Data and models for the paper: Computer Environments Elicit General Agentic Intelli. Feel free to open an issue if you have any questions or problems!
Daixuan Cheng
daixuancheng
AI & ML interests
I study LLMs, from Pre-Training to Agent.
Recent Activity
authored a paper 1 day ago
From Trial-and-Error to Improvement: A Systematic Analysis of LLM
Exploration Mechanisms in RLVR authored a paper 1 day ago
ClawGym: A Scalable Framework for Building Effective Claw Agents commentedon a paper 1 day ago
ClawGym: A Scalable Framework for Building Effective Claw AgentsOrganizations
None yet
models 144
daixuancheng/4-26-debugAdv_docker_rbs8_temp0.7_model-30BA3B-Sft_step99
31B • Updated • 12
daixuancheng/4-26-debugAdv_docker_rbs8_temp0.7_model-30BA3B-Sft_step39
31B • Updated • 15
daixuancheng/4-26-debugAdv_docker_rbs8_temp0.7_model-30BA3B-Sft_step9
31B • Updated • 16
daixuancheng/4-24-30BA3B-s233-final
31B • Updated • 15
daixuancheng/4-23_docker_rbs32_model-Coder_step15
Text Generation • 31B • Updated • 397
daixuancheng/0418_17K_filtered_10K_reward0.5_4b_inst
4B • Updated • 16
daixuancheng/4b-0404-2k
4B • Updated • 14
daixuancheng/Qwen3-4B-Instruct-2507-LLM-in-Sandbox-RL
4B • Updated • 101 • 1
daixuancheng/Qwen3-VL-8B-Thinking_stage1_MixAllRL
9B • Updated • 1
daixuancheng/Qwen3-VL-8B-Thinking_stage3_MixAllRL_and_dataMixRatio_and_easy2hard
9B • Updated • 3