activity
20212026
most citedMixed-Type Tabular Data Synthesis with Score-based Diffusion in Latent Space

23 citations · 43 across the 13 of their papers we have counts for

collaborators

14 papers

cs.CL2026

When LLMs Read Tables Carelessly: Measuring and Reducing Data Referencing Errors

Yuqing Yang, Qi Zhu, Zhen Han +5

While large language models (LLMs) perform well on table tasks, they still make data referencing errors (DREs), i.e., incorrectly citing or omitting table values, despite understan…

cs.LG2026

Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, recoverable deadzone, and an irreducible floor

Xiaocan Li, Shiliang Wu, Zheng Shen

MXFP4 arithmetic can dramatically accelerate reinforcement learning (RL) post-training of large language models (LLMs), yet the quantization error introduces severe accuracy degrad…

cs.AI2026

SQL-Trail: Multi-Turn Reinforcement Learning with Interleaved Feedback for Text-to-SQL

Harper Hua, Zhen Han, Zhengyuan Shen +9

While large language models (LLMs) have substantially improved Text-to-SQL generation, a pronounced gap remains between AI systems and human experts on challenging benchmarks such…

cs.LG2025

A-3PO: Accelerating Asynchronous LLM Training with Staleness-aware Proximal Policy Approximation

Xiaocan Li, Shiliang Wu, Zheng Shen

Decoupled PPO has been a successful reinforcement learning (RL) algorithm to deal with the high data staleness under the asynchronous RL setting. Decoupled loss used in decoupled P…

cs.LG2025

The Kinetics of Reasoning: How Chain-of-Thought Shapes Learning in Transformers?

Zihan Pengmei, Costas Mavromatis, Zhengyuan Shen +3

Chain-of-thought (CoT) supervision can substantially improve transformer performance, yet the mechanisms by which models learn to follow and benefit from CoT remain poorly understo…

cs.CL2025

BoundRL: Efficient Structured Text Segmentation through Reinforced Boundary Generation

Haoyuan Li, Zhengyuan Shen, Sullam Jeoung +6

Structured texts refer to texts containing structured elements beyond plain texts, such as code snippets and placeholders. Such structured texts increasingly require segmentation i…