LatticeTrack papers across labs, topics, and trends.

Papers Digest Topics Labs Collections FAQ

Created by Flynn Lachendro

Papers Digest Topics Labs Saved

Search

Search papers, labs, and topics across Lattice.

Built by Flynn Lachendro·𝕏 / Twitter·RSS··FAQ·Glossary·Privacy

Xingyu Lin

Normal University

Papers on Lattice

2

Total citations

0

Topics

3

Publication activitypapers/week, last 8 weeks

Research focus

Reasoning & Chain-of-Thought (2)Eval Frameworks & Benchmarks (1)RLHF & Preference Learning (1)

Frequent co-authors

Dehui Du (1)Yilin Wen (1)Du Su (1)Jinchang Hou (1)En Wang (1)Wenbin Liu (1)Chenfu Bao (1)Zhonghou Lv (1)

Papers (2)

Sep 20, 2026

Normal University3w ago

Collapse, Not Complexity: Failure-Conditioned Decomposition Repair for End-to-End Document Parsing

Xingyu Lin, Dehui Du

Eval Frameworks & Benchmarks Reasoning & Chain-of-Thought

Apr 14, 2026

Normal UniversityApr 14, 2026·also Affiliation:

Token-Level Policy Optimization: Linking Group-Level Rewards to Token-Level Aggregation via Sequence-Level Likelihood

LLMs can now learn mathematical reasoning 2x faster and with greater stability, thanks to a new token-level policy optimization method.

Xingyu Lin, Yilin Wen, Du Su +5

Reasoning & Chain-of-Thought RLHF & Preference Learning