Zheng Liu

starriver030515's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper 4 days ago

Tracing the Roots: A Multi-Agent Framework for Uncovering Data Lineage in Post-Training LLMs

Paper • 2604.10480 • Published 6 days ago • 19

upvoted a paper 9 days ago

OpenWorldLib: A Unified Codebase and Definition of Advanced World Models

Paper • 2604.04707 • Published 12 days ago • 200

upvoted a paper 11 days ago

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

Paper • 2604.04771 • Published 12 days ago • 120

upvoted a paper 15 days ago

DataFlex: A Unified Framework for Data-Centric Dynamic Training of Large Language Models

Paper • 2603.26164 • Published 21 days ago • 353

upvoted a paper about 1 month ago

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

Paper • 2603.07223 • Published Mar 7 • 13

upvoted a collection 3 months ago

MMFineReason

Collection

High-quality STEM reasoning dataset for Multimodal LLM post-training. • 8 items • Updated 17 days ago • 22

upvoted 2 papers 3 months ago

MMFineReason: Closing the Multimodal Reasoning Gap via Open Data-Centric Methods

Paper • 2601.21821 • Published Jan 29 • 62

Scientific Image Synthesis: Benchmarking, Methodologies, and Downstream Utility

Paper • 2601.17027 • Published Jan 17 • 42

upvoted a collection 3 months ago

ODA-Mixture

Collection

High-quality mixture datasets for post-training covering multiple domains. • 7 items • Updated 17 days ago • 5

upvoted 2 papers 3 months ago

ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratch

Paper • 2601.13606 • Published Jan 20 • 12

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

Paper • 2601.09733 • Published Dec 30, 2025 • 9

upvoted a collection 3 months ago

ChartVerse

Collection

8 items • Updated Mar 2 • 9

upvoted 2 papers 4 months ago

DataFlow: An LLM-Driven Framework for Unified Data Preparation and Workflow Automation in the Era of Data-Centric AI

Paper • 2512.16676 • Published Dec 18, 2025 • 222

OpenDataArena: A Fair and Open Arena for Benchmarking Post-Training Dataset Value

Paper • 2512.14051 • Published Dec 16, 2025 • 47

upvoted a paper 6 months ago

Scaling Code-Assisted Chain-of-Thoughts and Instructions for Model Reasoning

Paper • 2510.04081 • Published Oct 5, 2025 • 23

upvoted 2 papers 7 months ago

MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing

Paper • 2509.22186 • Published Sep 26, 2025 • 159

From Uniform to Heterogeneous: Tailoring Policy Optimization to Every Token's Nature

Paper • 2509.16591 • Published Sep 20, 2025 • 2

upvoted a paper 10 months ago

Reinforcement Pre-Training

Paper • 2506.08007 • Published Jun 9, 2025 • 265

upvoted 2 papers 11 months ago

Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning

Paper • 2506.01939 • Published Jun 2, 2025 • 190

Jigsaw-R1: A Study of Rule-based Visual Reinforcement Learning with Jigsaw Puzzles

Paper • 2505.23590 • Published May 29, 2025 • 25