About Me
Hello! I am currently serving as a Research Engineer at the FAIR team within Meta AI. Since joining the team in 2018, I have had the opportunity to delve into a broad spectrum of AI topics. My work has spanned Reinforcement Learning (RL), Generative Conversational AI (which has now evolved into LLM), Safety and Responsible AI, and Core Machine Learning research.
Before joining Meta AI, I earned a Master’s degree specializing in Machine Learning from Telecom ParisTech and University Paris 6. My academic journey has always been driven by a keen interest in both engineering and AI research. I am excited to continue exploring these fields and contributing to the advancements in AI technology.
I publish under the name Da JU.
Research Interest
NLP, GenAI, RAI
Academia
In 2015, I finished my undergraduate in Information Science and Engineering at Southeast University, Nanjing, China. I wrote my Bachelor’s thesis, “High-resolution LED Headlamps Testing System with Web GUI” at L-LAB, Lippstadt, Germany. It documents the development of WiFi based system for controlling cars. The thesis remains Intellectual property of HELLA.
In 2017, I finished my Master’s degree at Telecom ParisTech, Paris, France, focussing on Machine Learning and Humain Computer Interaction. I also recived a second master degree from UPMC university, Paris, France in Data Science.
Publications
2023
Jing Xu, Da Ju, Joshua Lane, Mojtaba Komeili, Eric Michael Smith, Megan Ung, Morteza Behrooz, William Ngan, Rashel Moritz, Sainbayar Sukhbaatar, Y-Lan Boureau, Jason Weston, Kurt Shuster. 2023. Improving Open Language Models by Learning from Organic Interactions.
2022
D Ju, S Roller, S Sukhbaatar, J Weston. 2021. Staircase Attention for Recurrent Processing of Sequences. Selected as Oral. NeurIPS 2022
D Ju, J Xu, YL Boureau, J Weston. 2022. Learning from data in the mixed adversarial non-adversarial case: Finding the helpers and ignoring the trolls.
K Shuster, J Xu, M Komeili, D Ju (Joint First Author), EM Smith, S Roller, M Ung, M Chen, … 2022. Blenderbot 3: a deployed conversational agent that continually learns to responsibly engage.Facebook AI Blog Post
2021
N Goyal, C Gao, V Chaudhary, PJ Chen, G Wenzek, D Ju, S Krishnan, … 2021. The FLORES-101 Evaluation Benchmark for Low-Resource and Multilingual Machine Translation. WMT21@EMNLP 2021
J Xu, D Ju, M Li, YL Boureau, J Weston, E Dinan. 2021. Bot-adversarial dialogue for safe conversational agents. NACL 2021
S Sukhbaatar, D Ju, S Poff, S Roller, A Szlam, J Weston, A Fan. 2021. Not All Memories are Created Equal: Learning to Forget by Expiring. ICML 2021
2020
Jing Xu, Da Ju, Margaret Li, Y.-Lan Boureau, Jason Weston, and Emily Dinan. 2020. Recipes for Safety in Open-domain Chatbots. arXiv:2010.07079 [cs], October. arXiv: 2010.07079.
Kurt Shuster, Eric Michael Smith, Da Ju, and Jason Weston. 2020. Multi-Modal Open-Domain Dialogue. arXiv:2010.01082 [cs], October. arXiv: 2010.01082. EMNLP 2021
Stephen Roller, Emily Dinan, Naman Goyal, Da Ju, Mary Williamson, Yinhan Liu, Jing Xu, Myle Ott, Kurt Shuster, Eric M. Smith, Y.-Lan Boureau, and Jason Weston. 2020. Recipes for building an open-domain chatbot. arXiv:2004.13637 [cs], April. arXiv: 2004.13637. Facebook AI Blog Post EACL 2021
2019
Da Ju, Kurt Shuster, Y.-Lan Boureau, and Jason Weston. 2020. All-in-One Image-Grounded Conversational Agents. arXiv:1912.12394 [cs], January. arXiv: 1912.12394.
Kurt Shuster, Da Ju, Stephen Roller, Emily Dinan, Y.-Lan Boureau, and Jason Weston. 2019. The Dialogue Dodecathlon: Open-Domain Knowledge and Image Grounded Conversational Agents. arXiv:1911.03768 [cs], November. arXiv: 1911.03768. ACL 2020
2018
Jonas Gehring, Da Ju, Vegard Mella, Daniel Gant, Nicolas Usunier, and Gabriel Synnaeve. 2018. High-Level Strategy Selection under Partial Observability in StarCraft: Brood War. arXiv:1811.08568 [cs, stat], November. arXiv: 1811.08568 (RLPO@NeurIPS 2018).