David Jurgens is now on BlueSky only
@david__jurgens
Followers
2K
Following
938
Media
17
Statuses
371
Associate Professor at @UMSI and @UMichCSE working in computational social science and NLP. PI of the Blablablab https://t.co/pt1UFJu3tm
Joined December 2010
RT @CanyuChen3: Excited to share our๐ฅNew #NeurIPS2024 Paper๐ฅ: ๐๐๐ง ๐๐๐ซ๐ ๐ ๐๐๐ง๐ ๐ฎ๐๐ ๐ ๐๐จ๐๐๐ฅ ๐๐ ๐๐ง๐ญ๐ฌ ๐๐ข๐ฆ๐ฎ๐ฅ๐๐ญ๐ ๐๐ฎ๐ฆ๐๐ง ๐๐ซ๐ฎ๐ฌ๐ญ ๐๐๐ก๐๐ฏ๐ข๐จ๐ซ? The project weโฆ.
0
30
0
RT @StellaLisy: Had so much fun presenting our work ValueScope ( at #EMNLP2024 and (mostly) enjoying Miami with friโฆ.
0
8
0
This is just the first release of SPoRC so if you're eager for more data, stay tuned. We'll be keeping the podcast data coming!.
0
0
4
Lots of folks listen to podcastsโan estimated half of Americans do monthly!โbut there's been limited data to study what happens on them. What do they talk about? Who are the guests? Who is influencing whom?? This new dataset is our attempt to bring light to an important medium.
Podcasts are a popular medium, but data for computational research is limited! We introduce the Structured Podcast Research Corpus (SPoRC - , a large, multimodal dataset of English podcasts ๐งต.
2
5
33
RT @MiriamSchirmer: Check out my poster on Trauma Event Detection tomorrow at 2pm (Jasmine) #EMNLP2024. /w Tobias Leemann @david__jurgens @โฆ.
aclanthology.org
Miriam Schirmer, Tobias Leemann, Gjergji Kasneci, Jรผrgen Pfeffer, David Jurgens. Findings of the Association for Computational Linguistics: EMNLP 2024. 2024.
0
4
0
RT @UMichCSE: ๐จ CSE is hiring faculty! ๐จ .Join a top 10 program where you can shape the future through the power of computing. Applicants fโฆ.
0
17
0
RT @WikiResearch: "A Test of Time: Predicting the Sustainable Success of Online Collaboration in Wikipedia" measures the ability of collaboโฆ.
0
10
0
This was a really fun collaboration and I'm super proud of @leczhang for all his amazing work on SPRIG. He's applying for PhD programs this fall so if you're looking for a great student, keep an eye out!.
[1/12] Optimizing prompts for specific tasks has been key to improving LLM performance, but what if we optimize prompts on system level to work well on *all* tasks? Check out ๐ฑSPRIG, a genetic system prompt optimizer that help unlock LLMs' full potential:
1
1
12
RT @leczhang: [1/12] Optimizing prompts for specific tasks has been key to improving LLM performance, but what if we optimize prompts on syโฆ.
0
9
0
RT @elisazmq_zheng: ๐๏ธ What if the way we prompt LLMs might actually hold it back?.๐จ Assigning personas like "helpful assistant" in systemโฆ.
0
94
0
RT @AndersGiovanni: ๐ฉ๐ผโ๐ป Real or Robotic? ๐ค Can LLMs accurately simulate qualities of human responses in dialogue?. Human conversations witโฆ.
0
11
0
RT @Prolific: How can we improve human and AI alignment? ๐. @UMichโs Dr @david__jurgens and Dr @huashen218 will discuss how we can get AI tโฆ.
0
1
0
RT @customnlp4u: The CFP for our workshop on Customizable NLP at #EMNLP2024 is now live with a deadline of 8/15: Wโฆ.
0
5
0
RT @armancohan: ๐ข My lab at Yale is looking for a postdoc to work on LLMs/NLP for health. This is a unique collaboration with a health staโฆ.
0
78
0
RT @MishaTeplitskiy: Let's talk about about REJECTION in science and. geography๐ค. New paper @_Hong_Chen @ProfChrisRider @david__jurgens dโฆ.
0
20
0
RT @vinodkpg: Super thrilled to have this paper out at the journal Social Networks, where we looked at linguistic cues that signal "formingโฆ.
0
1
0
RT @NaitianZhou: Hey NLPals, I'll be at #NAACL2024 this upcoming week! Let's chat about sociocultural NLP, what it means to study culture,โฆ.
0
3
0
RT @huashen218: ๐ขIs current โhuman-AI alignmentโ research clarified and comprehensive? ๐ค We systematically reviewed 400+ papers across HCI,โฆ.
0
67
0
This project was a ton of fun with a great group of folks, aimed at getting a better measurement of whether LLMs actually have human-like personalities (spoiler: no) and turned into a really interesting modeling question. @leczhang's thread has a nice summary of the highlights.
[1/13] LLMs are increasingly skilled at mimicking human agents in social settings, but have they truly developed a consistent personality? Check out our work accepted to #NAACL2024 where we question the reliability of persona tests applied to LLMs. Arxiv:
0
3
31