Sirui Chen·sirui.dev

Research

Research projects

Work I've done in research labs, written for engineers rather than reviewers.

University of Pennsylvania · Prof. Chris Callison-Burch · May 2025Aug 2025

Retrieval-Augmented Generation for DARPA SciFy and OpenScholar

Designed and evaluated retrieval-augmented generation workflows for scientific question answering on DARPA SciFy and OpenScholar, with a focus on retriever ranking quality and reproducible evaluation.

Contributions

  • Implemented score-based filtering on top of dense retrievers to reduce irrelevant context passed to the generator.
  • Prepared and standardized domain datasets so retriever swaps and generator swaps could be evaluated on the same evidence pool.
  • Wrote reusable training and evaluation scripts and lab-facing documentation so other researchers could reproduce results.

Methods

  • Contrastive retriever fine-tuning
  • Score-based passage filtering
  • RAG evaluation harness
  • Domain dataset curation
PythonPyTorchHugging FaceRetrievalRAGNLP

Washington University in St. Louis · Robotics Lab · Aug 2023May 2024

Vision-Assisted Self-Driving F1Tenth Car

Improved perception robustness for a 1/10-scale autonomous racing car under glare and shadow, and integrated new sensor input through ROS.

Contributions

  • Designed targeted data augmentation to stabilize perception under glare and shadow.
  • Applied self-attention distillation, improving model convergence speed by 20%.
  • Built a ROS accelerometer driver to add a new sensor modality to the perception stack.

Methods

  • Data augmentation
  • Self-attention distillation
  • ROS driver development
  • Experiment tracking
PythonPyTorchROSC++Computer Vision