About Me
I am a PhD student in Computer Science at the Baskin School of Engineering, UC Santa Cruz, advised by Prof. Ian Lane. My research focuses on natural language processing with an emphasis on building reliable and trustworthy language processing systems by detecting and mitigating hallucinations, improving generation faithfulness, and understanding the underlying causes of errors.
Outside of research, I enjoy playing the piano, skiing, ice skating, and climbing.
News
- 10-16-2025: I will give an oral presentation on “Context, Models and Prompt Optimization for Automated Hallucination Detection in LLM Output.” at BayLearn 2025.
- 08-01-2025: Our paper “Context, Models and Prompt Optimization for Automated Hallucination Detection in LLM Output.” won a best system paper award at SemEval-2025.
- 08-01-2025: Presented our paper at SemEval-2025 in Vienna, Austria.
Education
-
PhD in Computer Science, 2023 - present
University of California, Santa Cruz (Silicon Valley Campus)
Advised by Prof. Ian Lane -
MS in Computational Linguistics, 2019 - 2021
University of Washington, Seattle -
Exchange research student in Information and Communication Engineering, 2017 - 2018
Tokyo Institute of Technology, Tokyo
Advised by Prof. Manabu Okumura -
BS in Electrical and Computer Engineering, 2015 - 2019
University of Washington, Seattle
Experience
-
AI Resident, Meta · Menlo Park, CA · 09/2021 – 11/2022
Researched faithfulness metrics for dialogue summarization by improving BARTScore and proposing T0-Score, a new metric that better correlates with human judgments across domains. In addition, I contributed to Meta’s open-source PyTorch ecosystem (TorchRL, torchtnt). -
NLP Engineer Intern, Seasalt AI · Seattle, WA · 01/2021 – 05/2021
Designed, trained, and deployed a meeting summarization service for the SeaMeet product line. I also developed an automated speech-collection and annotation infrastructure. -
Student Researcher, University of Washington · Seattle, WA · 2019 – 2021
Engineered and selected biomedical knowledge features for NLI (ACL-BioNLP’19 Shared Task) and integrated them into a transformer model, ranking 1st with 99% test accuracy. In additionq, I helped create a medical dialogue summarization dataset from Chunyuyisheng and benchmarked SOTA summarizers.
Publications
-
UCSC at SemEval-2025 Task 3: Context, Models and Prompt Optimization for Automated Hallucination Detection in LLM Output
Sicong Huang, Jincheng He, Shiyuan Huang, Karthik Raja Anandan, Arkajyoti Chakraborty, Ian Lane
SemEval, 2025
💻 Code · 🔬 Project Page · 🏆 Best paper award -
Enhancing Faithfulness in Abstractive Summarization via Span-Level Fine-Tuning
Sicong Huang, Qianqi Yan, Shengze Wang, Ian Lane
arXiv preprint, 2025 -
Toward Faithful Dialogs: Evaluating and Improving the Faithfulness of Dialog Systems
Sicong Huang
Young Researchers’ Roundtable on Spoken Dialogue Systems, 2024 -
ED-FAITH: Evaluating dialogue summarization on faithfulness
Sicong Huang, Asli Celikyilmaz, Haoran Li
arXiv preprint, 2022 -
WTMED at MEDIQA 2019: A Hybrid Approach to Biomedical Natural Language Inference
Zhaofeng Wu, Yan Song, Sicong Huang, Yuanhe Tian, Fei Xia
BioNLP, 2019
Teaching
- TA for NLP 243 Deep Learning for NLP (Fall 2025)
- TA for NLP 243 Deep Learning for NLP (Fall 2024)
Last updated: October 2025