Post-Trained MoE Can Skip Half Experts via Self-Distillation
Paper • 2605.18643 • Published • 30
dynn-datasets
university
AI & ML interests
None defined yet.
Recent Activity
View all activity
XingtaiHF
authored a
paper 22 days ago
youyc22
authored a
paper 22 days ago
bambisheng
authored a
paper 22 days ago
XingtaiHF
submitted a
paper to Daily Papers 22 days ago
bambisheng
updated a
dataset 3 months ago
XingtaiHF
updated a
dataset 3 months ago
Upload aime-2026.jsonl
#1 opened 3 months ago
by 
XingtaiHF
luoxueheng
updated a
dataset 3 months ago
bambisheng
published a
dataset 3 months ago
XingtaiHF
authored a
paper 3 months ago
youyc22
authored a
paper 7 months ago
XingtaiHF
authored 2
papers 9 months ago
youyc22
authored a
paper about 1 year ago
bambisheng
authored a
paper about 1 year ago
XingtaiHF
authored 5
papers about 1 year ago
OpenDelta: A Plug-and-play Library for Parameter-efficient Adaptation of Pre-trained Models
Paper • 2307.03084 • Published • 1
Mastering Text, Code and Math Simultaneously via Fusing Highly Specialized Language Models
Paper • 2403.08281 • Published
Intuitive Fine-Tuning: Towards Unifying SFT and RLHF into a Single Process
Paper • 2405.11870 • Published
UltraMedical: Building Specialized Generalists in Biomedicine
Paper • 2406.03949 • Published
Fast and Slow Generating: An Empirical Study on Large and Small Language Models Collaborative Decoding
Paper • 2406.12295 • Published