collaborators

10 papers

cs.LG2026

ARBITER: Reasoning Trajectory Basins and Majority Vote Failures in Test-Time Sampling

Meng Cai, Lars Kulik, Farhana Choudhury

When language models use test-time sampling, they generate multiple reasoning trajectories and select an answer by majority vote. We show that these trajectories are not independen…

cs.CV2026

ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratch

Zheng Liu, Honglin Lin, Chonghan Qin +13

Chart reasoning is a critical capability for Vision Language Models (VLMs). However, the development of open-source models is severely hindered by the lack of high-quality training…

cs.CL2026

Heterogeneous Adaptive Policy Optimization: Tailoring Optimization to Every Token's Nature

Zheng Liu, Mengjie Liu, Siwei Wen +4

Using entropy as a measure of heterogeneity to guide optimization has emerged as a crucial research direction in Reinforcement Learning for LLMs. However, existing methods typicall…

cs.CV2026

MinerU2.5-Pro: Pushing the Limits of Data-Centric Document Parsing at Scale

Bin Wang, Tianyao He, Linke Ouyang +40

Current document parsing methods advance primarily through model architecture innovation, while systematic engineering of training data remains underexplored. Yet state-of-the-art…

cs.LG2026

Unlocking Data Value in Finance: A Study on Distillation and Difficulty-Aware Training

Chuxue Cao, Honglin Lin, Zhanping Zhong +5

Large Language Models (LLMs) have demonstrated strong general capabilities, yet their deployment in finance remains challenging due to dense domain-specific terminology, stringent…

cs.CL2025

Closing the Data Loop: Using OpenDataArena to Engineer Superior Training Datasets

Xin Gao, Xiaoyang Wang, Yun Zhu +3

The construction of Supervised Fine-Tuning (SFT) datasets is a critical yet under-theorized stage in the post-training of Large Language Models (LLMs), as prevalent practices often…