LLMSys-PaperList
Curated list of academic papers related to Large Language Model systems
GraphCanon updated 2w · GitHub synced 2w
Decision brief
LLMSys-PaperList offers a comprehensive list of papers and resources tailored specifically to Large Language Model (LLM) systems.
Good fit when
- - When you need a curated list focusing on technical advancements in pre-training, post-training, serving, and multi-modal LLM systems.
- - If your interest lies specifically in recent developments and cutting-edge research by leading industry players like Google, ByteDance, and Sea AI Lab, from conferences like SOSP' 24 and NSDI' 24.
Avoid when
- - If you are looking for a general repository of machine learning papers rather than specific developments related to Large Language Models.
- - When your primary need is documentation or code examples rather than academic papers and project insights.
- Hosting:
- unknown - (repository does not specify hosting environment)
Observed Jul 11, 2026 · Source: enrich:decision_facts
Verify the decision
Maintenance and security
Full trust report- Maintenance
- Active (12d since push)
- As of 2w
- Provenance
- Not a fork · Personal account
- As of 2w
- Security (OSV)
- No lockfile
- As of 1mo
Public GitHub metadata and optional OSV scans. Signals, not a guarantee. Trust methodology.
Install
pip install LLMSys-PaperListSimilar tools
Same-category neighbours. No typed graph edges are catalogued for this tool yet.
Evidence and technical details
Sourced facts, taxonomy, compatibility claims, README excerpt, and machine-readable endpoints.
Overview
A repository containing a curated collection of academic papers, articles, tutorials, and projects related to Large Language Model (LLM) systems in categories such as training, serving, multi-modal systems, LLM frameworks, ML conferences, survey papers, benchmarks, and more.
Capability facts
- Languages
- python
Source: github.language · Aug 6, 2026
Categories
Tags
README
Awesome LLM Systems Papers
A curated list of Large Language Model systems related academic papers, articles, tutorials, slides and projects. Star this repository, and then you can keep abreast of the latest developments of this booming research field.
Trends at a Glance (2024 → 2026)
Three eras, one lens: what unit of work the system optimizes — a request (2024), a session or reasoning trace (2025), a whole agent trajectory (2026).
Serving is still the largest area in absolute terms, but its share of the list fell from 49% to 33% — the growth went to kernel/model co-design (3 → 29 papers), agentic systems (4 → 22), AI-for-systems (5 → 16) and edge (2 → 14).
The fastest-rising techniques by share of the year's papers: agentic / multi-agent (1.0% → 13.1%), compiler / kernel / megakernel (0.0% → 10.2%), speculative decoding (1.0% → 5.7%), sparse attention (1.0% → 4.5%) and energy / power (1.0% → 3.3%).
→ Full analysis, per-technique numbers and reproduction scripts: trends/
Table of Contents
- Trends at a Glance
- LLM Systems
- Training
- Pre-training
- Post Training
- Fault Tolerance / Straggler Mitigation
- Serving
- LLM serving
- Agent Systems
- Serving at the edge
- System Efficiency Optimization - Model Co-design
- Multi-Modal Training Systems
- Multi-Modal Serving Systems
- Training
- LLM for Systems
- Industrial LLM Technical Report
- ML Conferences
- NeurIPS 2025
- LLM Frameworks
- Training
- Post-Training
- Serving
- ML Systems
- Survey Paper
- LLM Benchmark / Leaderboard / Traces
- Related ML Readings
- MLSys Courses
- Other Reading
LLM Systems
Training
Pre-training
Before 2024
- Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
- Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM
- Reducing Activation Recomputation in Large Transformer Models
- Optimized Network Architectures for Large Language Model Training with Billions of Parameters | MIT
- Carbon Emissions and Large Neural Network Training | Google, UCB
2024
- Perseus: Removing Energy Bloat from Large Model Training | SOSP' 24
- MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs | ByteDance
- DISTMM: Accelerating distributed multimodal model training | NSDI' 24
- Pipeline Parallelism with Controllable Memory | Sea AI Lab
- Boosting Large-scale Parallel Training Efficiency with C4: A Communication-Driven Approach
- Scaling Beyond the GPU Memory Limit for Large Mixture-of-Experts Model Training | ICML' 24
- Alibaba HPN: A Data Center Network for Large Language ModelTraining
- [The Llama 3 Herd o
For agents
This page has a .md twin and JSON over the API.