Shadowgraph
TrendingAbout
Sign in

SJTU-ReArch-Group/Paper-Reading-List

158

216 commits

updated Jul 24, 2026

rearch.github.io

See the code

README

ReArch Group Paper Reading List

 

Seminars

Spring 2021

DatePaper TitlePresenterNotes
03.01Training for Multi-resolution Inference Using Reusable Quantization TermsCong Guo
03.08Toward Efficient Interactions between Python and Native LibrariesYuxian Qiu
03.15SpAtten: Efficient Natural Language ProcessingYue Guan
03.22X-Stream: Edge-centric Graph Processing using Streaming PartitionsZhihui Zhang
03.29Loop Nested Optimization, Polyhedral Model and Micro-2020 Best Paper (Optimizing the Memory Hierarchy by Compositing Automatic Transformations on Computations and Data)Zihan LiuSlides
04.12Defensive Approximation: Securing CNNs using Approximate ComputingYakai WangRelated Work
05.17Commutative Data Reordering: A New Technique to Reduce Data Movement Energy on Sparse Inference WorkloadsYangjie ZhouISCA 2020
05.31Large Graph Convolutional Network Training with GPU-Oriented Data Communication ArchitectureZhihui ZhangVLDB 2021
06.07DynamicViT: Efficient Vision Transformers with Dynamic Token SparsificationYue GuanNeurIPS 2021

Summer 2021

DatePaper TitlePresenterNotes
07.14AKG: automatic kernel generation for neural processing units using polyhedral transformations (PLDI 2021)Yuxian QiuSlides
07.21Floating-Point Format and Quantization for Deep Learning ComputationCong Guo
07.28P-OPT: Practical Optimal Cache Replacement for Graph AnalyticsYangjie ZhouSlides
08.04Rubik: A Hierarchical Architecture for Efficient Graph Neural Network TrainingZhihui Zhang
08.11A Useful Tool CKA: Similarity of Neural Network Representations Revisited and It's application: Uncovering How Neural Network Representations Vary with Width and DepthZhengyi LiSlides
08.18Ansor: Generating High-Performance Tensor Programs for Deep LearningZihan LiuSlides

Fall 2021

DatePaper TitlePresenterNotes
10.11Adaptive numeric type for DNN quantizationCong Guo
10.18Compiling Graph Applications for GPUs with GraphItYangjie ZhouSlides
11.01TENET: A Framework for Modeling Tensor Dataflow Based on Relation-centric NotationZihan LiuSlides
11.08Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient SparsityZhengyi LiSlides (code: zdea)
11.22Dynamic Tensor Rematerialization
Checkmate: Breaking The Memory Wall with Optimal Tensor Rematerialization		Yue GuanSlides
Slides
11.29GraphPulse: An Event-Driven Hardware Accelerator for Asynchronous Graph ProcessingZhihui ZhangPresentation
12.06CheckFreq: Frequent, Fine-Grained DNN CheckpointingGuandong LuSlides
12.13PipeDream: generalized pipeline parallelism for DNN trainingRunzhe ChenSlides
12.20Towards Scalable Distributed Training of Deep Learning on Public Cloud ClustersYakai WangSlides

Spring 2022

DatePaper TitlePresenterNotes
3.10Speculation Attack: Meltdown, Spectre, Pinned-LoadsZihan LiuSlides
3.24SparTA: Deep-Learning Model Sparsity via Tensor-with-Sparsity-AttributeYue Guan
3.31ROLLER: Fast and Efficient Tensor Compilation for Deep LearningYijia DiaoLink
4.07Adaptable Register File Organization for Vector ProcessorsZhihui Zhang
4.14CORTEX: A COMPILER FOR RECURSIVE DEEP LEARNING MODELSYangjie ZhouSlides
4.21Zero-Knowledge Succinct Non-Interactive Argument of KnowledgeShuwen LuSlides
5.05Alpa: Automating Inter- and Intra-Operator Parallelism for Distributed Deep LearningRunzhe ChenSlides

Fall 2022

DatePaper TitlePresenterNotes
9.20ANT: Exploiting Adaptive Numerical Data Type for Low-bit Deep Neural Network QuantizationCong GuoSlides
9.27X-cache: a modular architecture for domain-specific cachesZihan LiuSlides
10.18Automatically Discovering ML OptimizationsYangjie ZhouSlides
11.8Privacy Preserving Machine Learning--inferenceZhengyi LiSlides
11.15Dynamic Tensor CompilersYijia DiaoSlides

Spring 2023

DatePaper TitlePresenterNotes
3.30JUNO: Algorithm-Hardware Mapping Co-design for Efficient\Approximate Nearest Neighbour Search in High Dimensional SpaceZihan Liu
4.6LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale; SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models; Optimal Brain Compression: A Framework for Accurate Post-Training Quantization and Pruning; GPTQ: ACCURATE POST-TRAINING QUANTIZATION FOR GENERATIVE PRE-TRAINED; SparseGPT: Massive Language Models Can be Accurately Pruned in One-Shot; P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks; Offsite-Tuning: Transfer Learning without Full Model; LoRA: Low-Rank Adaptation of Large Language ModelsJiaming TangSlides
4.13SMG: Towards Efficient Execution and Adequate Encryption of Private DNN Inference via Secure Micro-GraphZhengyi LiSlides
5.04FlexGen and FlashAttentionYue GuanSlides
5.11Multi-Tenant DNN Inference: Spatial GPU SharingYijia DiaoSlides
5.25Chimera: An Analytical Optimizing Framework for Effective Compute-intensive Operators FusionYangjie ZhouTVMConf Video

Fall 2023

DatePaper TitlePresenterNotes
9.21GPU Warp Scheduling and Control CodeWeiming HuSlides
9.28Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured SparsityYue GuanSlides
10.12Shared SIMD unit: Occamy, Two Out-of-Order Commit CPU: NOREBA and OrinocoZihan LiuSlides
10.19Multitasking on GPU: PreemptionYijia DiaoSlides
10.26SecretFlow-SPU: A Performant and User-Friendly Framework for Privacy-Preserving Machine LearningZhengyi LiSlides
11.09Efficient large-scale language model training on GPU clusters using megatron-LM; ZeRO: Memory Optimizations Toward Training Trillion Parameter Models; ZeRO-Offload: Democratizing Billion-Scale Model Training; ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep LearningJiale XuSlides
11.16ATOM: LOW-BIT QUANTIZATION FOR EFFICIENT AND ACCURATE LLM SERVINGHaoyan ZhangSlides
11.23DFU: Dataflow Processing UnitRenyang GuanSlides
12.07WaveScalar;Think Fast: A Tensor Streaming Processor (TSP) for Accelerating Deep Learning WorkloadsGonglin XuSlides
12.14Fast Inference from Transformers via Speculative Decoding;SpecInfer: Accelerating Generative Large Language Model Serving with Speculative Inference and Token Tree Verification;LLMCad: Fast and Scalable On-device Large Language Model InferenceChangming YuSlides
12.28A Framework for Fine-Grained Synchronization of Dependent GPU Kernels;Fast Fine-Grained Global Synchronization on GPUs;AutoScratch: ML-Optimized Cache Management for Inference-Oriented GPUsZiyu HuangSlides

Spring 2024

DatePaper TitlePresenterNotes
03.14LLM Attack and DefenseZhengyi LiSlides
03.21Transparent GPU Sharing in Container Clouds for Deep Learning WorkloadsYijia DiaoLink
03.28DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model ServingShuwen LuSlide
05.098-bit Transformer Inference and Fine-tuning for Edge AcceleratorsWeiming HuSlide

Fall 2024

DatePaper TitlePresenterNotes
07.26TEE-SGX IntroductionZhengyi LiSlides
08.15Accelerating mixture of experts model InferenceShuwen LuSlides
08.22Accelerating Stable Diffusion-based Video GenerationYuge ChengSlides
09.05TCP: A Tensor Contraction Processor for AI WorkloadsWeiming HuSlides
10.11dLoRA: Dynamically Orchestrating Requests and Adapters for LoRA LLM ServingHaoyan ZhangSlides
10.18Dataflow Chips and a CompilerRenyang GuanSlides
11.01ByteCheckpoint A Unified Checkpointing System for Large Foundation Model DevelopmentGonglin XuSlides
11.15Opensora architecture and its computational reuseHaosong LiuSlides
11.22LLM QuantizationWenxuan MiaoSlides
11.29Survey: Large-scale 3DGSZheng LiuSlides
12.05Enhance Efficiency: 3D Gaussian Splatting for Speed and Memory OptimizationXiaotong HuangSlides
12.13Stealing Part of a Production Language ModelZhengyi LiSlides
12.20Communication-Compute Co-Optimization in Distributed TrainingYijia DiaoSlides
12.27Byte Latent Transformer: Patches Scale Better Than TokensShuyong BaoSlides
01.03Gemini Mapping and Architecture Co-exploration for Large-scale DNN Chiplet AcceleratorsRenyang GuanSlides

Spring 2025

DatePaper TitlePresenterNotes
01.17HybridFlow: A Flexible and Efficient RLHF FrameworkGonglin XuSlides
02.28Chimera: An Analytical Optimizing Framework for Effective Compute-intensive Operators FusionZiyu HuangSlides
03.14Auto-Vectorization in Compilers: Leveraging SIMD for HighPerformance ComputingShihan FangSlides
03.21Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse AttentionXing MaSlides
03.28Taming Load Balancing in Distributed LLM TrainingJiale XuSlides
04.11SparseAttn for Video GenerationYulin SunSlides
04.18Towards End-to-End Optimization of LLM-based Applications with AyoJiawei HuangSlides
04.25FSMoE: A Flexible and Scalable Training System for Sparse Mixture-of-Experts ModelsXiaotong HuangSlides
05.23EXION: Exploiting Inter- and Intra-Iteration Output Sparsity for Diffusion ModelsYuge ChenSlides
05.30Modern Programming Model for Writing Kernels on GPUsXinhao LuoSlides
06.06Prefix Sharing LLM Inference and SGLangYitong DingSlides
06.13Ditto: Accelerating Diffusion Model via Temporal Value Similarity CMC: Video Transformer Acceleration via CODEC Assisted Matrix CondensingHaosong LiuSlides
06.21-25ISCA conference notesLink
06.27Speeding up LLM and GEMMWenxuan MiaoSlides
07.25Modeling and SimulationWeiming HuSlides

Fall 2025

DatePaper TitlePresenterNotes
09.18Fine Grained Comm Comp OverlapZiyu HuangSlides
09.26A Sample-Free Compilation Framework for Efficient Dynamic Tensor ComputationYangjie ZhouSlides
10.17Hydra: Harnessing Expert Popularity for Efficient Mixture-of-Expert Inference on Chiplet SystemGonglin XuSlides
10.24LLM for cuda codegenMa XingSlides
11.14A Survey of 3DGS SLAMXiaotong HuangSlides
11.21Survey of Vision-Language-Action (VLA) ModelsZheng LiuSlides
11.28Modern DSLs Compile WorkflowXinhao LuoSlides

Spring 2026

DatePaper TitlePresenterNotes
03.06Breaking the Layer Barrier: Remodeling Private Transformer Inference with Hybrid CKKS and MPCZhengyi LiSlides
03.13Where LLMs Fit, and Where We Still MatterYijia DiaoSlides1, Paper2
03.27GPU Profiling for OptimizationWu SunSlides
05.08WarpDrive: GPU-Based Fully Homomorphic Encryption Acceleration Leveraging Tensor and CUDA CoresLiukun YuSlides1,Paper2
05.14New Paradigms for KV Cache ReuseXing MaSlides
05.29从 prompt 到 context 到 harnessJiawei HuangSlides
06.06From MHA to DSA: The Evolution of Attention Paradigms in Long-Context LLMsXiaotong HuangSlides
06.12Attention Quantization in Video Diffusion ModelsYuge ChengSlides
07.03Sort-free 3dgsHe ZhuSlides
07.17LLM for Kernel generationXinhao LuoSlides

DNN Architecture

Link

 

Deep Learning Compiler

List Contributed by Zihan Liu

 

Past Architecture Papers

List Contributed by Jingwen Leng

 

List Contributed by Shuwen Lu

Quantization, Data Type, Compression, Acceleration

List Contributed by Weiming Hu

LLM for Coding Papers

List Contributed by Ma Xing and Yangjie Zhou

Reading List From Other Groups

architecture
paperlist

Contributors

2436824987

58 commits

ziyuhuang123

51 commits

huweim

19 commits

SubjectNoi

13 commits

Similar repos

lzyrapx/Papers-Books-Reading

📚Some papers & books I’ve read.

31

xlite-dev/Awesome-LLM-Inference

📚A curated list of Awesome LLM/VLM Inference Papers with Codes: Flash-Attention, Paged-Attention,…

5,515

tiingweii-shii/Awesome-Resource-Efficient-LLM-Papers

a curated list of high-quality papers on resource-efficient LLMs 🌱

165

PasaLab/LLM_Paper_Learning

LLM Papers We Recommend to Read

59

HuangOwen/Awesome-LLM-Compression

Awesome LLM compression research papers and tools.

1,879

AmadeusChan/Awesome-LLM-System-Papers

645

horseee/Awesome-Efficient-LLM

A curated list for Efficient Large Language Models

2,043

byungsoo-oh/ml-systems-papers

Curated collection of papers in machine learning systems

655

SJTU-ReArch-Group/Paper-Reading-List

158

216 commits

updated Jul 24, 2026

rearch.github.io

See the code

README

ReArch Group Paper Reading List

 

Seminars

Spring 2021

DatePaper TitlePresenterNotes
03.01Training for Multi-resolution Inference Using Reusable Quantization TermsCong Guo
03.08Toward Efficient Interactions between Python and Native LibrariesYuxian Qiu
03.15SpAtten: Efficient Natural Language ProcessingYue Guan
03.22X-Stream: Edge-centric Graph Processing using Streaming PartitionsZhihui Zhang
03.29Loop Nested Optimization, Polyhedral Model and Micro-2020 Best Paper (Optimizing the Memory Hierarchy by Compositing Automatic Transformations on Computations and Data)Zihan LiuSlides
04.12Defensive Approximation: Securing CNNs using Approximate ComputingYakai WangRelated Work
05.17Commutative Data Reordering: A New Technique to Reduce Data Movement Energy on Sparse Inference WorkloadsYangjie ZhouISCA 2020
05.31Large Graph Convolutional Network Training with GPU-Oriented Data Communication ArchitectureZhihui ZhangVLDB 2021
06.07DynamicViT: Efficient Vision Transformers with Dynamic Token SparsificationYue GuanNeurIPS 2021

Summer 2021

DatePaper TitlePresenterNotes
07.14AKG: automatic kernel generation for neural processing units using polyhedral transformations (PLDI 2021)Yuxian QiuSlides
07.21Floating-Point Format and Quantization for Deep Learning ComputationCong Guo
07.28P-OPT: Practical Optimal Cache Replacement for Graph AnalyticsYangjie ZhouSlides
08.04Rubik: A Hierarchical Architecture for Efficient Graph Neural Network TrainingZhihui Zhang
08.11A Useful Tool CKA: Similarity of Neural Network Representations Revisited and It's application: Uncovering How Neural Network Representations Vary with Width and DepthZhengyi LiSlides
08.18Ansor: Generating High-Performance Tensor Programs for Deep LearningZihan LiuSlides

Fall 2021

DatePaper TitlePresenterNotes
10.11Adaptive numeric type for DNN quantizationCong Guo
10.18Compiling Graph Applications for GPUs with GraphItYangjie ZhouSlides
11.01TENET: A Framework for Modeling Tensor Dataflow Based on Relation-centric NotationZihan LiuSlides
11.08Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient SparsityZhengyi LiSlides (code: zdea)
11.22Dynamic Tensor Rematerialization
Checkmate: Breaking The Memory Wall with Optimal Tensor Rematerialization		Yue GuanSlides
Slides
11.29GraphPulse: An Event-Driven Hardware Accelerator for Asynchronous Graph ProcessingZhihui ZhangPresentation
12.06CheckFreq: Frequent, Fine-Grained DNN CheckpointingGuandong LuSlides
12.13PipeDream: generalized pipeline parallelism for DNN trainingRunzhe ChenSlides
12.20Towards Scalable Distributed Training of Deep Learning on Public Cloud ClustersYakai WangSlides

Spring 2022

DatePaper TitlePresenterNotes
3.10Speculation Attack: Meltdown, Spectre, Pinned-LoadsZihan LiuSlides
3.24SparTA: Deep-Learning Model Sparsity via Tensor-with-Sparsity-AttributeYue Guan
3.31ROLLER: Fast and Efficient Tensor Compilation for Deep LearningYijia DiaoLink
4.07Adaptable Register File Organization for Vector ProcessorsZhihui Zhang
4.14CORTEX: A COMPILER FOR RECURSIVE DEEP LEARNING MODELSYangjie ZhouSlides
4.21Zero-Knowledge Succinct Non-Interactive Argument of KnowledgeShuwen LuSlides
5.05Alpa: Automating Inter- and Intra-Operator Parallelism for Distributed Deep LearningRunzhe ChenSlides

Fall 2022

DatePaper TitlePresenterNotes
9.20ANT: Exploiting Adaptive Numerical Data Type for Low-bit Deep Neural Network QuantizationCong GuoSlides
9.27X-cache: a modular architecture for domain-specific cachesZihan LiuSlides
10.18Automatically Discovering ML OptimizationsYangjie ZhouSlides
11.8Privacy Preserving Machine Learning--inferenceZhengyi LiSlides
11.15Dynamic Tensor CompilersYijia DiaoSlides

Spring 2023

DatePaper TitlePresenterNotes
3.30JUNO: Algorithm-Hardware Mapping Co-design for Efficient\Approximate Nearest Neighbour Search in High Dimensional SpaceZihan Liu
4.6LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale; SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models; Optimal Brain Compression: A Framework for Accurate Post-Training Quantization and Pruning; GPTQ: ACCURATE POST-TRAINING QUANTIZATION FOR GENERATIVE PRE-TRAINED; SparseGPT: Massive Language Models Can be Accurately Pruned in One-Shot; P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks; Offsite-Tuning: Transfer Learning without Full Model; LoRA: Low-Rank Adaptation of Large Language ModelsJiaming TangSlides
4.13SMG: Towards Efficient Execution and Adequate Encryption of Private DNN Inference via Secure Micro-GraphZhengyi LiSlides
5.04FlexGen and FlashAttentionYue GuanSlides
5.11Multi-Tenant DNN Inference: Spatial GPU SharingYijia DiaoSlides
5.25Chimera: An Analytical Optimizing Framework for Effective Compute-intensive Operators FusionYangjie ZhouTVMConf Video

Fall 2023

DatePaper TitlePresenterNotes
9.21GPU Warp Scheduling and Control CodeWeiming HuSlides
9.28Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured SparsityYue GuanSlides
10.12Shared SIMD unit: Occamy, Two Out-of-Order Commit CPU: NOREBA and OrinocoZihan LiuSlides
10.19Multitasking on GPU: PreemptionYijia DiaoSlides
10.26SecretFlow-SPU: A Performant and User-Friendly Framework for Privacy-Preserving Machine LearningZhengyi LiSlides
11.09Efficient large-scale language model training on GPU clusters using megatron-LM; ZeRO: Memory Optimizations Toward Training Trillion Parameter Models; ZeRO-Offload: Democratizing Billion-Scale Model Training; ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep LearningJiale XuSlides
11.16ATOM: LOW-BIT QUANTIZATION FOR EFFICIENT AND ACCURATE LLM SERVINGHaoyan ZhangSlides
11.23DFU: Dataflow Processing UnitRenyang GuanSlides
12.07WaveScalar;Think Fast: A Tensor Streaming Processor (TSP) for Accelerating Deep Learning WorkloadsGonglin XuSlides
12.14Fast Inference from Transformers via Speculative Decoding;SpecInfer: Accelerating Generative Large Language Model Serving with Speculative Inference and Token Tree Verification;LLMCad: Fast and Scalable On-device Large Language Model InferenceChangming YuSlides
12.28A Framework for Fine-Grained Synchronization of Dependent GPU Kernels;Fast Fine-Grained Global Synchronization on GPUs;AutoScratch: ML-Optimized Cache Management for Inference-Oriented GPUsZiyu HuangSlides

Spring 2024

DatePaper TitlePresenterNotes
03.14LLM Attack and DefenseZhengyi LiSlides
03.21Transparent GPU Sharing in Container Clouds for Deep Learning WorkloadsYijia DiaoLink
03.28DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model ServingShuwen LuSlide
05.098-bit Transformer Inference and Fine-tuning for Edge AcceleratorsWeiming HuSlide

Fall 2024

DatePaper TitlePresenterNotes
07.26TEE-SGX IntroductionZhengyi LiSlides
08.15Accelerating mixture of experts model InferenceShuwen LuSlides
08.22Accelerating Stable Diffusion-based Video GenerationYuge ChengSlides
09.05TCP: A Tensor Contraction Processor for AI WorkloadsWeiming HuSlides
10.11dLoRA: Dynamically Orchestrating Requests and Adapters for LoRA LLM ServingHaoyan ZhangSlides
10.18Dataflow Chips and a CompilerRenyang GuanSlides
11.01ByteCheckpoint A Unified Checkpointing System for Large Foundation Model DevelopmentGonglin XuSlides
11.15Opensora architecture and its computational reuseHaosong LiuSlides
11.22LLM QuantizationWenxuan MiaoSlides
11.29Survey: Large-scale 3DGSZheng LiuSlides
12.05Enhance Efficiency: 3D Gaussian Splatting for Speed and Memory OptimizationXiaotong HuangSlides
12.13Stealing Part of a Production Language ModelZhengyi LiSlides
12.20Communication-Compute Co-Optimization in Distributed TrainingYijia DiaoSlides
12.27Byte Latent Transformer: Patches Scale Better Than TokensShuyong BaoSlides
01.03Gemini Mapping and Architecture Co-exploration for Large-scale DNN Chiplet AcceleratorsRenyang GuanSlides

Spring 2025

DatePaper TitlePresenterNotes
01.17HybridFlow: A Flexible and Efficient RLHF FrameworkGonglin XuSlides
02.28Chimera: An Analytical Optimizing Framework for Effective Compute-intensive Operators FusionZiyu HuangSlides
03.14Auto-Vectorization in Compilers: Leveraging SIMD for HighPerformance ComputingShihan FangSlides
03.21Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse AttentionXing MaSlides
03.28Taming Load Balancing in Distributed LLM TrainingJiale XuSlides
04.11SparseAttn for Video GenerationYulin SunSlides
04.18Towards End-to-End Optimization of LLM-based Applications with AyoJiawei HuangSlides
04.25FSMoE: A Flexible and Scalable Training System for Sparse Mixture-of-Experts ModelsXiaotong HuangSlides
05.23EXION: Exploiting Inter- and Intra-Iteration Output Sparsity for Diffusion ModelsYuge ChenSlides
05.30Modern Programming Model for Writing Kernels on GPUsXinhao LuoSlides
06.06Prefix Sharing LLM Inference and SGLangYitong DingSlides
06.13Ditto: Accelerating Diffusion Model via Temporal Value Similarity CMC: Video Transformer Acceleration via CODEC Assisted Matrix CondensingHaosong LiuSlides
06.21-25ISCA conference notesLink
06.27Speeding up LLM and GEMMWenxuan MiaoSlides
07.25Modeling and SimulationWeiming HuSlides

Fall 2025

DatePaper TitlePresenterNotes
09.18Fine Grained Comm Comp OverlapZiyu HuangSlides
09.26A Sample-Free Compilation Framework for Efficient Dynamic Tensor ComputationYangjie ZhouSlides
10.17Hydra: Harnessing Expert Popularity for Efficient Mixture-of-Expert Inference on Chiplet SystemGonglin XuSlides
10.24LLM for cuda codegenMa XingSlides
11.14A Survey of 3DGS SLAMXiaotong HuangSlides
11.21Survey of Vision-Language-Action (VLA) ModelsZheng LiuSlides
11.28Modern DSLs Compile WorkflowXinhao LuoSlides

Spring 2026

DatePaper TitlePresenterNotes
03.06Breaking the Layer Barrier: Remodeling Private Transformer Inference with Hybrid CKKS and MPCZhengyi LiSlides
03.13Where LLMs Fit, and Where We Still MatterYijia DiaoSlides1, Paper2
03.27GPU Profiling for OptimizationWu SunSlides
05.08WarpDrive: GPU-Based Fully Homomorphic Encryption Acceleration Leveraging Tensor and CUDA CoresLiukun YuSlides1,Paper2
05.14New Paradigms for KV Cache ReuseXing MaSlides
05.29从 prompt 到 context 到 harnessJiawei HuangSlides
06.06From MHA to DSA: The Evolution of Attention Paradigms in Long-Context LLMsXiaotong HuangSlides
06.12Attention Quantization in Video Diffusion ModelsYuge ChengSlides
07.03Sort-free 3dgsHe ZhuSlides
07.17LLM for Kernel generationXinhao LuoSlides

DNN Architecture

Link

 

Deep Learning Compiler

List Contributed by Zihan Liu

 

Past Architecture Papers

List Contributed by Jingwen Leng

 

List Contributed by Shuwen Lu

Quantization, Data Type, Compression, Acceleration

List Contributed by Weiming Hu

LLM for Coding Papers

List Contributed by Ma Xing and Yangjie Zhou

Reading List From Other Groups

architecture
paperlist

Similar repos

lzyrapx/Papers-Books-Reading

📚Some papers & books I’ve read.

31

xlite-dev/Awesome-LLM-Inference

📚A curated list of Awesome LLM/VLM Inference Papers with Codes: Flash-Attention, Paged-Attention,…

5,515

tiingweii-shii/Awesome-Resource-Efficient-LLM-Papers

a curated list of high-quality papers on resource-efficient LLMs 🌱

165

PasaLab/LLM_Paper_Learning

LLM Papers We Recommend to Read

59

HuangOwen/Awesome-LLM-Compression

Awesome LLM compression research papers and tools.

1,879

AmadeusChan/Awesome-LLM-System-Papers

645

horseee/Awesome-Efficient-LLM

A curated list for Efficient Large Language Models

2,043

byungsoo-oh/ml-systems-papers

Curated collection of papers in machine learning systems

655

Contributors

2436824987

58 commits

ziyuhuang123

51 commits

huweim

19 commits

SubjectNoi

13 commits