Hanyu Wang's picture

7

Hanyu Wang

hywww24

AI & ML interests

Diffusion Models, LLMs

Recent Activity

upvoted a paper 3 days ago

When Does Trajectory-Level Supervision Permit Efficient Offline Reinforcement Learning?

authored a paper 25 days ago

SkillGrad: Optimizing Agent Skills Like Gradient Descent

upvoted a paper 25 days ago

The Illusion of Reasoning: Exposing Evasive Data Contamination in LLMs via Zero-CoT Truncation

View all activity

Organizations

None yet

Papers 1

arxiv:2605.27760

models 1

hywww24/llama2-7b-chat-iti

Updated Oct 22, 2024 • 1

datasets 3

hywww24/mistral-v0.3-tqa-seed42-greedy-probe-layer13

Viewer • Updated Nov 5, 2024 • 817 • 3 • 1

hywww24/llama-3-tqa-seed42-greedy-probe-layer13

Viewer • Updated Nov 5, 2024 • 817 • 4 • 1

hywww24/win_lose_pairs_tqa_layer13

Viewer • Updated Oct 22, 2024 • 817 • 16 • 1