Ada
@adadtur.bsky.social
she/her
McGillNLP & Mila
'can you be an expat if you were never a pat?' - leo hertzler
occasionally live on ckut 90.3 fm :-)
adadtur.github.io
McGillNLP & Mila
'can you be an expat if you were never a pat?' - leo hertzler
occasionally live on ckut 90.3 fm :-)
adadtur.github.io
We won a Senior Area Chair Award at NAACL!! Many thanks again to my amazing coauthors Gaurav Kamath and @sivareddyg.bsky.social :-)
May 3, 2025 at 3:51 PM
We won a Senior Area Chair Award at NAACL!! Many thanks again to my amazing coauthors Gaurav Kamath and @sivareddyg.bsky.social :-)
1. While models align with human preferences overall, base models do so more strongly than instruction-tuned counterparts
2. While different metrics of constituent complexity inform model preferences, syllable length is often the most informative metric, surprisingly more so than token length
2. While different metrics of constituent complexity inform model preferences, syllable length is often the most informative metric, surprisingly more so than token length
February 19, 2025 at 7:34 PM
1. While models align with human preferences overall, base models do so more strongly than instruction-tuned counterparts
2. While different metrics of constituent complexity inform model preferences, syllable length is often the most informative metric, surprisingly more so than token length
2. While different metrics of constituent complexity inform model preferences, syllable length is often the most informative metric, surprisingly more so than token length
Super excited to finally announce our NAACL 2025 main conference paper “Language Models Largely Exhibit Human-like Constituent Ordering Preferences”!
We examine constituent ordering preferences between humans and LLMs; we present two main findings… 🧵
We examine constituent ordering preferences between humans and LLMs; we present two main findings… 🧵
February 19, 2025 at 7:31 PM
Super excited to finally announce our NAACL 2025 main conference paper “Language Models Largely Exhibit Human-like Constituent Ordering Preferences”!
We examine constituent ordering preferences between humans and LLMs; we present two main findings… 🧵
We examine constituent ordering preferences between humans and LLMs; we present two main findings… 🧵