Liliang Ren

prof_pic.jpeg

Hi! I am a Senior Researcher at Microsoft GenAI, and a graduating PhD candidate in Computer Science at University of Illinois Urbana-Champaign. I am fortunate to be advised by Prof. Chengxiang Zhai, and work with Prof. Heng Ji and Prof. Julia Hockenmaier. I finished my master degree at University of California San Diego advised by Prof. Julian McAuley, and my undergraduate degree at Shanghai Jiao Tong University working with Prof. Kai Yu.

My long-term research goal is to develop artifical neural architectures with superhuman efficiency and interpretability, and apply them to discover new structured knowledge from natural signals. Specifically, I am interested in the following areas:

  • Efficient and Interpretable Sequence Modeling: dynamic and modular sparsity for long sequence modeling, language modeling with latent type learning.
  • Structured Text Generation: text-to-graph extraction, dialogue state tracking and generation.

For a complete list of publications, please check here.

News

Jun 12, 2024 Excited to announce our Samba-3.8B paper! It outperforms Phi-3-mini on major benchmarks by a large margin, while having infinite context length with linear complexity!🚀 Checkout my twitter for a brief introduction!
Sep 21, 2023 One paper on Efficient Sequence Modeling got accepted by NeurIPS 2023!
Jul 31, 2023 Gave a talk on my recent work of Sparse Modular Activation at Microsoft Research!
Jul 18, 2023 Proud to announce the paper and the code of my previous internship work on Efficient Sequence Modeling at Microsoft Research!
May 23, 2023 One paper on Dialogue Evaluation is accepted for oral presentation at the DialDoc workshop at ACL 2023!
May 15, 2023 Started my summer research internship at Amazon Science, working on a confidential project of Large Language Model!
Oct 6, 2022 One paper on Language Model Pre-training got accepted as Oral presentation by EMNLP 2022!
May 15, 2022 Started my summer Research Intern at the Knowledge and Language Team of Microsoft Research! Working with Yang Liu, Shuohang Wang and Yichong Xu.

Selected Publications

  1. NeurIPS
    Sparse Modular Activation for Efficient Sequence Modeling
    Liliang Ren, Yang Liu, Shuohang Wang, Yichong Xu, Chenguang Zhu, and ChengXiang Zhai
    Proceedings of the Thirty-seventh Conference on Neural Information Processing Systems, Dec 2023
  2. ACL DialDoc
    C-PMI: Conditional Pointwise Mutual Information for Turn-level Dialogue Evaluation
    Liliang Ren, Mankeerat Sidhu, Qi Zeng, Revanth Gangi Reddy, Heng Ji, and ChengXiang Zhai
    Proceedings of the ACL2023 Workshop on Document-grounded Dialogue and Conversational Question Answering, Jul 2023
  3. EMNLP
    Oral
    Language Model Pre-Training with Sparse Latent Typing
    Liliang Ren*, Zixuan Zhang*, Han Wang, Clare Voss, Chengxiang Zhai, and Heng Ji
    Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, Dec 2022
  4. ACL
    HySPA: Hybrid Span Generation for Scalable Text-to-Graph Extraction
    Liliang Ren, Chenkai Sun, Heng Ji, and Julia Hockenmaier
    Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, Aug 2021
  5. EMNLP
    Scalable and Accurate Dialogue State Tracking via Hierarchical Sequence Generation
    Liliang Ren, Jianmo Ni, and Julian McAuley
    Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), Nov 2019
  6. EMNLP
    Oral
    Towards Universal Dialogue State Tracking
    Liliang Ren, Kaige Xie, Lu Chen, and Kai Yu
    Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, Oct 2018