# Less is More: Lightweight Prompt Compression for Question Answering Applications on Edge Devices

arXiv: https://arxiv.org/abs/2606.20571 · HTML: https://mlsanity.com/paper/2606.20571/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Generation and context integration, one of 15 directions in Retrieval-augmented generation.
- Builds on 22 of its 83 references that are in the map.
- First posted on arXiv 27 Apr 2026.

## Placed in

- Retrieval-augmented generation / [Generation and context integration](https://mlsanity.com/rag/generation-and-context-integration/)
- Efficient LLM inference and serving / [Prompt and context compression](https://mlsanity.com/inference/prompt-and-context-compression/)

## Builds on (most cited first, 22 of 22)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Multilingual Denoising Pre-training for Neural Machine Translation](https://mlsanity.com/paper/2001.08210/) (2020)
- [A Dataset of Information-Seeking Questions and Answers Anchored in Research Papers](https://mlsanity.com/paper/2105.03011/) (2021)
- [Adapting Language Models to Compress Contexts](https://mlsanity.com/paper/2305.14788/) (2023)
- [Retrieval-Augmented Generation for AI-Generated Content: A Survey](https://mlsanity.com/paper/2402.19473/) (2024)
- [HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models](https://mlsanity.com/paper/2405.14831/) (2024)
- [SpargeAttention: Accurate and Training-free Sparse Attention Accelerating Any Model Inference](https://mlsanity.com/paper/2502.18137/) (2025)
- [A Comprehensive Survey on Long Context Language Modeling](https://mlsanity.com/paper/2503.17407/) (2025)
- [Prompt Compression for Large Language Models: A Survey](https://mlsanity.com/paper/2410.12388/) (2024)
- [Unlocking Context Constraints of LLMs: Enhancing Context Efficiency of LLMs with Self-Information-Based Content Filtering](https://mlsanity.com/paper/2304.12102/) (2023)
- [Fast On-device LLM Inference with NPUs](https://mlsanity.com/paper/2407.05858/) (2024)
- [Injecting Domain-Specific Knowledge into Large Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2502.10708/) (2025)
- [The What, Why, and How of Context Length Extension Techniques in Large Language Models -- A Detailed Survey](https://mlsanity.com/paper/2401.07872/) (2024)
- [Llama-3.1-FoundationAI-SecurityLLM-Base-8B Technical Report](https://mlsanity.com/paper/2504.21039/) (2025)
- [Extending LLMs' Context Window with 100 Samples](https://mlsanity.com/paper/2401.07004/) (2024)
- [Enhancing Large Language Model Performance To Answer Questions and Extract Information More Accurately](https://mlsanity.com/paper/2402.01722/) (2024)
- [Recent Advances in Named Entity Recognition: A Comprehensive Survey and Comparative Study](https://mlsanity.com/paper/2401.10825/) (2024)
- [Efficient Prompt Compression with Evaluator Heads for Long-Context Transformer Inference](https://mlsanity.com/paper/2501.12959/) (2025)

## Abstract (arXiv)

In agent-driven question answering (QA) applications, retrieval-augmented generation (RAG) is commonly introduced to enhance the response accuracy of large language models (LLMs) by providing additional context. Due to the inherent noise in retrieval results and the coarse granularity of document-level retrieval, the retrieved context often contains substantial redundant information. In this setting, the agent prompt, consisting of the user query and the associated retrieved context, leads to unnecessary computational overhead during LLM inference. Existing prompt compression methods typically rely on auxiliary small language models (SLMs) to estimate context importance. However, such approaches introduce significant memory and computational overhead, which limits their deployment on resource-constrained edge devices. In this paper, we propose CORE, a two-stage sentence-level prompt compression method that eliminates the need for SLMs. In the first stage, CORE constructs an answer set via named entity recognition (NER) and a clue set via semantic matching. In the second stage, CORE refines the clue set using an orthogonal residual retrieval strategy and designs a spatial proximity-based metric to filter the answer set. The two sets are then combined to form the final compressed context. We implement CORE on an NVIDIA Jetson AGX Orin edge device and a Huawei Nova smartphone. Experimental results demonstrate that within a 2000-token budget, CORE improves accuracy by at least 30.19% compared to state-of-the-art baselines, while reducing memory usage by at least 50.47% and achieving at least 1.94 times speedup on the edge device. Moreover, compared to the state-of-the-art LLMLingua2 method, CORE achieves a substantial energy reduction of 95.74% on the smartphone, highlighting its practicality and generalizability for mobile deployments.