Baixi Sun

Baixi Sun

he/his

Baixi Sun is a Ph.D. Candidate at Luddy School of Informatics, Computing, and Engineering.

About Me

I am a Ph.D. candidate at the Luddy School of Informatics, Computing, and Engineering at Indiana University, Bloomington. I obtained a master’s degree (MS) in Computer Science at the University of California, Riverside (UCR). My research interests include machine learning systems, data compression, high-performance computing and cloud computing.

News

I am excited to start my internship at ByteDance in San Jose this May!
COMPSO paper got accepted to PPoPP 2025!

Selected Publications

  1. B. Sun, W. Liu, J. Pauloski, J. Tian, J. Jia, D. Wang, B. Zhang, M. Zheng, S. Di, S. Jin, Z. Zhang, X. Yu, K. Iskra, P. Beckman, G. Tan, and D. Tao, “COMPSO: Optimizing Gradient Compression for Distributed Training with Second-Order Optimizers,” accepted to ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP) 2025.
  2. (Co-1st author) S. Huang, B. Sun, L. Fan, X. Chen, J. Tian, Q. Shen, H.-Y. Wan, D. Tao, and E. Bao, “Ufn: User-friendly navigation framework based on systematic utilization of user-friendliness features from historical traffic data,” Journal of Computer Science and Technology, 2024. [Online]. Available: https://jcst.ict.ac.cn/en/article/doi/10.1007/s11390-024-4020-9.
  3. B. Sun, X. Yu, C. Zhang, J. Tian, S. Jin, K. Iskra, T. Zhou, T. Bicer, P. Beckman, and D. Tao, “Solar: A highly optimized data loading framework for distributed training of cnn-based scientific surrogates,” 2022. [Online]. Available: https://arxiv.org/abs/2211.00224.
  4. B. Sun, X. Yu, K. Iskra, D. Tao. 2022. SurrogateTrain: Drastically Improving Performance of Data Loading for Training Scientific Surrogate Models. The International Conference for High-Performance Computing, Networking, Storage, and Analysis (SC’22) ACM Student Research Competition Poster. https://sc22.supercomputing.org/proceedings/src_poster/src_poster_pages/spostg106.html.
  5. B. Sun, X. Yu, D. Tao. 2024. KVSort: Drastically Improving LLM Inference Performance via KV Cache Compression. Accepted The International Conference for High-Performance Computing, Networking, Storage, and Analysis (SC) 2024 ACM Student Research Competition Poster.

Recent Awards

2025 NSF PPoPP Travel Grant
2024 SIGHPC SC Student Travel Grants (1st)
2024 SIGHPC SC Student Travel Grants (2nd)
2022 SIGHPC Student Research Competition Travel Grants

Teaching

Fall 2024 Associate Instructor, ISE (Indiana University Bloomington)
ENGR_E 516: Engineering Cloud Computing
Spring 2021 Teaching assistant, EECS (Washington State University)
CPT_S 460: Operating Systems and Computer Architecture\

Full Publications

  1. B. Sun, W. Liu, J. Pauloski, J. Tian, J. Jia, D. Wang, B. Zhang, M. Zheng, S. Di, S. Jin, Z. Zhang, X. Yu, K. Iskra, P. Beckman, G. Tan, and D. Tao, “COMPSO: Optimizing Gradient Compression for Distributed Training with Second-Order Optimizers,” accepted to ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP) 2025.
  2. J. Jia, C. Xie, H. Lu, D. Wang, H. Feng, C. Zhang, B. Sun, H. Lin, Z. Zhang, X. Liu, and D. Tao, “SDP4Bit: Toward 4-bit Communication Quantization in Sharded Data Parallelism for LLM Training,” accepted by the Thirty-eighth Annual Conference on Neural Information Processing Systems (NeurIPS) 2024.
  3. (Co-1st author) S. Huang, B. Sun, L. Fan, X. Chen, J. Tian, Q. Shen, H.-Y. Wan, D. Tao, and E. Bao, “Ufn: User-friendly navigation framework based on systematic utilization of user-friendliness features from historical traffic data,” Journal of Computer Science and Technology, 2024. Available: https://jcst.ict.ac.cn/en/article/doi/10.1007/s11390-024-4020-9.
  4. X. Wei, F. Ye, O. Yonay, X. Chen, B. Sun, D. Tao, and T. Yang, “Fastclip: A suite of optimization techniques to accelerate clip training with limited resources,” submitted to International Conference on Learning Representations (ICLR) 2025. Available: https://arxiv.org/abs/2407.01445.
  5. B. Sun, X. Yu, C. Zhang, J. Tian, S. Jin, K. Iskra, T. Zhou, T. Bicer, P. Beckman, and D. Tao, “Solar: A highly optimized data loading framework for distributed training of cnn-based scientific surrogates,” 2022. Available: https://arxiv.org/abs/2211.00224.
  6. B. Sun, X. Yu, K. Iskra, D. Tao. 2022. SurrogateTrain: Drastically Improving Performance of Data Loading for Training Scientific Surrogate Models. The International Conference for High-Performance Computing, Networking, Storage, and Analysis (SC’22) ACM Student Research Competition Poster.
  7. B. Sun, X. Yu, D. Tao. 2024. KVSort: Drastically Improving LLM Inference Performance via KV Cache Compression. Accepted The International Conference for High-Performance Computing, Networking, Storage, and Analysis (SC) 2024 ACM Student Research Competition Poster.
  8. Q. Shen, S. Huang, B. Sun, X. Chen, D. Tao, H. Wan, and E. Bao, “Pvii: A pedestrian-vehicle interactive and iterative prediction framework for pedestrian’s trajectory,” Applied Intelligence, pp. 1–11, 2024.
  9. C. Zhang, S. Smith, B. Sun, J. Tian, J. Soifer, X. Yu, S. L. Song, Y. He, and D. Tao, “Heat: A highly efficient and affordable training system for collaborative filtering based recommendation on CPUs,” in Proceedings of the 37th ACM International Conference on Supercomputing, ser. ICS ’23. New York, NY, USA: Association for Computing Machinery, 2023, p.324–335.
  10. C. Zhang, B. Sun, X. Yu, Z. Xie, W. Zheng, K. A. Iskra, P. Beckman, and D. Tao, “Benchmarking and in-depth performance study of large language models on habana gaudi processors,” in Proceedings of the SC’23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis, ser. SC-W ’23. New York, NY, USA: Association for Computing Machinery, 2023, p. 1759–1766.
  11. A. V. Babu, H. Chan, M. J. Cherukara, J. M. Monsalve Diaz, J. Doerfert, J. Feinstein, I. Foster, R. J. Harder, K. Hickey, J. H¨uckelheim, S. Kandel, R. Kettimuthu, T. Kumar, Z. Liu, M. MacDonell, A. Miceli, M. Ngom, P. Pal, N. Paulson, K. Picel, K. Raghavan, A. Ramanathan, E. Rangel, S. Raskar, V. Sastry, G. Sivaraman, B. Sun, M. Trovato, L. Valentino, Z. Xie, E. Yan, Y. Yao, K. Yoshii, X. Yu, and T. Zhou, “2022 ai testbed expeditions report,” Argonne National Laboratory (ANL), Argonne, IL (United States), Tech. Rep., 2022.

Hobbies

Table Tennis, Basketball, Swimming, Go, Acoustic Guitar, Video games.