mlsanityData to 14 Oct 2026

Prompt-SAW: Leveraging Relation-Aware Graphs for Textual Prompt Compression

arXiv 2404.00489 ↗First posted 30 Mar 2024cs.CL cs.AI cs.LGcited by 18

On the map: Efficient LLM inference and serving → Prompt and context compression

Abstract (arXiv)

Large Language Models (LLMs) have shown exceptional abilities for multiple different natural language processing tasks. While prompting is a crucial tool for LLM inference, we observe that there is a significant cost associated with exceedingly lengthy prompts. Existing attempts to compress lengthy prompts lead to substandard results in terms of readability/interpretability of the compressed prompt, with a detrimental impact on prompt utility. To address this, we propose PromptSAW: Prompt compresSion via Relation AWare graphs, an effective strategy for prompt compression over task-agnostic and task-aware prompts. Prompt-SAW uses the prompt's textual information to build a graph and later extracts key information elements in the graph to come up with the compressed prompt. We also propose GSM8K-aug, i.e., an extended version of the existing GSM8K benchmark for task-agnostic prompts in order to provide a comprehensive evaluation platform. Experimental evaluation using benchmark datasets shows that prompts compressed by Prompt-SAW are not only better in terms of readability, but they also outperform the best-performing baseline models by up to 10.1 and 77.1, respectively, for task-agnostic and task-aware settings while compressing the original prompt text by 34.9 and 56.7.

Key facts. Cited by 18 papers in the map (data through 14 Oct 2026). Placed in Prompt and context compression, one of 12 directions in Efficient LLM inference and serving. Builds on 36 of its 50 references that are in the map. First posted on arXiv 30 Mar 2024.

Builds on

36 of 50 references.

Closest ancestors · 3

Same direction, newest first

  1. 2023LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression10 Oct
  2. 2023Discrete Prompt Compression with Reinforcement Learning17 Aug
  3. 2022Prompt Compression and Contrastive Conditioning for Controllability and Toxicity Reduction in Language Models6 Oct

Other references · 33

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  4. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  5. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  6. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,795
  7. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,568
  8. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
  9. 2022Locating and Editing Factual Associations in GPTcited by 2,691
  10. 2022A Survey on In-context Learningcited by 721
  11. 2023Unifying Large Language Models and Knowledge Graphs: A Roadmapcited by 668
  12. 2020A Survey on Knowledge Graphs: Representation, Acquisition and Applicationscited by 643
21 more
  1. 2024A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applicationscited by 431
  2. 2023Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoningcited by 328
  3. 2023LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Modelscited by 287
  4. 2023An LLM can Fool Itself: A Prompt-Based Adversarial Attackcited by 116
  5. 2023Unlocking Context Constraints of LLMs: Enhancing Context Efficiency of LLMs with Self-Information-Based Content Filteringcited by 65
  6. 2024MoRAL: MoE Augmented LoRA for LLMs' Lifelong Learningcited by 63
  7. 2021High Dimensional Differentially Private Stochastic Optimization with Heavy-tailed Datacited by 42
  8. 2023KG-GPT: A General Framework for Reasoning on Knowledge Graphs Using Large Language Modelscited by 32
  9. 2023Understanding In-Context Learning from Repetitionscited by 29
  10. 2023Compress, Then Prompt: Improving Accuracy-Efficiency Trade-off of LLM Inference with Transferable Promptcited by 28
  11. 2022SEAT: Stable and Explainable Attentioncited by 25
  12. 2023Differentially Private Natural Language Models: Recent Advances and Future Directionscited by 20
  13. 2023Practical Differentially Private and Byzantine-resilient Federated Learningcited by 18
  14. 2020Fine-Grained Named Entity Typing over Distantly Supervised Data Based on Refined Representationscited by 12
  15. 2021Faster Rates of Private Stochastic Convex Optimizationcited by 12
  16. 2020OpenIE6: Iterative Grid Labeling and Coordination Analysis for Open Information Extractioncited by 11
  17. 2023Differentially Private Non-convex Learning for Multi-layer Neural Networkscited by 10
  18. 2019Antonym-Synonym Classification Based on New Sub-space Embeddingscited by 9
  19. 2019Estimating Smooth GLM in Non-interactive Local Differential Privacy Model with Public Unlabeled Datacited by 9
  20. 2023GRI: Graph-based Relative Isomorphism of Word Embedding Spacescited by 3
  21. 2023GARI: Graph Attention for Relative Isomorphism of Arabic Word Embeddingscited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 9; 2025: 4; 2026: 5.

18 papers.

Sort:

By area:

  1. 2024Efficient Prompting Methods for Large Language Models: A Surveycited by 47
  2. 2024Merge, Ensemble, and Cooperate! A Survey on Collaborative Strategies in the Era of Large Language Modelscited by 44
  3. 2025A Survey of Multimodal Retrieval-Augmented Generationcited by 37
  4. 2024Editable Concept Bottleneck Modelscited by 23
  5. 2024Characterizing Prompt Compression Methods for Long Context Inferencecited by 18
  6. 2024Leveraging Logical Rules in Knowledge Editing: A Cherry on the Topcited by 12
  7. 2024NTSEBENCH: Cognitive Reasoning Benchmark for Vision Language Modelscited by 9
  8. 2024From Reading to Compressing: Exploring the Multi-document Reader for Prompt Compressioncited by 7
  9. 2024Squid: Long Context as a New Modality for Energy-Efficient On-Device Language Modelscited by 6
  10. 2025SCOPE: A Generative Approach for LLM Prompt Compressioncited by 6
  11. 2024The Compressor-Retriever Architecture for Language Model OScited by 5
  12. 2025CompLLM: Compression for Long Context Q&Acited by 5
6 more
  1. 2025Thoughtbubbles: an Unsupervised Method for Parallel Thinking in Latent Spacecited by 4
  2. 2026Controllable Concept Bottleneck Modelscited by 2
  3. 2026Mapping Text to Multiplex Graph: Prompt Compression as L'evy Walk-Guided Graph Pruningcited by 1
  4. 2026Backdoor Learning in Language Models and Vision-Language Modelscited by 0
  5. 2026Prompt Minimization: Reducing Input Redundancy Without Sacrificing Output Fidelitycited by 0
  6. 2026Learning from Research: Toward Lifelong Agent Harness Evolutioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Fundamental Limits of Prompt Compression: A Rate-Distortion Framework for Black-Box Language Models
  2. 2025Understanding and Improving Information Preservation in Prompt Compression for LLMs
  3. 2024LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression
  4. 2024Style-Compress: An LLM-Based Prompt Compression Framework Considering Task-Specific Styles
  5. 2024Prompt Compression for Large Language Models: A Survey
  6. 2026Lexical Prompt Compression for Large Language Models: A Training-Free, Deterministic Pipeline with Empirical Pareto Analysis Across Eleven Task Categories
  7. 2025Task-agnostic Prompt Compression with Context-aware Sentence Embedding and Reward-guided Task Descriptor
  8. 2025Dynamic Compressing Prompts for Efficient Inference of Large Language Models