HackerTrans
TopNewTrendsCommentsPastAskShowJobs

anarmorarm

no profile record

Submissions

Show HN: WaveletLM – wavelet-based, attention-free model with O(n log n) scaling

github.com
7 points·by anarmorarm·قبل 3 أشهر·1 comments

comments

anarmorarm
·قبل 3 أشهر·discuss
Edit: GPT-2, not GPT-2 Medium. The 2nd paragraph should read:

"With 23.8 PPL on WikiText-103, WaveletLM beats both GPT-2, which was trained on 80× more data, and Transformer-XL Standard, which..."
anarmorarm
·قبل 4 سنوات·discuss
This deserves to be on the front page. Fantastic read.