publications

Publications in reversed chronological order. *indicates joint co-authorship. Generated by jekyll-scholar.

Conference and Journal Papers

2026

  1. OSDI
    VTC: DNN Compilation with Virtual Tensors for Data Movement Elimination
    Muyan Hu , Ahan Gupta , Jiachen Yuan , Vima Gupta , Taeksang Kim , Xin Xu , Janardhan Kulkarni , Ofer Dekel , Vikram Adve , and Charith Mendis
    In 20th USENIX Symposium on Operating Systems Design and Implementation (OSDI) (to appear) , Jul 2026
    Novel optimization that goes beyond operator fusion
  2. SAS: Sparse Attention Synthesizer for Efficient Language Model Inference
    Yuan Zhou , Shaojie Xiang , Lingfan Yu , Zhenyu Song , Charith Mendis, and Yida Wang
    In European Conference on Computer Systems (EuroSys) (to appear) , 2026
  3. CGO
    GRANII : Selection and Ordering of Primitives in GRAph Neural Networks using Input Inspection
    Damitha Lenadora , Vimarsh Sathia , Gerasimos Gerogiannis , Serif Yesil , Josep Torrellas , and Charith Mendis
    In International Symposium on Code Generation and Optimization (CGO) , 2026
    func

2025

  1. TensorRight: Automated Verification of Tensor Graph Rewrites
    Jai Arora , Sirui Lu , Devansh Jain , Tianfan Xu , Farzin Houshmand , Phitchaya Mangpo Phothilimthana , Mohsen Lesani , Praveen Narayanan , Karthik Srinivasa Murthy , Rastislav Bodik , Amit Sabne , and Charith Mendis
    In 52nd ACM SIGPLAN Symposium on Principles of Programming Languages , Jan 2025
    Distinguished Paper Award
    First verification work on production-level tensor compilers used in ML
    func
    func
  2. TAIDL: Tensor Accelerator ISA Definition Language with Auto-generation of Scalable Test Oracles
    Devansh Jain , Marco Frigo , Jai Arora , Akash Pardeshi , Zhihao Wang , Krut Patel , and Charith Mendis
    In Proceedings of the 58th IEEE/ACM International Symposium on Microarchitecture , Oct 2025
    First ISA definition language for accelerators
    Used by Amazon for their accelerator offerings
    func
    func
    func
  3. COGNATE: Acceleration of Sparse Tensor Programs on Emerging Hardware using Transfer Learning
    Chamika Sudusinghe , Gerasimos Gerogiannis , Damitha Lenadora , Charles Block , Josep Torrellas , and Charith Mendis
    In Proceedings of the 42nd International Conference on Machine Learning , 13–19 jul 2025
  4. GALA: A High Performance Graph Neural Network Acceleration LAnguage and Compiler
    Damitha Lenadora , Nikhil Jayakumar , Chamika Sudusinghe , and Charith Mendis
    In PACMPL(OOPSLA) , Oct 2025
    func
    func
    func
  5. SPLAT: A framework for optimised GPU code-generation for SParse reguLar ATtention
    Ahan Gupta , Yueming Yuan , Devansh Jain , Yuhao Ge , David Aponte , Yanqi Zhou , and Charith Mendis
    In PACMPL(OOPSLA) , Apr 2025
    func
    func
    func
  6. Automated Verification of Soundness of DNN Certifiers
    Avaljot Singh , Yasmin Sarita , Charith Mendis, and Gagandeep Singh
    In PACMPL(OOPSLA) , Apr 2025
    func
    func
  7. SIGMOD
    PilotDB: Database-Agnostic Online Approximate Query Processing with A Priori Error Guarantees
    Yuxuan Zhu , Tengjun Jin , Stefanos Baziotis , Chengsong Zhang , Charith Mendis, and Daniel Kang
    In Proceedings of ACM on Management of Data (SIGMOD) , Jun 2025
  8. MISAAL: Synthesis-Based Automatic Generation of Efficient and Retargetable Semantics-Driven Optimizations
    Abdul Rafae Noor , Dhruv Baronia , Akash Kothari , Muchen Xu , Charith Mendis, and Vikram Adve
    In 46th ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI) , Jun 2025

2024

  1. SIGMOD
    Dias: Dynamic Rewriting of Pandas Code
    Stefanos Baziotis , Daniel Kang , and Charith Mendis
    In Proc. ACM Manag. Data (SIGMOD) , Mar 2024
    Honorable Mention for the Best Artifact Award
    func
    func
    func
  2. TGLite: A Lightweight Programming Framework for Continuous-Time Temporal Graph Neural Networks
    Yufeng Wang , and Charith Mendis
    In Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) , 2024
    func
    func
    func
  3. Hydride: A Retargetable and Extensible Synthesis-based Compiler for Modern Hardware Architectures
    Akash Kothari , Abdul Rafae Noor , Muchen Xu , Hassam Uddin , Dhruv Baronia , Stefanos Baziotis , Sudipta Sengupta , Charith Mendis, and Vikram Adve
    In Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) , 2024
  4. Two-Face: Combining Collective and One-Sided Accesses for Efficient Distributed SpMM
    Charles Block , Gerasimos Gerogiannis , Charith Mendis, Ariful Azad , and Josep Torrellas
    In Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) , 2024
  5. SAS
    ConstraintFlow: A DSL for Specification and Verification of Neural Network Analyses
    Avaljot Singh , Yasmin Sarita , Charith Mendis, and Gagandeep Singh
    In Static Analysis Symposium , 2024
  6. Towards Efficient Temporal Graph Learning: Algorithms, Frameworks, and Tools
    Ruijie Wang , Wanyu Zhao , Dachun Sun , Charith Mendis, and Tarek Abdelzaher
    In Proceedings of the 33rd ACM International Conference on Information and Knowledge Management , 2024
  7. COMET: X86 Cost Model Explanation Framework
    Isha Chaudhary , Alex Renda , Charith Mendis, and Gagandeep Singh
    2024
  8. TGOnline: Enhancing Temporal Graph Learning with Adaptive Online Meta-Learning
    Ruijie Wang , Jingyuan Huang , Yutong Zhang , Jinyang Li , Yufeng Wang , Wanyu Zhao , Shengzhong Liu , Charith Mendis, and Tarek Abdelzaher
    In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval , 2024

2023

  1. Learning Large Graph Property Prediction via Graph Segment Training
    Kaidi Cao , Mangpo Phothilimthana , Sami Abu-El-Haija , Dustin Zelle , Yanqi Zhou , Charith Mendis, Jure Leskovec , and Bryan Perozzi
    In Advances in Neural Information Processing Systems , 2023
  2. TpuGraphs: A Performance Prediction Dataset on Large Tensor Computational Graphs
    Mangpo Phothilimthana , Sami Abu-El-Haija , Kaidi Cao , Bahare Fatemi , Michael Burrows , Charith Mendis, and Bryan Perozzi
    In Advances in Neural Information Processing Systems , 2023
  3. MetaCom
    Challenges in Metaverse Research: An Internet of Things Perspective
    Tarek Abdelzaher , Matthew Caesar , Charith Mendis, Klara Nahrstedt , Mani Srivastava , and Minlan Yu
    In IEEE International Conference on Metaverse Computing, Networking and Applications (MetaCom) , 2023
  4. Unified Convolution Framework: A compiler-based approach to support sparse convolutions
    Jaeyeon Won , Changwan Hong , Charith Mendis, Joel Emer , and Saman Amarasinghe
    In Proceedings of Machine Learning and Systems (MLSys) , 2023
    func
    func
  5. SPADE: A Flexible and Scalable Accelerator for SpMM and SDDMM
    Gerasimos Gerogiannis , Serif Yesil , Damitha Lenadora , Dingyuan Cao , Charith Mendis, and Josep Torrellas
    In Proceedings of the 50th Annual International Symposium on Computer Architecture (ISCA) , 2023
    IEEE Top Picks Honorable Mention
  6. TGOpt: Redundancy-Aware Optimizations for Temporal Graph Attention Networks
    Yufeng Wang , and Charith Mendis
    In Proceedings of the 28th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming (PPoPP) , 2023
    func
    func
    func
  7. WACO: Learning Workload-Aware Co-optimization of the Format and Schedule of a Sparse Tensor Program
    Jaeyeon Won , Charith Mendis, Joel S. Emer , and Saman P. Amarasinghe
    In Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) , 2023
    func
    func

2022

  1. IISWC
    GRANITE: A Graph Neural Network Model for Basic Block Throughput Estimation
    Ondrej Sýkora , Phitchaya Mangpo Phothilimthana , Charith Mendis, and Amir Yazdanbakhsh
    In IEEE International Symposium on Workload Characterization (IISWC) , 2022
  2. All You Need is Superword-Level Parallelism: Systematic Control-Flow Vectorization with SLP
    Yishen Chen , Charith Mendis, and Saman Amarasinghe
    In Proceedings of the 43rd ACM SIGPLAN International Conference on Programming Language Design and Implementation (PLDI) , 2022
    func
    func

2021

  1. VeGen: A Vectorizer Generator for SIMD and Beyond
    Yishen Chen , Charith Mendis, Michael Carbin , and Saman Amarasinghe
    In Proceedings of the 26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) , 2021
    func
    func
  2. A Learned Performance Model for Tensor Processing Units
    Sam Kaufman , Phitchaya Phothilimthana , Yanqi Zhou , Charith Mendis, Sudip Roy , Amit Sabne , and Mike Burrows
    In Proceedings of Machine Learning and Systems (MLSys) , 2021
    Used in production at Google in the XLA TPU compiler

2020

  1. DiffTune: Optimizing CPU Simulator Parameters with Learned Differentiable Surrogates
    Alex Renda , Yishen Chen , Charith Mendis, and Michael Carbin
    In 53rd Annual IEEE/ACM International Symposium on Microarchitecture (MICRO) , 2020

2019

  1. CC
    Revec: program rejuvenation through revectorization
    Charith Mendis, Ajay Jain , Paras Jain , and Saman P. Amarasinghe
    In Proceedings of the 28th International Conference on Compiler Construction (CC) , 2019
  2. Ithemal: Accurate, Portable and Fast Basic Block Throughput Estimation using Deep Neural Networks
    Charith Mendis, Alex Renda , Saman P. Amarasinghe , and Michael Carbin
    In Proceedings of the 36th International Conference on Machine Learning (ICML) , 2019
    Best Paper award at the ML for systems workshop co-located with ISCA’19
  3. IISWC
    BHive: A Benchmark Suite and Measurement Framework for Validating x86-64 Basic Block Performance Models
    Yishen Chen , Ajay Brahmakshatriya , Charith Mendis, Alex Renda , Eric Atkinson , Ondrej Sýkora , Saman P. Amarasinghe , and Michael Carbin
    In IEEE International Symposium on Workload Characterization (IISWC) , 2019
  4. Compiler Auto-Vectorization with Imitation Learning
    Charith Mendis, Cambridge Yang , Yewen Pu , Saman P. Amarasinghe , and Michael Carbin
    In Advances in Neural Information Processing Systems 32: Annual Conference on Neural Information Processing Systems (NeurIPS) , 2019

2018

  1. goSLP: globally optimized superword level parallelism framework
    Charith Mendis, and Saman P. Amarasinghe
    Proc. ACM Program. Lang. (OOPSLA), 2018
    func

2017

  1. BigData
    Making caches work for graph analytics
    Yunming Zhang , Vladimir Kiriansky , Charith Mendis, Saman P. Amarasinghe , and Matei Zaharia
    In IEEE International Conference on Big Data (BigData) , 2017
    Best Student Paper award

2016

  1. ICASSP
    Parallelizing WFST speech decoders
    Charith Mendis, Jasha Droppo , Saeed Maleki , Madanlal Musuvathi , Todd Mytkowicz , and Geoffrey Zweig
    In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , 2016

2015

  1. Helium: lifting high-performance stencil kernels from stripped x86 binaries to halide DSL code
    Charith Mendis, Jeffrey Bosboom , Kevin Wu , Shoaib Kamil , Jonathan Ragan-Kelley , Sylvain Paris , Qin Zhao , and Saman P. Amarasinghe
    In Proceedings of the 36th ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI) , 2015