Dongyeop Kang (DK)
@dongyeopkang
Followers
965
Following
1K
Statuses
328
Assistant professor @UMNComputerSci, a member of @MinnesotaNLP group, research on human-centered language technologies, he/him/his
Minneapolis, MN
Joined April 2018
RT @vishakh_pk: If you work on writing with LLMs, consider submitting to the In2Writing workshop at #NAACL2025 that alternates between *CL…
0
8
0
RT @windx0303: The always-amazing #In2Writng workshop will be at #NAACL2025 @naaclmeeting this time! #NLProc #HCI Paper deadline: Feb 28,…
0
22
0
Very excited to host @nlpnoah today in our Cray Distinguished Speaker series, and hear about his team's work (@allen_ai) on Open-sourced LLMs for Science!
Colloquium continues Monday with @nlpnoah from @uwcse as part of our Cray Distinguished Speaker Series. The talk is titled "OLMo, Tülu, and Friends: Accelerating the Science of Language Modeling" and will be held in Keller Hall 3-180. Learn more here:
1
1
16
RT @UMNComputerSci: Colloquium continues Monday with @nlpnoah from @uwcse as part of our Cray Distinguished Speaker Series. The talk is tit…
0
2
0
RT @im_kooryan: How can we produce generations that are well-balanced in irony, formality, and positivity? Or between negativity and formal…
0
2
0
(2) How can we control multiple (possibly conflicting) styles or rewards in RLHF? We developed a simple but effect dynamic multi-weighting scheme.
If you have any questions, visit Karin at the In-Person Session 4 (4-5:30 PM) on Tuesday! paper: code:
0
0
0
0
0
1
1
0
3
RT @pluralistic_ai: 🚨 Travel Support Alert! 💰Our workshop offers a limited number of grants to help cover NeurIPS registration and accommod…
0
7
0
0
0
2
Check out the paper, if you are interested in! "Shallow Synthesis of Knowledge in GPT-Generated Texts: A Case Study in Automatic Related Work Composition"
0
0
0
RT @shannonzshen: @SakanaAILabs For all these AI generated papers, they have very few (<10), "old" (before 2023), and only-"famous" citatio…
0
10
0
[5/5] Identifying and improving multi-modal AND multi-hop reasoning in VQA. @AmazonScience Internship project led by @Jihyung_Kil and managed by @supakjk
0
1
3
[4/5] Benchmarking cognitive biases when using LLMs as evaluators (led by @im_koogusry and collaboration with @Grammarly @_vipulraheja )
LLMs have proven to outperform humans on a multitude of tasks. Does this also mean they are more biased too? In our work, we benchmark several different LLMs as automatic evaluators for various cognitive biases.
0
1
0
[3/5] machine-text detection using discourse patterns in long-text (led by @zaemyung and collaboration with @Grammarly @_vipulraheja )
"Paraphrasing attacks" can compromise the effectiveness of AI content detectors. 🙀 Can hierarchical structures in texts help build a more robust detector? Our research reveals a resounding💡YES!💡Delighted to share our work on merging discourse frameworks with graph analysis.
0
1
1