-
oaimli/pgpo_grpo_full_scitrek_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 14 -
oaimli/pgpo_grpo_full_scitrek_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 4 -
oaimli/pgpo_grpo_proxy_scitrek_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 12 -
oaimli/pgpo_grpo_proxy_scitrek_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 10
Miao Li
oaimli
AI & ML interests
Natural Language Processing
Recent Activity
liked a dataset 3 days ago
oaimli/SciTrek liked a dataset 3 days ago
oaimli/ProxyCoT-HotpotQA liked a dataset 3 days ago
oaimli/ProxyCoT-SciTrekOrganizations
None yet
PGPO
-
oaimli/pgpo_grpo_full_scitrek_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 14 -
oaimli/pgpo_grpo_full_scitrek_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 4 -
oaimli/pgpo_grpo_proxy_scitrek_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 12 -
oaimli/pgpo_grpo_proxy_scitrek_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 10
LongPT
Long-context post-training
models 37
oaimli/sft_longpt_full_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 143 • 1
oaimli/sft_longpt_full_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 101
oaimli/sft_longpt_full_qwen3_4b_instruct_2507_no_starting_think
Text Generation • 4B • Updated • 10
oaimli/sft_longpt_full_gemma3_4b_it_no_starting_think
Image-Text-to-Text • 4B • Updated • 10
oaimli/grpo_longpt_direct_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 12
oaimli/scitrek_grpo_full_longrlvr_qwen3_4b_instruct_2507_660
Text Generation • 4B • Updated • 9
oaimli/scitrek_grpo_full_longrlvr_qwen3_4b_instruct_2507_280
Text Generation • 4B • Updated • 10
oaimli/pgpo_grpo_proxy_scitrek_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 10
oaimli/scitrek_grpo_full_loongrl_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 18
oaimli/pgpo_grpo_proxy_scitrek_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 12