🔄 In a Training Loop

weiliu

thinkwee

5 28 17

https://thinkwee.top/about/

AI & ML interests

LLM reasoning, agents

Recent Activity

submitted a paper 27 days ago

Large Language Models Hack Rewards, and Society

upvoted a paper 27 days ago

Large Language Models Hack Rewards, and Society

upvoted a paper 30 days ago

Linear Ensembles Wash Away Watermarks: On the Fragility of Distributional Perturbations in LLMs

View all activity

Organizations

None yet

Collections 3

View 3 collections

Papers 13

spaces 4

BibGuard

🛡

Generate a bibliography health report from .bib and .tex files

DDR Bench

🚀

Deep Data Research Benchmark

NOVER

🧠

Reasoning on ANYTHING

IAgents

🧠

Connecting Humans and Agents

models 2

thinkwee/NOVER1-Qwen2.5-7B

Question Answering • 8B • Updated Aug 20, 2025 • 7 • 2

thinkwee/NOVER1-Qwen3-4B

Question Answering • 4B • Updated Aug 20, 2025 • 8 • 2

datasets 5

weiliu

AI & ML interests

Recent Activity

Organizations

Collections 3

Hunt Instead of Wait: Evaluating Deep Data Research on Large Language Models

DDR Bench

thinkwee/DDRBench_10K

thinkwee/DDRBench_10K_trajectory

NOVER: Incentive Training for Language Models via Verifier-Free Reinforcement Learning

thinkwee/NOVER1-Qwen2.5-7B

thinkwee/NOVER1-Qwen3-4B

Hunt Instead of Wait: Evaluating Deep Data Research on Large Language Models

DDR Bench

thinkwee/DDRBench_10K

thinkwee/DDRBench_10K_trajectory

NOVER: Incentive Training for Language Models via Verifier-Free Reinforcement Learning

thinkwee/NOVER1-Qwen2.5-7B

thinkwee/NOVER1-Qwen3-4B

Papers 13

spaces 4

BibGuard

DDR Bench

NOVER

IAgents

models 2

thinkwee/NOVER1-Qwen2.5-7B

thinkwee/NOVER1-Qwen3-4B

datasets 5

thinkwee/DDRBench_10K_trajectory

thinkwee/DDRBench_10K

thinkwee/NOVEReason_full

thinkwee/NOVEReason_2k

thinkwee/NOVEReason_5k

weiliu

AI & ML interests

Recent Activity

Organizations

Collections 3

DDR Bench

DDR Bench

Papers 13

spaces 4 Sort: Recently updated

BibGuard

DDR Bench

NOVER

IAgents

models 2 Sort: Recently updated

datasets 5 Sort: Recently updated

spaces 4

models 2

datasets 5