Topic · 45 recaps

LLM Pretraining & Post-training

How large language models are built: data curation and mixtures, pretraining objectives, scaling laws, instruction tuning, preference learning, and the full post-training stack.

Sort

Newest

J-Zero: Unified Challenger--Solver--Judge Co-Evolution from Zero Data

Evaluation · Aug 27

Puro-2B: Poor Lab's Qwen2-1.5B Trained on RTX 5090 within $5090

Inference Optimization · Aug 27

Fast Weight Attention for Continual Learning

LLM Training · Aug 27

TLive-Omni: An Omni-Modal Understanding Model for E-Commerce Live Streaming

Inference Optimization · Aug 21

DART-SD: Diamond-topology Aware Retrieval and Tuning for Self-Distillation of Multi-Turn Tool-Calling Agents

Agents · Aug 19

Unlocking the Potential of Image Editing via Concept Scaling and Dense Supervision

Image Generation · Aug 17

On-Policy Self-Distillation without Any Supervision

LLM Training · Aug 9

SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs

LLM Training · Aug 6

AgentOPSD: Recursive Self-Distillation for Agentic Reinforcement Learning

Agents · Aug 6

Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes

LLM Training · Aug 5