Liu L. Sijia

Email  /  Google Scholar  /  Twitter  /  Github

prof_pic.jpg

sijia.liu@{myschool}.edu

8 Lawrence Dr

Princeton, NJ 08540

PhD Student of Computer Science at Princeton University

Research Areas: ML / NLP / LLM

Hello! I am a 1st year CS PhD at Princeton University where I am fortunate to be advised by Prof. Karthik Narasimhan at Princeton Language and Intelligence. I am fascinated by the interactive nature of language, and am actively working on topics including AI alignment, reasoning and agentic behavior especially in multi-agent systems. Previously, I worked as a research scientist in Amazon Foundation Model (Amazon AGI) Post-Training team where I contributed as a founding member to the development of Nova – the state-of-the-art text and multimodal models and Alexa Prize SocialBot with Dr. Yang Liu and Prof. Dilek Hakkani-Tur in Sunnyvale, CA. Before that, I received my Bachelors from Peking University and my Masters from Carnegie Mellon University.

Outside of work, I enjoy badminton, running, reading and exploring local hidden gems. Being a language lover at heart, I’ve been pre-trained with Chinese and fine-tuned on English, Korean, German and Japanese, but I might suffer from catastrophic forgetting sometimes. I enjoy photography and have also touched upon scriptwriting for a bit, and wrote a short film based on a real story during the pandemic time and won several film festival awards.


news

Dec 3, 2024 The Amazon Nova Family of Models have been released. :rocket::sparkles:
Oct 9, 2024 Our work on LLM Self-Correction with DeCRIM: Decompose, Critique, and Refine for Enhanced Following of Instructions with Multiple Constraints has been accepted by EMNLP 2024 and also NeurIPS System 2 Reasoning Workshop. See you in Miami! :tada:
Sep 3, 2024 I started my PhD journey at Princeton University.
Oct 7, 2023 Our work on Aligning Dialogue Model Behavior with Guidelines has been accepted by EMNLP 2023.
Sep 12, 2023 Our proceedings on the Fifth Alexa Prize SocialBot Grand Challenge are out :rocket::sparkles:

selected publications

2024

  1. Model Card
    The Amazon Nova family of models: Technical report and model card
    Amazon Artificial General Intelligence
    Amazon Technical Reports, 2024
  2. EMNLP
    LLM Self-Correction with DeCRIM: Decompose, Critique, and Refine for Enhanced Following of Instructions with Multiple Constraints
    Thomas Palmeira Ferraz, Kartik Mehta, Yu-Hsiang Lin, and 7 more authors
    EMNLP, NeurIPS Workshop on System 2 Reasoning at Scale (Oral), 2024

2023

  1. AAAI
    Towards Credible Human Evaluation of Open-Domain Dialog Systems Using Interactive Setup
    Sijia Liu, Patrick Lange, Behnam Hedayatnia, and 5 more authors
    AAAI (Oral), NeurIPS Workshop on Human Evaluation of Generative Models (Oral), EMNLP Workshop on Natural Language Generation, Evaluation, and Metrics (Oral), Jun 2023

2022

  1. SigDial
    Improving Bot Response Contradiction Detection via Utterance Rewriting
    Di Jin, Sijia Liu, Yang Liu, and 1 more author
    SIGDIAL (Oral), Jul 2022