arxiv:2607.15161
Sangdoo Yun
oodgnas
AI & ML interests
None yet
Recent Activity
upvoted a paper 3 days ago
On-Policy Distillation with Negative-Policy Rollouts upvoted a paper 3 days ago
DLoop: Looped Speculative Decoding upvoted a paper 29 days ago
Revisiting Complete Reasoning Traces for Post-TrainingOrganizations
models 33
oodgnas/Qwen3-1.7B-SFT-100k-MTP-s2-w005
2B • Updated • 5
oodgnas/Qwen3-1.7B-SFT-100k
2B • Updated • 9
oodgnas/Qwen3-1.7B-Base
2B • Updated • 8
oodgnas/Qwen2.5-1.5B-Instruct-SFT-30k-NTP-MTP-w001
2B • Updated • 3
oodgnas/Qwen2.5-1.5B-Instruct-SFT-30k-MTP-s3-w001
2B • Updated • 7
oodgnas/Qwen2.5-1.5B-Instruct-SFT-30k-NTP-s1-w001
2B • Updated • 11
oodgnas/Qwen2.5-1.5B-Instruct-SFT-30k-MTP-w001
2B • Updated • 5
oodgnas/Qwen2.5-1.5B-Instruct-SFT-30k-MTP-w005
2B • Updated • 2
oodgnas/Qwen2.5-1.5B-Instruct-SFT-30k
2B • Updated • 3
oodgnas/Qwen3-1.7B-OLMo3-midtrain-mtp-pos21
2B • Updated • 6
datasets 0
None public yet