activity
20242026
collaborators

7 papers

cs.LG2026

PithTrain: A Compact and Agent-Native MoE Training System

Ruihang Lai, Hao Kang, Haozhan Tang +6

Mixture-of-Experts (MoE) has become the dominant architecture for frontier language models. To meet this demand, production frameworks have built optimized MoE training stacks over…

cs.CL2026

Generating Pretraining Tokens from Organic Data for Data-Bound Scaling

Zichun Yu, Chenyan Xiong

LLM pretraining is shifting from a compute-bound to a data-bound regime, where available human (organic) text falls far short of scaling demands. However, reaching the data-bound r…

cs.CL2026

Efficient Multi-Agent System Training with Data Influence-Oriented Tree Search

Wentao Shi, Zichun Yu, Fuli Feng +2

Monte Carlo Tree Search (MCTS) based methods provide promising approaches for generating synthetic data to enhance the self-training of Large Language Model (LLM) based multi-agent…

cs.CL2025

RePro: Training Language Models to Faithfully Recycle the Web for Pretraining

Zichun Yu, Chenyan Xiong

High-quality pretraining data is the fossil fuel of large language models (LLMs), yet its reserves are running low for frontier models. In this paper, we introduce RePro, a novel w…

cs.CL2025

FLAME-MoE: A Transparent End-to-End Research Platform for Mixture-of-Experts Language Models

Hao Kang, Zichun Yu, Chenyan Xiong

Recent large language models such as Gemini-1.5, DeepSeek-V3, and Llama-4 increasingly adopt Mixture-of-Experts (MoE) architectures, which offer strong efficiency-performance trade…

cs.CL2024

MATES: Model-Aware Data Selection for Efficient Pretraining with Data Influence Models

Zichun Yu, Spandan Das, Chenyan Xiong

Pretraining data selection has the potential to improve language model pretraining efficiency by utilizing higher-quality data from massive web data corpora. Current data selection…