honored to have given a plenary address at the Society for Language Development annual symposium titled "Whence insights? The value of delineating human and machine CogSci". It's a synthesis of a few years of thoughts, recently concretized with @aditya-yedetore.bsky.social and @kanishka.bsky.social
November 9, 2025 at 1:27 PM
honored to have given a plenary address at the Society for Language Development annual symposium titled "Whence insights? The value of delineating human and machine CogSci". It's a synthesis of a few years of thoughts, recently concretized with @aditya-yedetore.bsky.social and @kanishka.bsky.social
hello NAACL friends I'm giving a keynote today at RepL4NLP at 1:30PM local time, come say hi! I'll mostly be musing about things with light research discussions
May 4, 2025 at 1:48 PM
hello NAACL friends I'm giving a keynote today at RepL4NLP at 1:30PM local time, come say hi! I'll mostly be musing about things with light research discussions
"Gaming Linguists" good bigram
February 4, 2025 at 7:42 PM
"Gaming Linguists" good bigram
have many tasks but immobilized by car
December 1, 2024 at 5:41 PM
have many tasks but immobilized by car
this is a first... suffering maximized...
November 22, 2024 at 9:15 PM
this is a first... suffering maximized...
so proud of Hayley Ross for getting a best paper award at GenBench this year!! 🎉🪅🎉 I'm sure Kate would be too :) check out our paper and share if you think homemade cats are cats!
arxiv.org/abs/2410.17482
arxiv.org/abs/2410.17482
November 16, 2024 at 11:47 PM
so proud of Hayley Ross for getting a best paper award at GenBench this year!! 🎉🪅🎉 I'm sure Kate would be too :) check out our paper and share if you think homemade cats are cats!
arxiv.org/abs/2410.17482
arxiv.org/abs/2410.17482
(3) Cat(s)! And also maybe science.
Thanks for reading all the way through if you did!
Thanks for reading all the way through if you did!
November 8, 2024 at 5:04 PM
(3) Cat(s)! And also maybe science.
Thanks for reading all the way through if you did!
Thanks for reading all the way through if you did!
(2) Potentially working together in my lab (or via other collaborations)! I'm sure lab members (Yukyung Lee, Yulu Qin, Hayley Ross & Aditya Yedetore) will also be around & willing to chat :)
najoung.kim/students/
najoung.kim/students/
November 8, 2024 at 5:04 PM
(2) Potentially working together in my lab (or via other collaborations)! I'm sure lab members (Yukyung Lee, Yulu Qin, Hayley Ross & Aditya Yedetore) will also be around & willing to chat :)
najoung.kim/students/
najoung.kim/students/
11/16 Fri 10:00 (Brickell): @HayleyRossLing
will give a talk at GenBench on "Is artificial intelligence still intelligence? LLMs generalize to novel adjective-noun pairs, but don’t mimic the full human distribution" where we find that LLMs do often generalize compositionally to unseen AN combos!
will give a talk at GenBench on "Is artificial intelligence still intelligence? LLMs generalize to novel adjective-noun pairs, but don’t mimic the full human distribution" where we find that LLMs do often generalize compositionally to unseen AN combos!
November 8, 2024 at 5:01 PM
11/16 Fri 10:00 (Brickell): @HayleyRossLing
will give a talk at GenBench on "Is artificial intelligence still intelligence? LLMs generalize to novel adjective-noun pairs, but don’t mimic the full human distribution" where we find that LLMs do often generalize compositionally to unseen AN combos!
will give a talk at GenBench on "Is artificial intelligence still intelligence? LLMs generalize to novel adjective-noun pairs, but don’t mimic the full human distribution" where we find that LLMs do often generalize compositionally to unseen AN combos!
11/12 Tues 11:00 (Riverfront Hall, Session A): Aditya Yedetore and I will present "Semantic Training Signals Promote Hierarchical Syntactic Generalization in Transformers" where we find that training on mapping from form to meaning leads to improved hierarchical generalization.
November 8, 2024 at 5:00 PM
11/12 Tues 11:00 (Riverfront Hall, Session A): Aditya Yedetore and I will present "Semantic Training Signals Promote Hierarchical Syntactic Generalization in Transformers" where we find that training on mapping from form to meaning leads to improved hierarchical generalization.