Knowledge Engineer Group @ Tsinghua University

university

https://keg.cs.tsinghua.edu.cn/

Activity Feed Request to join this org

AI & ML interests

None defined yet.

Recent Activity

mwh24 published a model about 21 hours ago

THU-KEG/SAEVerbalizer-Adapter-1B-to-27B

mwh24 updated a collection about 21 hours ago

mwh24 updated a collection about 21 hours ago

View all activity

Papers

Memory for Large Language Models

EurekAgent: Agent Environment Engineering is All You Need For Autonomous Scientific Discovery

View all Papers

published a model about 21 hours ago

THU-KEG/SAEVerbalizer-Adapter-1B-to-27B

Updated 6 days ago

updated a collection about 21 hours ago

SAEVerbalizer

SAEVerbalizer: Generating Explanations for Sparse Autoencoder Features via Representation Verbalization • 3 items • Updated about 21 hours ago

updated a dataset 6 days ago

THU-KEG/SAEVerbalizer-Data

Viewer • Updated 6 days ago • 4.4k • 84

updated a model 6 days ago

THU-KEG/SAEVerbalizer-Adapter-1B-to-27B

Updated 6 days ago

submitted 3 papers to Daily Papers 18 days ago

HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?

Paper • 2609.01437 • Published 20 days ago • 119

Aspire: Can Models Self-Evolve from Vague Goals?

Paper • 2608.31111 • Published 21 days ago • 83

S3Gym: Can LLMs Turn Self-Testing and Self-Judging into Self-Improvement?

Paper • 2608.31100 • Published 21 days ago • 41

published a dataset 19 days ago

THU-KEG/SAEVerbalizer-Data

Viewer • Updated 6 days ago • 4.4k • 84

published a model 19 days ago

THU-KEG/SAEVerbalizer-27B

Text Generation • 27B • Updated 19 days ago • 576

updated a model 19 days ago

THU-KEG/SAEVerbalizer-27B

Text Generation • 27B • Updated 19 days ago • 576

submitted a paper to Daily Papers 20 days ago

CogEvol: Towards Efficient and Reliable Learning Environment Generation

Paper • 2608.30968 • Published 21 days ago • 32

submitted a paper to Daily Papers about 2 months ago

Memory for Large Language Models

Paper • 2607.25380 • Published Jul 28 • 13

authored a paper 3 months ago

EurekAgent: Agent Environment Engineering is All You Need For Autonomous Scientific Discovery

Paper • 2606.13662 • Published Jun 11 • 33

authored a paper 3 months ago

EurekAgent: Agent Environment Engineering is All You Need For Autonomous Scientific Discovery

Paper • 2606.13662 • Published Jun 11 • 33

submitted a paper to Daily Papers 3 months ago

EurekAgent: Agent Environment Engineering is All You Need For Autonomous Scientific Discovery

Paper • 2606.13662 • Published Jun 11 • 33

submitted a paper to Daily Papers 4 months ago

Diagnosing Harmful Continuation in Answer-Correct Long-CoT Training Traces

Paper • 2605.29288 • Published May 28 • 8

updated a collection 4 months ago

LongTraceRL

LongTraceRL: Learning Long-Context Reasoning from Search Agent Trajectories with Rubric Rewards • 5 items • Updated Jun 1 • 1

updated a model 4 months ago

THU-KEG/LongTraceRL-30B

Reinforcement Learning • 31B • Updated Jun 1 • 8 • 1