-
oaimli/pgpo_grpo_full_scitrek_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 160 -
oaimli/pgpo_grpo_full_scitrek_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 32 -
oaimli/pgpo_grpo_proxy_scitrek_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 274 -
oaimli/pgpo_grpo_proxy_scitrek_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 90
Miao Li
oaimli
AI & ML interests
Natural Language Processing
Recent Activity
updated a collection about 18 hours ago
LongPT updated a model about 18 hours ago
oaimli/sft_longpt_full_qwen3_4b_instruct_2507_no_starting_think published a model about 18 hours ago
oaimli/sft_longpt_full_qwen3_4b_instruct_2507_no_starting_thinkOrganizations
None yet
-
oaimli/pgpo_grpo_full_scitrek_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 160 -
oaimli/pgpo_grpo_full_scitrek_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 32 -
oaimli/pgpo_grpo_proxy_scitrek_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 274 -
oaimli/pgpo_grpo_proxy_scitrek_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 90
Long-context post-training
models 35
oaimli/sft_longpt_full_qwen3_4b_instruct_2507_no_starting_think
Text Generation • 4B • Updated
oaimli/sft_longpt_full_gemma3_4b_it_no_starting_think
Image-Text-to-Text • 4B • Updated
oaimli/grpo_longpt_direct_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 9
oaimli/scitrek_grpo_full_longrlvr_qwen3_4b_instruct_2507_660
Text Generation • 4B • Updated • 31
oaimli/scitrek_grpo_full_longrlvr_qwen3_4b_instruct_2507_280
Text Generation • 4B • Updated • 62
oaimli/pgpo_grpo_proxy_scitrek_gemma3_4b_it
Image-Text-to-Text • 4B • Updated • 90
oaimli/scitrek_grpo_full_loongrl_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 430
oaimli/pgpo_grpo_proxy_scitrek_qwen3_4b_instruct_2507
Text Generation • 4B • Updated • 274
oaimli/longtune_scitrek_grounding_reinforcement_gemma_0
Image-Text-to-Text • 4B • Updated • 13
oaimli/longtune_scitrek_simple_sft_gemma
Image-Text-to-Text • 4B • Updated • 18