Alan Li
@alanli2020
Followers
72
Following
96
Statuses
32
PhD @YaleNLP | undergrad @UWcse @nlpnoah | interned at @kotoba_tech
Joined November 2021
Thank you @kotoba_tech and special thanks to @jungokasai and @noriyuki_kojima! Wonderful and rewarding experience in Tokyo for the summer, surrounded by such a passionate team of talented engineers. Always excited about Kotoba's next release and look forward to keeping in touch!
Kotoba's former intern, Alan Li (@alanli2020), is starting his CS PhD at @Yale. Best of luck on your PhD journey, and we'll stay in touch!
1
2
6
RT @m_att_dunn: @ThereBeLyte @jenboland @pelaseyed We built it out of necessity, but Alan Li, et from @nlpnoah's group put together a nice…
0
1
0
@ThereBeLyte @m_att_dunn @jenboland @pelaseyed @nlpnoah Thanks for the interest! Combination doesnt matter cuz the model was trained and directly generate doc id rather than prompting model with any query-doc pair. the LLM retriever engine could return multiple relevant docs as well, e.g. by descending order of beam search scores
1
0
2
RT @zhang_muru: Running your model on multiple GPUs but often found the speed not satisfiable? We introduce Ladder-residual, a parallelism-…
0
52
0
@AnsongNi Any specific usecase? I actually unsubscribed perplexity as I found gpt search is quite decent
2
0
0
@Banana3Stocks As nlp researcher in the field I def agree the deepseek news is actually good for nvda and meta. Don’t know why people panic about it lol
34
3
37
@_apoorvnandan It’s good if it exists and readers like me do care lol but unfortunately I’m no expert in HPC and network etc
0
0
0
@_apoorvnandan For educational code that’s true. I’m also wondering is there a simulating environment for this kind of dev. Ig people in industry don’t care as they have abundant resources anyway. Articles should be fine tho
1
0
0
RT @AIatMeta: Today we’re announcing the recipients of our LLM Evaluation research grants. These four projects will each receive $200K in g…
0
62
0
🕯️ GLUE was an important benchmark in my first project, alongside all other nice work from Felix ... So sorry to hear this
0
0
1
RT @JeffDean: @moderncpp7 @clu_cheng @NeurIPSConf @drfeifei @jhyuxm @edchi I didn't see the talk, but the images I've seen of the slide see…
0
166
0
Happy thanksgiving! Thanks to all my mentors and collaborators, friends and family keep supporting me along the way. It’s been a journey and look forward to the next chapter!
0
0
2
RT @soldni: OLMo 2 is out 🥳 7B and 13B trained on 5T tokens, and meticulousy instruction tuned using Tulu 3 recipe. Simply the best fully…
0
29
0
RT @xwang_lk: I was in my mid 20s, I was a CS PhD student in AI, I wanted to build tech startups, I wanted to be a faculty, I wanted to bui…
0
23
0
RT @jacobcares: I'm so excited that we're finally releasing Tülu 3, our new post-training recipe! We're releasing models built on top of Ll…
0
18
0
RT @Tim_Dettmers: Just to clarify this benchmark. This is an apple to oranges comparison. - Cerebras is fast for batch size 1 but slow for…
0
84
0
RT @HanSineng: Super excited about the upcoming talk by @May_F1_ on responsible and robust AI systems! Come join us at Kline Tower next Tu…
0
1
0
RT @jungokasai: Just removed the waitlist for the playground with a turbo version release! Enjoy the fast and naturally sounding speech gen…
0
2
0
RT @armancohan: And a few papers in finding/demo/industry tracks. Come find us! A lot of us are here including @AnsongNi @HanSineng @Yixin…
0
3
0