About me

I am currently a final-year Ph.D. candidate at the UCLA Computer Science Department. I have the privilege of working with Prof. Songwu Lu and the members of the WiNG. My research broadly covers machine learning and systems, with a current focus on LLM post-training [SibylSense][DRO][RLTHF] and ML for systems [DeepSpecs][CloudEval-YAML]. I have a background in LLM inference systems [preprint], network systems [NSDI ’26][ATC ’25][SIGCOMM ’22][NSDI ’21][TMC], and data mining algorithms [KDD ’20][SIGCOMM ’22 Workshop][SANER 2021][TNSE]. Prior to UCLA, I received my B.S. in Computer Science from Peking University, where I studied sketch-based algorithms under the supervision of Prof. Tong Yang.

What’s New

[Feb. 2026] Check out our new work SibylSense on adaptive rubric learning available at preprint!

[Jan. 2026] Check out our new version of DRO that incorporates rubric-gated constraints available at preprint!

[Dec. 2025] Our work AnyPro on anycast optimization is accepted to NSDI ’26. Congrats to the team, and see you in Renton!

[Nov. 2025] Check out our new work DeepSpecs on telecom QA available at preprint!

[Jul. 2025] Check out our new work ECO-LLM on edge-cloud LLM orchestration available at preprint!

[Jul. 2025] Excited to continue with Microsoft Copilot Tuning Research as a Researcher, starting this summer and through my PhD final year, advancing our research on LLM post-training. See you in Redmond!

[Jun. 2025] Check out our new work DRO on LLM reasoning for open-ended tasks available at preprint!

[May 2025] Our work RLTHF is accepted to ICML 2025. See you in Vancouver!

Publications

Preprints

SibylSense: Adaptive Rubric Learning via Memory Tuning and Adversarial Probing
Yifei Xu, Guilherme Potje, Shivam Shandilya, Tiancheng Yuan, Leonardo de Oliveira Nunes, Rakshanda Agarwal, Saeid Asgari, Adam Atkinson, Emre Kıcıman, Songwu Lu, Ranveer Chandra, Tusher Chakraborty
[arXiv]

DeepSpecs: Expert-Level Question Answering in 5G
Aman Ganapathy Manvattira*, Yifei Xu*, Ziyue Dang, Songwu Lu
[arXiv]

Direct Reasoning Optimization: Constrained RL with Token-Level Dense Reward and Rubric-Gated Constraints for Open-ended Tasks
Yifei Xu*, Tusher Chakraborty*, Srinagesh Sharma, Leonardo Nunes, Swati Sharma, Kate Drakos Demopulos, Emre Kıcıman, Songwu Lu, Ranveer Chandra
[arXiv]

Orchestration for Domain-specific Edge-Cloud Language Models
Prasoon Patidar, Alex Crown, Kevin Hsieh, Yifei Xu, Tusher Chakraborty, Ranveer Chandra, Yuvraj Agarwal
[arXiv]

Conference Papers

AnyPro: Preference-Preserving Anycast Optimization based on Strategic AS-Path Prepending
Minyuan Zhou*, Yuning Chen*, Jiaqi Zheng, Yifei Xu, Guihai Chen, Wanchun Dou, Pan Hu, Yongping Tang, Wendong Yin, Jie Lin, Qingyan Yu, Yuanchao Su, Songwu Lu, Wan Du
NSDI ’26 [paper]

RLTHF: Targeted Human Feedback for LLM Alignment
Yifei Xu, Tusher Chakraborty, Emre Kıcıman, Bibek Aryal, Eduardo Rodrigues, Srinagesh Sharma, Roberto Estevao, Maria Angels de Luis Balaguer, Jessica Wolk, Rafael Padilha, Leonardo Nunes, Shobana Balakrishnan, Songwu Lu, Ranveer Chandra
ICML 2025 [arXiv]

Roaming Free in the VR World with MP2
Yifei Xu, Xumiao Zhang, Yuning Chen, Pan Hu, Xuan Zeng, Zhilong Zheng, Xianshang Lin, Yanmei Liu, Songwu Lu, Z. Morley Mao, Wan Du, Dennis Cai, Ennan Zhai, Yunfei Ma
ATC ’25 [paper]

CloudEval-YAML: A Practical Benchmark for Cloud Configuration Generation
Yifei Xu*, Yuning Chen*, Xumiao Zhang*, Xianshang Lin, Pan Hu, Yunfei Ma, Songwu Lu, Wan Du, Z. Morley Mao, Ennan Zhai, Dennis Cai
MLSys 2024 [pdf] [code]

SEED: a SIM-based solution to 5G failures
Jinghao Zhao, Zhaowei Tan, Yifei Xu, Zhehui Zhang, Songwu Lu
SIGCOMM 2022 [pdf]

Device-Based LTE Latency Reduction at the Application Layer
Zhaowei Tan, Jinghao Zhao, Yuanjie Li, Yifei Xu, Songwu Lu
NSDI ’21 [pdf]

A Multi-Metric Ranking Approach for Library Migration Recommendations
Hao He, Yulin Xu, Yixiao Ma, Yifei Xu, Guangtai Liang, Minghui Zhou
IEEE SANER 2021 [pdf]

WavingSketch: An Unbiased and Generic Sketch for Finding Top-k Items in Data Streams
Jizhou Li*, Zikun Li*, Yifei Xu*, Shiqi Jiang, Tong Yang, Bin Cui, Yafei Dai and Gong Zhang
SIGKDD 2020 [pdf]

Journals

Unbiased Real-time Traffic Sketching
Yuhan Wu*, Shiqi Jiang*, Yifei Xu*, Siyuan Dong, Kaicheng Yang, Peiqing Chen, Tong Yang
IEEE TNSE [pdf]

LDRP: Device-Centric Latency Diagnostic and Reduction for Cellular Networks Without Root
Zhaowei Tan, Jinghao Zhao, Yuanjie Li, Yifei Xu, Yunqi Guo, Songwu Lu
IEEE TMC [pdf]

Workshops, Posters, and Demos

CloudEval-YAML: A Realistic and Scalable Benchmark for Cloud Configuration Generation
Yifei Xu*, Yuning Chen*, Xumiao Zhang*, Xianshang Lin, Pan Hu, Yunfei Ma, Songwu Lu, Wan Du, Z. Morley Mao, Ennan Zhai, Dennis Cai
NeurIPS 2023 Workshop on ML for Systems [pdf]

PISketch: finding persistent and infrequent flows
Zhuochen Fan, Zhoujing Hu, Yuhan Wu, Jiarui Guo, Wenrui Liu, Tong Yang, Hengrui Wang, Yifei Xu, Steve Uhlig, Yaofeng Tu
SIGCOMM 2022 Workshop on FFSPIN [pdf]

* co-first author

Teaching

TA in UCLA CS Dept since 2022, courses include:
CS 180 - Introduction to Algorithms and Complexity
CS 118 - Computer Network Fundamentals
CS 31 - Introduction to Computer Science I
CS 35L - Software Construction