Xingxing Zhang

Researcher@Microsoft Research Asia·中国·OpenReview

研究方向

synthetic data · post-training · reasoning for large language models · Large Language Models · Transformer Architecture · Pre-training · Text Summarization · Text Generation · Natural Language Processing · Machine Learning

5.0

Synthetic Data (Almost) from Scratch: Generalized Instruction Tuning for Language Models

合作者 (20)

Xingxing Zhang

Learning To Draft: Adaptive Speculative Decoding with Reinforcement Learning

WildLong: Synthesizing Realistic Long-Context Instruction Data at Scale

QueST: Incentivizing LLMs to Generate Difficult Problems

Two Pathways to Truthfulness: On the Intrinsic Encoding of LLM Hallucinations

Think Only When You Need with Large Hybrid-Reasoning Models

Preference Optimization for Reasoning with Pseudo Feedback

Self-Boosting Large Language Models with Synthetic Preference Data

Scaling Laws of Synthetic Data for Language Model

Synthetic Data (Almost) from Scratch: Generalized Instruction Tuning for Language Models