Ofir Press(@OfirPress):I started doing language modeling research in Jan 2016, using the LSTM from the "RNN Regularization" model by @woj_zaremba @OriolVinyalsML and @ilyasut, trying to improve its perplexity on Penn Treebank (1M toks). This led to the weight tying method, sometimes still used today.

Ofir Press

@OfirPress

I push the AI frontier by building tough benchmarks with amazing people. SWE-bench, ProgramBench, SWE-agent, SciCode. Postdoc @Princeton. PhD @nlpnoah @UW.

加入 June 2016

9.8K 正在關注 19.8K 粉絲

Ofir Press@OfirPress

2026.06.26 15:55

I started doing language modeling research in Jan 2016, using the LSTM from the "RNN Regularization" model by @woj_zaremba @OriolVinyalsML and @ilyasut, trying to improve its perplexity on Penn Treebank (1M toks). This led to the weight tying method, sometimes still used today.

214

轉發到社區

熱門用戶