VisPhyWorld
Collection
An execution-based framework that evaluates physical reasoning by generating executable simulator code from visual inputs. • 4 items • Updated
AI & ML interests
Natural Language Processing, Image Generation
Recent Activity
Papers
RationalRewards: Reasoning Rewards Scale Visual Generation Both Training and Test Time
OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis
lllqaq
updated a
collection 1 day ago
lllqaq
updated a
Space 1 day ago
lllqaq
published a
Space 1 day ago
lllqaq
updated a
collection 1 day ago
lllqaq
updated a
dataset 1 day ago
lllqaq
published a
dataset 1 day ago
JasperHaozhe
updated 3
datasets 3 days ago
JasperHaozhe
authored 11
papers 7 days ago
Emergent Hierarchical Reasoning in LLMs through Reinforcement Learning
Paper • 2509.03646 • Published • 33
Reverse-Engineered Reasoning for Open-Ended Generation
Paper • 2509.06160 • Published • 151
VideoScore2: Think before You Score in Generative Video Evaluation
Paper • 2509.22799 • Published • 26
Dr. Bench: A Multidimensional Evaluation for Deep Research Agents, from Answers to Reports
Paper • 2510.02190 • Published • 19
Infinity Parser: Layout Aware Reinforcement Learning for Scanned Document Parsing
Paper • 2510.15349 • Published
TR-DQ: Time-Rotation Diffusion Quantization
Paper • 2503.06564 • Published
From Illusion to Intention: Visual Rationale Learning for Vision-Language Reasoning
Paper • 2511.23031 • Published • 1
CogDoc: Towards Unified thinking in Documents
Paper • 2512.12658 • Published
EvoCUA: Evolving Computer Use Agents via Learning from Scalable Synthetic Experience
Paper • 2601.15876 • Published • 92
Understanding by Reconstruction: Reversing the Software Development Process for LLM Pretraining
Paper • 2603.11103 • Published • 9
SWE-QA-Pro: A Representative Benchmark and Scalable Training Recipe for Repository-Level Code Understanding
Paper • 2603.16124 • Published • 3