Rutgers University

Junlin Chen.

:++++++++++++++++++:::+++++++++++++++++++++++
:::::++++**##@@@@@#*+++++++:::++++++++++++++:
::::::+*@@@@@@@@@@@@@@*+:::.:::::::::::::::::
::::+*@@@@@@@#*+*##@@@@#+..::::::::::::::::::
:::+@@@@#+::::.  .:+*#@@#+:::::::::::::::::::
.::*@@#*:..         :#@@@#+::::::::::::::::::
..:#@@#*+:.          +#@@@*::::::::::::.:::::
...+@#####*+::***++. :#@@@*..................
   .#@@@@@#*.*@@@#*+.:*+::.   ..........
    *#####*: .*###*. .:.
    :#*:.:**:..::.. ...
    .##*+*##++..::...:..
     .#####**++++:..:+::++++++::::::.......
      .*###*++. ..:++::++++++++++++++++++++++
+++++:++###*+::::+++::::++++++++++++:::::+:+:
++++++**#######**+:...:::++++++++++++++++++++
****##@#######*+:.        .::+++*************
*############+:.               .:+***********
  .:*#####*+.                     .:+********
     .:::.                           +*******
                                      +******
                                       +*****
                                       .*****
                                        :**++
              .                 ..       :***
...           .                :::   .    +**
::.  .        .                .::   ..   .+*
++..:   .     .      .         .++:..:....:+#
*+.:.        .        .        :++::+***+****
+::.   .     .        .        .+:.     :****

Efficient language models & ML systems.

Papers

PHOENIX Resilient LLM Training with Hot-Swapping via Zero-Overhead Checkpoint arXiv

ALPS Adaptive Lineage-Aware Parallel Search for LLM-Driven Optimization

PR2 Predictive Routing Replay for MoE-Based LLM Reinforcement Learning arXiv

SCAPE Accurate and Efficient LLM Training with Extreme Sparse Communication arXiv

CUBE-GD Communication-Efficient Byzantine-Resilient Learning

CV

Education

Rutgers University–New Brunswick

B.S. in Computer Science & Mathematics

Expected May 2027

Research experience

IFM, MBZUAI

AI Researcher

Feb 2026 – Present

Rutgers University

Research Assistant

Oct 2025 – Present

INSPIRE Lab, Rutgers University

Research Intern

Dec 2024 – Oct 2025

Contact

chenjunlin110@gmail.com

GitHub LinkedIn

Junlin Chen.

PHOENIX Resilient LLM Training with Hot-Swapping via Zero-Overhead Checkpoint arXiv ↗

ALPS Adaptive Lineage-Aware Parallel Search for LLM-Driven Optimization

PR2 Predictive Routing Replay for MoE-Based LLM Reinforcement Learning arXiv ↗

SCAPE Accurate and Efficient LLM Training with Extreme Sparse Communication arXiv ↗

CUBE-GD Communication-Efficient Byzantine-Resilient Learning

Education

Rutgers University–New Brunswick

Research experience

IFM, MBZUAI

Rutgers University

INSPIRE Lab, Rutgers University

PHOENIX Resilient LLM Training with Hot-Swapping via Zero-Overhead Checkpoint arXiv

PR2 Predictive Routing Replay for MoE-Based LLM Reinforcement Learning arXiv

SCAPE Accurate and Efficient LLM Training with Extreme Sparse Communication arXiv