To appear @ #ICLR2025! We show that LMs represent semantically-equivalent inputs across languages, modalities, etc. similarly. This shared representation space is structured by the LM's dominant language, which is also relevant to recent phenomena where LMs "think" in ChineseποΈ in Englishπ contexts
22.01.2025 18:10 β π 11 π 2 π¬ 0 π 0
We have released our code at github.com/ZhaofengWu/s.... We hope that this could be useful for future studies understanding the how LMs work!
17.12.2024 15:26 β π 3 π 1 π¬ 0 π 0
Would love to be added! Thank youuu π
05.12.2024 22:09 β π 1 π 0 π¬ 0 π 0
π‘We find that models βthinkβ π in English (or in general, their dominant language) when processing distinct non-English or even non-language data types π€― like texts in other languages, arithmetic expressions, code, visual inputs, & audio inputsβΌοΈ π§΅β¬οΈ arxiv.org/abs/2411.04986
02.12.2024 18:08 β π 12 π 1 π¬ 2 π 2
Hi Marc! Do you mind adding me to the pack? Thanks!
01.12.2024 04:37 β π 1 π 0 π¬ 1 π 0
πthank you thank you!!
25.11.2024 04:14 β π 0 π 0 π¬ 0 π 0
PhD in computer vision/AI - medical imagens. Electronics projects, astronomy, and the history of science. Father, husband, electrical engineer and engineering manager.
https://www.vita-group.space/ π¨βπ« UT Austin ML Professor (on leave)
https://www.xtxmarkets.com/ π¦ XTX Markets Research Director (NYC AI Lab)
Superpower is trying everything πͺ
Newest focus: training next-generation super intelligence - Preview above πΆ
Researcher at AWS AI
https://shuaichenchang.github.io
CS PhD JohnsHopkins | Ex NLProc @ Genentech | Information Seeking | Disinformation Agents | Copilots for Social Good | PhD
@JHUCLSP @JHUMCEH
#NLProc
CS PhD @Harvard β’ Pre-doc @GoogleDeepMind β’ Anything `science', ~cosmos, and Oxford commas
PhD student USC NLP working on generalization and reasoning, prev UMassAmherst, IITG (he/him)
A LLN - large language Nathan - (RL, RLHF, society, robotics), athlete, yogi, chef
Writes http://interconnects.ai
At Ai2 via HuggingFace, Berkeley, and normal places
Postdoc in AI at the Allen Institute for AI & the University of Washington.
π https://valentinapy.github.io
PhD student/research scientist intern at UCL NLP/Google DeepMind (50/50 split). Previously MS at KAIST AI and research engineer at Naver Clova. #NLP #ML π https://soheeyang.github.io/
AI scientist & consultant :: prev Amazon Alexa, Toshiba, Cam Uni :: voice & language tech :: powered by coffee :: photographer :: Cambridge UK
https://www.catherinebreslin.co.uk
French mathematician doing XAI and Fairness for #NLProc.
Prev: PhD on CS at ANITI Toulouse & MSc on maths at Γcole polytechnique Paris.
https://fanny-jourdan.github.io/
PhD student @ UAEU & KU Leuven
I develop tough benchmarks for LMs and then I build agents to try and beat those benchmarks. Postdoc @ Princeton University.
https://ofir.io/about
π PhD student @uwcse @uwnlp. π© Private pilot. Previously: π§βπ» @oculus, π @IllinoisCS. π π₯Ύ π΄ββοΈ π΅ β οΈ
PhD student at NYU, working on NLP.
https://timchen0618.github.io
ML PhD @ CMU | Prev: research intern @ Microsoft Research New England, Amazon, Bosch.
Final year Ph.D. candidate in NLP, CV at JHU. Researching reasoning systems, multimodality, and AI for science. On the job market for full-time industry positions! #NLProc
https://katesanders9.github.io/
Applied Scientist @ Amazon
(Posts are my own opinion)
Previously PhD@JHU
