Large Language Model (LLM)

I am actively maintaining this list.

LLM Training

Hybrid parallelism

• Accelerating the Training of Large Language Models using Efficient Activation Rematerialization and Optimal Hybrid Parallelism (ATC 2024) [Paper] [Code]

  • Kuaishou

• Alpa: Automating Inter- and Intra-Operator Parallelism for Distributed Deep Learning (OSDI 2022) [Paper] [Code] [Docs]

  • UC Berkeley & AWS & Google & SJTU & CMU & Duke

  • Generalize the search through parallelism strategies.

Fault tolerance

• Oobleck: Resilient Distributed Training of Large Models Using Pipeline Templates (SOSP 2023) [Paper] [arXiv] [Code]

  • UMich SymbioticLab & AWS & PKU

• Gemini: Fast Failure Recovery in Distributed Training with In-Memory Checkpoints (SOSP 2023) [Paper]

  • Rice & AWS

• Bamboo: Making Preemptible Instances Resilient for Affordable Training of Large DNNs (NSDI 2023) [Paper] [Code]

  • UCLA & CMU & MSR & Princeton

  • Resilient distributed training

LLM Inference

• CacheGen: KV Cache Compression and Streaming for Fast Large Language Model Serving (SIGCOMM 2024) [arXiv] [Code] [Video]

  • UChicago & Microsoft & Stanford

• Tender: Accelerating Large Language Models via Tensor Decomposition and Runtime Requantization (ISCA 2024)
• ALISA: Accelerating Large Language Model Inference via Sparsity-Aware KV Caching (ISCA 2024)
• LLM in a flash: Efficient Large Language Model Inference with Limited Memory (arXiv 2312.11514) [arXiv]

  • Apple

• SpotServe: Serving Generative Large Language Models on Preemptible Instances (ASPLOS 2024) [Personal Notes] [arXiv] [Code]

  • CMU & PKU & CUHK

• Fast Distributed Inference Serving for Large Language Models (arXiv 2305.05920) [Paper]

  • PKU

  • Skip-join multi-level feedback queue scheduling instead of first-come-frist-serve.

  • Proactive KV cache swapping.

  • Compared to Orca

• AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving (OSDI 2023) [Paper] [Code]

  • UC Berkeley & PKU & UPenn & Stanford & Google

  • Trade-off between the overhead of model parallelism and reduced serving latency by statistical multiplexing.

• Efficiently Scaling Transformer Inference (MLSys 2023) [Paper]

  • Google

  • Outstanding Paper Award

  • Model partitioning; PaLM; TPUv4

• DeepSpeed-Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale (SC 2022) [Paper] [Code] [Homepage]

  • Microsoft DeepSpeed

  • Leverage CPU/NVMe/GPU memory.

LLM-based Applications

• Teola: Towards End-to-End Optimization of LLM-based Applications (ASPLOS 2025) [arXiv]

  • CUHK

  • An orchestration framework for LLM-based applications: utilize task primitives as the basic units; represent each query’s workflow as a primitive-level dataflow graph.

  • Enable larger design space for optimization including graph optimization (i.e., parallelization and pipelining) and application-aware scheduling.

• Parrot: Efficient Serving of LLM-based Applications with Semantic Variable (OSDI 2024) [Paper] [Code]

  • SJTU & MSRA

• SGLang: Efficient Execution of Structured Language Model Programs (NeurIPS 2024) [Personal Notes] [Paper] [arXiv] [Code]

  • UC Berkeley & Stanford

  • Co-design the front-end programming interface and back-end serving runtime

  • SGLang; SGVM w/ RadixAttention

  • Reuse KV cache across multiple calls and programs

Retrieval-Augmented Generation (RAG)

• CacheFocus: Dynamic Cache Re-Positioning for Efficient Retrieval-Augmented Generation (arXiv:2502.11101) [arXiv]

  • Jeonbuk National University & Seoul National University

  • Leverage query-independent, offline caching to reuse a context KV cache store.

  • Cache Re-Positioning: shift keys to different positions in the encoding space.

  • Layer-Adaptive Cache Pruning: discard low-relevance caches for documents during pre-filling.

  • Adaptive Positional Allocation: adjust cache positions to maximize the use of the available positional encoding range.

• Cache-Craft: Managing Chunk-Caches for Efficient Retrieval-Augmented Generation (SIGMOD 2025) [arXiv]

  • Adobe Research & IIT Bombay & IIT Kanpur

  • Identify the reusability of chunk-caches; perform a small fraction of recomputation to fix the cache to maintain output quality; store and evict chunk-caches.

  • A wrapper around vLLM; built on Xformers backend optimized with Triton.

• RAGCache: Efficient Knowledge Caching for Retrieval-Augmented Generation (arXiv:2404.12457) [arXiv]

  • PKU & ByteDance

  • Organize the intermediate states of retrieved knowledge in a knowledge tree; cache them in the GPU and host memory.

  • Replacement policy: evaluate each node based on its access frequency, size, and access cost.

    • Priority= Clock + (Frequency × Cost Size) / Size

    • Nodes with lower priority are evicted first.

  • Built on vLLM.

Request Scheduling

• Llumnix: Dynamic Scheduling for Large Language Model Serving (OSDI 2024) [Paper] [Code]

  • Alibaba

• Orca: A Distributed Serving System for Transformer-Based Generative Models (OSDI 2022) [Personal Notes] [Paper]

  • Seoul National University & FriendliAI

  • Iteration-level scheduling; selective batching.

KV Cache Management

• Efficient Memory Management for Large Language Model Serving with PagedAttention (SOSP 2023) [Paper] [arXiv] [Code] [Homepage]

  • UC Berkeley & Stanford & UCSD

  • vLLM, PagedAttention

  • Partition the KV cache of each sequence into blocks, each block containing the keys and values for a fixed number of tokens

Prefill-Decode (PD) Disaggregation

• Mooncake: A KVCache-centric Disaggregated Architecture for LLM Serving (FAST 2025) [Paper] [arXiv] [Slides] [Code]

  • Mootshot AI & Tsinghua

  • Best Paper Award

  • Separate the prefill and decoding clusters; prediction-based early rejection.

  • Distributed multi-layer KVCache pool; prefix-hashed KVCache object storage.

• Inference without Interference: Disaggregate LLM Inference for Mixed Downstream Workloads (arXiv:2401.11181) [arXiv]

  • ICT, CAS & Huawei Cloud

• DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model Serving (OSDI 2024) [Paper] [Code]

  • PKU & UCSD

• Splitwise: Efficient Generative LLM Inference Using Phase Splitting (ISCA 2024) [Paper] [arXiv] [Blog]

  • UW & Microsoft

  • Best Paper Award

  • Split the two phases (i.e., prefill and decode) of a LLM inference request to separate machines

Chunked Prefill

• Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve (OSDI 2024) [Paper] [Code] [arXiv]

  • MSR India & GaTech

  • Sarathi-Serve

Serverless Inference

• λScale: Enabling Fast Scaling for Serverless Large Language Model Inference (arXiv:2502.09922) [arXiv]

  • CUHK-SZ & UVA & HKUST & Alibaba & Nokia Bell Labs

• ServerlessLLM: Low-Latency Serverless Inference for Large Language Models (OSDI 2024) [Paper] [Code] [arXiv]

  • Edinburgh

LoRA Serving

• dLoRA: Dynamically Orchestrating Requests and Adapters for LoRA LLM Serving (OSDI 2024) [Paper]

  • PKU & Shanghai AI Lab

• CaraServe: CPU-Assisted and Rank-Aware LoRA Serving for Generative LLM Inference (arXiv:2401.11240) [arXiv]

  • HKUST & CUHK-Shenzhen & Shanghai AI Lab & Huawei Cloud

• S-LoRA: Serving Thousands of Concurrent LoRA Adapters (MLSys 2024) [arXiv] [Code]

  • UC Berkeley

• Punica: Multi-Tenant LoRA Serving (MLSys 2024) [arXiv] [Code]

  • UW & Duke

Position-Independent Caching (PIC)

• EPIC: Efficient Position-Independent Context Caching for Serving Large Language Models (ICML 2025) [arXiv]

  • PKU & NJU & Huawei Cloud

  • Key insight: the initial tokens of each chunk separately absorb a disproportionate amount of attention, preventing subsequent tokens from attending to relevant parts.

  • Propose an algorithm named LegoLink to recompute k (≤ 32) initial tokens on each chunk (except the first chunk) → Recognize their non-initial status and cripple their attention-absorbing ability.

  • Compared to CacheBlend, LegoLink reduces recomputation complexity and relies on static attention sparsity.

• CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusion (arXiv:2405.16444) [arXiv] [Code]

  • UChicago

  • For an LLM input including multiple text chunks, reuse all KV caches but re-compute a small fraction of KV.

  • Objective: have both the speed of full KV reuse and the generation quality of full KV recompute.

Sparsity

• InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management (OSDI 2024) [Paper]

  • Seoul National University

• PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU (SOSP 2024) [Paper] [arXiv] [Code]

  • SJTU

  • A GPU-CPU hybrid inference engine

  • Hot-activated neurons are preloaded onto the GPU for fast access; cold-activated neurons are computed on the CPU

• Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time (ICML 2023) [Paper] [Code]

  • Rice & ZJU & Stanford & UCSD & ETH & Adobe & Meta AI & CMU

  • A system to predict contextual sparsity (small, input-dependent sets that yield approximately the same output).

Speculative Decoding

• Online Speculative Decoding (ICML 2024) [arXiv]

  • UC Berkeley & UCSD & Sisu Data & SJTU

• SpecInfer: Accelerating Generative LLM Serving with Speculative Inference and Token Tree Verification (ASPLOS 2024) [arXiv] [Code]

  • CMU

• Speculative Decoding with Big Little Decoder (NeurIPS 2023) [Paper]

  • UC Berkeley & ICSI & LBNL

• Fast Inference from Transformers via Speculative Decoding (ICML 2023) [Paper]

  • Google Research

Offloading

• FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU (ICML 2023) [Personal Notes] [Paper] [Code]

  • Stanford & UC Berkeley & ETH & Yandex & HSE & Meta & CMU

  • High-throughput serving; only use a single GPU.

Heterogeneous Environment

• Demystifying Cost-Efficiency in LLM Serving over Heterogeneous GPUs (arXiv:2502.00722) [arXiv]

  • Cambridge & HKUST & PKU & ETH & Purdue

• HexGen-2: Disaggregated Generative Inference of LLMs in Heterogeneous Environment (ICLR 2025) [Paper] [arXiv]

  • HKUST

• HexGen: Generative Inference of Foundation Model over Heterogeneous Decentralized Environment (ICML 2024) [Personal Notes] [arXiv] [Code]

  • HKUST & ETH & CMU

  • Support asymmetric tensor model parallelism and pipeline parallelism under the heterogeneous setting (i.e., each pipeline parallel stage can be assigned with a different number of layers and tensor model parallel degree)

  • Propose a heuristic-based evolutionary algorithm to search for the optimal layout

Fairness

• Locality-aware Fair Scheduling in LLM Serving (arXiv:2501.14312) [arXiv]

  • UC Berkeley

• Fairness in Serving Large Language Models (OSDI 2024) [Paper] [Code]

  • UC Berkeley

LLM Alignment

• PUZZLE: Efficiently Aligning Large Language Models through Light-Weight Context Switch (ATC 2024) [Paper]

  • THU

Acronyms

• LLM: Large Language Model

• LoRA: Low-Rank Adaptation