.
I have a brief section about it in my Big LLM Architecture Comparison article: https://
magazine.sebastianraschka.com/i/168650848/11
-glm-45
…

Model-Agnostic Approach Recovers 90% Sequence Packing Gains

By

–

11 February 2026 15h19

5/5 Because this approach is model-agnostic, it applies to any architecture. Even on transformers (like Qwen2.5-7B by @alibaba_cloud
, this method recovers ~90% of the gains of sequence packing, without relying on specific attention implementations. Full breakdown +

Model-Agnostic Approach: Minimizing Padding with Truncation and Batching

By

–

11 February 2026 15h19

4/5 So we took a different, model agnostic path: minimize padding before it reaches the model using truncation combined with padding-aware batching.

Sequence packing: limitations for hybrid models like Jamba

By

–

11 February 2026 15h19

3/5 Sequence packing helps for transformers, but it relies on architecture-specific support that is often missing and introduces implementation risks for non-transformer or hybrid models like @AI21Labs' Jamba.

AI21 Labs Cuts LLM Online-RL Training Time by 70% with Padding Minimization

By

–

11 February 2026 15h19

1/5 As part of our work on improving the efficiency of our LLM online-RL training pipelines, we cut policy update step time by ~70% by introducing a model-agnostic padding minimization method.

Padding Waste in RL Training: Impact on Compute and Cost

By

–

11 February 2026 15h19

2/5 When sequence lengths vary a lot, as in RL training, padding can consume a large fraction of the tokens. This wastes compute and memory and drives up training time and cost.