Yuchen Zhuang

Yuchen Zhuang

yczhuang [at] gatech (dot) edu

Twitter LinkedIn GitHub Google Scholar

Hello World! My name is Yuchen Zhuang. I am a final-year Machine Learning Ph.D. candidate in Georgia Institute of Technology, advised by Prof. Chao Zhang. I am also very fortunate to work closely with Prof. Le Song and Prof. Bo Dai. My principal interest lies in the language intelligence, aiming for developing language model-based agents capable of exhibiting human-like reasoning and planning when tackling challenging real-world problems. My experience spans various stages of LLM development, including model pre-training, instruction fine-tuning, RLHF, and evaluation. My recent research covers the following directions:

  • Language intelligence with human-level reasoning and planning capabilities: [ICLR'24, NeurIPS'23a, NeurIPS'23b, EMNLP'22];
  • Adapting or aligning language models to possess human-level capabilities on specific tasks: [ICML'24, NeurIPS'24];
  • Data-centric approaches that can offer high-quality data for effort-light model training and faithful evaluation: [NeurIPS'23c, KDD'23].

News

  • [--Pinned--] I am actively seeking industrial R&D opportunities, including both internship (Spring 2025, starting from Dec 2024) and full-time occupations (2025). I am open to topics and happy to engage in discussions regarding potential opportunities!
  • [09/2024] Two papers got accepted in NeurIPS'24, discussing LLM personalization with model factorization, and LLM alignment with representation editing. See you in Vancouver!
  • [09/2024] Three papers got accepted in EMNLP'24, introducing RAG, LLM Agent, and domain adaptation in medical applications.
  • [07/2024] Our paper got accepted in COLM'24, introducing principle discovery for LLM reasoning. See you in Philadelphia, PA!
  • [06/2024] Our ICML paper has been selected as Spotlight! Congratulations to all the collaborators.
  • [05/2024] One paper got accepted in ICML'24, introducing a novel method for black-box LLM adaptation. See you (virtually) in Vienna!
  • [03/2024] Will join Amazon Rufus Team as Applied Scientist Intern during Summer 2024. See you in Palo Alto!
  • [01/2024] One paper got accepted in ICLR'24, discussing efficient navigation of LLM agent in action space. Congratulations to my collaborators in Adobe Research! See you (maybe virtually) in Vienna!
  • [10/2023] Humbled to receive NeurIPS 2023 Scholar Award. See you in New Orleans!
  • [09/2023] Three papers got accepted in NeurIPS'23, discussing a closed-loop LLM-based autonomous agent, a tool-use dataset for LLMs, and attributed dataset generation with LLMs. See you in New Orleans!

Selected Publications and Manuscripts

Please refer to publications or my Google Scholar profile for the full list. ("*" stands for equal contribution)
HYDRA: Model Factorization Framework for Black-Box LLM Personalization
BBox-Adapter: Lightweight Adapting for Black-Box Large Language Models
ToolChain*: Efficient Action Space Navigation in Large Language Models with A* Search
AdaPlanner: Adaptive Planning from Feedback with Language Models
ToolQA: A Dataset for LLM Question Answering with External Tools
Large Language Model as Attributed Training Data Generator: A Tale of Diversity and Bias
DyGen: Learning from Noisy Labels via Dynamics-Enhanced Generative Modeling
Retrieval-Augmented Large Language Models for Adolescent Idiopathic Scoliosis Patients in Shared Decision-Making
End-to-end Stochastic Optimization with Energy-based Model
ReSel: N-ary Relation Extraction from Scientific Text and Tables by Learning to Retrieve and Select

Industrial Experience

Amazon (May 2024-Aug 2024)
Topic: Pre-Training Agent LLM to Enhance the Foundamental Agentic Capabilities
Adobe Research (May 2023-Aug 2023)
Topic: ToolChain*: Efficient Action Space Navigation in Large Language Models with A* Search [ICLR'24]
Amazon (May 2022-Aug 2022)
Topic: G-STO: Sequential Main Shopping Intention Detection via Graph-Regularized Stochastic Transformer [CIKM'23]

Academic Services

  • Reviewer for conferences: ARR (2023-), EMNLP (2022-), ICLR (2024), NeurIPS (2023), KDD (2021-), ACL (2023), AAAI (2023-), ICML (2021), AISTATS (2024), SDM (2024).
  • Reviewer for workshops: FMDM@NeurIPS (2023), DMLR@ICML (2023), SPIGM@ICML (2023).
  • Reviewer for Journals: IEEE Journal on Selected Topics in Signal Processing (JSTSP).
  • Teaching Assistant, Georgia Tech NLP Bootcamp, Fall 2023-2024.
  • Teaching Assistant, Georgia Tech Big Data Bootcamp, Fall 2021-2024.
  • Graduate Teaching Assistant, CSE8803 DLT: Deep Learning for Text Data, Fall 2021.
  • Graduate Teaching Assistant, CSE7641 A: Machine Learning, Fall 2020.

Selected Awards

  • [10/2023] NeurIPS 2023 Scholar Award;
  • [09/2023] Best Paper Award, ACM Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM BCB);
  • [08/2023] KDD 2023 Student Travel Grant;
  • [06/2020] Second Prize of Excellent Undergraduate Student Graduation Thesis in Jiangsu Province;
  • [06/2019] Most Influential Graduate Award Nomination (20/4000), Southeast University;
  • [12/2017] Excellent Paper Award, International Collaboration Symposium on Information, Production & Systems, Japan.