Experiences
21'-Now : AI Engineer @
24'.5-8 : Research Intern @
23'.5-8 : Research Intern @
22'.5-8 : Research Intern @

Dong-Ho Lee (이동호)

PhD Candidate @ (dongho.lee[at]usc.edu)

I am a PhD candidate in Computer Science at USC and USC/ISI, advised by Prof. Jay Pujara, with active collaboration with Prof. Xiang Ren. During my PhD, I interned at & (2024 with Adam Kraft, Long Jin, and Xinyang Yi), (2023 with Francesco Barbieri), and (2022 with Sujay Jauhar); I am also a founding AI engineer at , closely working with Sungjoon Park and Jihyung Moon. Prior to USC, I conducted ML research at Upstage AI and Sungkyunkwan University.

I have served as an area chair and reviewer for leading conferences in NLP (ARR - Area Chair, ACL, EMNLP, NAACL, EACL, COLM, LREC), ML (ICML, ICLR, NeurIPS), and IR (KDD, WWW, SDM).

I work on contextual reasoning to improve LLMs as effective reasoners and social intelligence to enable LLMs to create meaningful societal impact. In specific,

Contextual reasoning: Contextual cues are important in recent LM research, enabling models to reason effectively, and adapt to complex tasks. My research improves LM performance across diverse tasks by integrating a variety of contextual cues, including: (a) human-provided explanations [TriggerNER (ACL 2020), AutoTriggER (EACL 2023 & TrustNLP 2021 Best Paper), LEAN-LIFE (ACL 2020 Demo), XMD (ACL 2023 Demo)]; (b) in-context examples [FewNER (ACL 2022), LLM-Data-Creation (EMNLP 2023), TKG-LLM (EMNLP 2023)]; and (c) dialogue context [NormVio-RT (EMNLP 2023)].

Social Intelligence System: A goal-oriented dialogue system must engage in meaningful conversations to achieve a desired outcome, such as persuading a user to make a purchase or facilitating better learning. My research focuses on (a) decision support systems to assist users in making informed decisions through tailored recommendations [STAR (Google Deepmind, 2024)]; (b) creating realistic user simulations to evaluate dialogue systems in diverse scenarios [LoCoMo (ACL 2023)], and (c) deploying dialogue systems in real business to make societal impact [Our product (MRR $10K) at ].

I actively engage in cross-disciplinary collaboration to apply machine learning techniques to various domains, including 6G networks [On-device Semantic Communication (IEEE TWC 2024), Seq2Seq-SC (IEEE Asilomar 2023)], medical science [Suicide content detection], etc.

News

  • [2024-02-15] I passed my qualifying exam and officially became a PhD Candidate.
  • [2023-12-07] I will serve as an Area Chair at NAACL 2024 (ARR December 2023).
  • [2023-10-30] I will serve as an Area Chair at EACL 2024 (ARR October 2023).
  • [2023-10-07] Three first-authored papers (Paper 1, Paper 2, Paper 3) have been accepted to EMNLP 2023!
  • [2023-05-08] XMD, Explanation-based model debugging framework, has been accepted to ACL 2023 Demo!
  • [2023-01-21] One first-authored paper (AutoTriggER) has been accepted to EACL 2023!
  • [2022-08-24] Invited talks on Explanation based Learning at POSTECH.
  • [2022-02-24] Two first-authored papers (Paper 1, Paper 2) have been accepted to ACL 2022!

Talks

  • Explanation-based Learning [link], 2022, Invited Talks @ POSTECH
  • Explanation-based Learning [link], 2022, Invited Talks @ Microsoft