Xiyao Wang
russwang
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
4 days ago
SPICE: Self-Play In Corpus Environments Improves Reasoning
upvoted
a
paper
5 days ago
PRISM-Bench: A Benchmark of Puzzle-Based Visual Tasks with CoT Error
Detection
upvoted
a
paper
5 days ago
Vision-Zero: Scalable VLM Self-Improvement via Strategic Gamified
Self-Play
Organizations
-
SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvement
Paper • 2504.07934 • Published • 20 -
russwang/ThinkLite-VL-7B
8B • Updated • 1.74k • 15 -
russwang/ThinkLite-VL-70k
Viewer • Updated • 70k • 166 • 5 -
russwang/ThinkLite-VL-hard-11k
Viewer • Updated • 11k • 97 • 3
-
SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvement
Paper • 2504.07934 • Published • 20 -
russwang/ThinkLite-VL-7B
8B • Updated • 1.74k • 15 -
russwang/ThinkLite-VL-70k
Viewer • Updated • 70k • 166 • 5 -
russwang/ThinkLite-VL-hard-11k
Viewer • Updated • 11k • 97 • 3
datasets
27
russwang/mixed_bg_136k_rdm
Viewer
•
Updated
•
136k
•
7
russwang/mixed_bg_136k
Viewer
•
Updated
•
136k
•
10
russwang/Math-Reasoning-BG
Viewer
•
Updated
•
97.3k
•
31
russwang/Math-Reasoning-Mixed-100k
Viewer
•
Updated
•
100k
•
9
russwang/Math-Reasoning-Mixed-116k
Viewer
•
Updated
•
116k
•
30
russwang/Math-Reasoning
Viewer
•
Updated
•
94.6k
•
26
russwang/Mixed-Critic-Thinklite-Hard
Viewer
•
Updated
•
51.2k
•
14
russwang/ThinkLite-70k-new
Viewer
•
Updated
•
70k
•
8
russwang/MMMU_diff
Viewer
•
Updated
•
100
•
16
russwang/MathVista_diff
Viewer
•
Updated
•
84
•
10