Models of On-Policy Residual Learning (ORL): QwenDLM, LLaDA-2.1-mini and DiffusionGemma trained with ORL.
Yifan Yu
yifanyu
AI & ML interests
None yet
Recent Activity
upvoted a paper about 9 hours ago
On-Policy Residual Learning for Diffusion Language Models updated a collection about 11 hours ago
ORL: On-Policy Residual Learning for Diffusion LMs updated a dataset about 11 hours ago
yifanyu/ORL-training-dataOrganizations
models 10
yifanyu/DiffusionGemma-26B-A4B-it-ORL-NVFP4
Text Generation • 26B • Updated • 225
yifanyu/DiffusionGemma-26B-A4B-it-ORL
Text Generation • 27B • Updated • 459
yifanyu/LLaDA2.1-mini-ORL-NVFP4
Text Generation • 9B • Updated • 273
yifanyu/LLaDA2.1-mini-ORL
Text Generation • 16B • Updated • 485
yifanyu/QwenDLM-4B-ORL
Text Generation • 4B • Updated • 468
yifanyu/QwenDLM-4B
Text Generation • 4B • Updated • 473
yifanyu/idlm-8b-n8
8B • Updated • 10
yifanyu/I-DLM-8B-lora-r128
Text Generation • Updated • 34 • 2
yifanyu/I-DLM-32B
Text Generation • 33B • Updated • 690 • 6
yifanyu/I-DLM-8B
Text Generation • 8B • Updated • 1.34k • 14