Research

See my Google Scholar for most the up-to-date list.

Preprints

• Full-bandwidth Transformer
  Xi Wang, Ziyang Cai, Zheng Zhan, Harry Dong, Ying Fan, Gustavo de Rosa, Tim Pearce, John Langford
  arXiv, 2026

• Almost Free State Prediction Separation
  John Langford, Nathan Godey, Giovanni Monea, Yoav Artzi, Harry Dong, Ying Fan, Gustavo de Rosa, Zheng Zhan
  arXiv, 2026

• Multimodal Diffusion to Mutually Enhance Polarized Light and Low Resolution EBSD Data
  Harry Dong, Timofey Efimov, Megna Shah, Jeff Simmons, Sean Donegan, Marc De Graef, Yuejie Chi
  arXiv, 2026

Journals

• A Lightweight Transformer for Faster and Robust EBSD Data Collection
  Harry Dong, Sean Donegan, Megna Shah, Yuejie Chi
  Scientific Reports, 2024

  • Oral presentation at the Machine Learning for Scientific Imaging Conference at Electronic Imaging, 2024.

  • Poster presentation at the Joint Workshop at the Intersection of Materials Science and Machine Learning, 2023.

• Fast and Provable Tensor Robust Principal Component Analysis via Scaled Gradient Descent
  Harry Dong, Tian Tong, Cong Ma, Yuejie Chi
  Information and Inference: A Journal of the IMA, 2023

  • Contributed talk at SIAM Conference on Mathematics of Data Science, 2022.

Conferences

• Generalized Parallel Scaling with Interdependent Generations
  Harry Dong, David Brandfonbrener, Eryk Helenowski, Yun He, Mrinal Kumar, Han Fang, Yuejie Chi, Karthik Abinav Sankararaman
  International Conference on Learning Representations (ICLR), 2026

  • Oral and best paper nomination at NeurIPS Workshop on Efficient Reasoning, 2025.

• STEM: Scaling Transformers with Embedding Modules
  Ranajoy Sadhukhan, Sheng Cao, Harry Dong, Changsheng Zhao, Attiano Purpura-Pontoniere, Yuandong Tian, Zechun Liu, Beidi Chen
  International Conference on Learning Representations (ICLR), 2026

• Scalable LLM Reasoning Acceleration with Low-rank Distillation
  Harry Dong, Bilge Acun, Beidi Chen, Yuejie Chi
  Conference on Parsimony and Learning (CPAL), 2026

  • Early version presented at ICML Workshop on Long-Context Foundation Models, 2025.

• ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference
  Hanshi Sun, Li-Wen Chang, Wenlei Bao, Size Zheng, Ningxin Zheng, Xin Liu, Harry Dong, Yuejie Chi, Beidi Chen
  International Conference on Machine Learning (ICML) spotlight, 2025

• Leveraging Multimodal Diffusion Models to Accelerate Imaging with Side Information
  Timofey Efimov, Harry Dong, Megna Shah, Jeff Simmons, Sean Donegan, Yuejie Chi
  IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2025

  • Oral presentation/poster at the Computational Imaging Conference at Electronic Imaging, 2025.

• Prompt-prompted Adaptive Structured Pruning for Efficient LLM Generation
  Harry Dong, Beidi Chen, Yuejie Chi
  Conference on Language Modeling (COLM), 2024

  • Oral presentation at ICML Workshop on Efficient Systems for Foundation Models, 2024.

• Get More with LESS: Synthesizing Recurrence with KV Cache Compression for Efficient LLM Inference
  Harry Dong, Xinyu Yang, Zhenyu Zhang, Zhangyang Wang, Yuejie Chi, Beidi Chen
  International Conference on Machine Learning (ICML), 2024

• Deep Unfolded Tensor Robust PCA with Self-supervised Learning
  Harry Dong, Megna Shah, Sean Donegan, Yuejie Chi
  IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2023

  • Also presented at the Third Workshop on Seeking Low‑Dimensionality in Deep Neural Networks, 2023.

• Learning Optimal Traffic Routing Behaviors using Markovian Framework in Microscopic Simulation
  Theophile Cabannes, Jiayi Li, Fangyu Wu, Harry Dong, Alexandre Bayen
  TRB 99th Annual Meeting, 2020

Workshops

• Towards Low-bit Communication for Tensor Parallel LLM Inference
  Harry Dong, Tyler Johnson, Minsik Cho, Emad Soroush
  NeurIPS Workshop on Efficient Natural Language and Speech Processing IV, 2024

• Towards Structured Sparsity in Transformers for Efficient Inference
  Harry Dong, Beidi Chen, Yuejie Chi
  ICML Workshop on Efficient Systems for Foundation Models, 2023

Talks

• Chain-of-Thought Reasoning in Transformers: From Theory to Practice [Slides]
  Conference on Uncertainty in Artificial Intelligence Tutorial, 2026

• Towards Efficient and Performant Generative AI Inference
  Ph.D. Thesis Defense, 2026

• Squeezing the Most Out of LLMs: The Interplay Between Efficiency and Performance
  CPAL Rising Stars Talk, 2026

• Generalized Parallel Scaling with Interdependent Generations
  NeurIPS Workshop on Efficient Reasoning, 2025

• Leveraging Multimodal Diffusion Models to Accelerate Imaging with Side Information
  IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2025
  Data-driven Discovery of Optimized Multifunctional Material Systems Funding Review, 2024

• Prompt-prompted Adaptive Structured Pruning for Efficient LLM Generation
  ICML Workshop on Efficient Systems for Foundation Models, 2024

• A Lightweight Transformer for Faster and Robust EBSD Data Collection
  Machine Learning for Scientific Imaging Conference at Electronic Imaging, 2024
  Data-driven Discovery of Optimized Multifunctional Material Systems Funding Review, 2023

• Recovering Missing Electron Backscatter Diffraction Microscopy Slices using Transformers
  MIrACLE Forum, 2023

• Fast and Provable Tensor Robust Principal Component Analysis via Scaled Gradient Descent
  SIAM MDS, 2022

• Learning Missing EBSD Serial Section Slices
  Data-driven Discovery of Optimized Multifunctional Material Systems Funding Review, 2022