Efficient Prompt Compression with Evaluator Heads for Long-Context Transformer Inference

• URL: http://arxiv.org/abs/2501.12959v2
• Date: Wed, 05 Feb 2025 09:35:38 GMT
• Title: Efficient Prompt Compression with Evaluator Heads for Long-Context Transformer Inference
• Authors: Weizhi Fei, Xueyan Niu, Guoqing Xie, Yingqing Liu, Bo Bai, Wei Han,
• Abstract summary: Long-context inputs are crucial for the effective utilization of large language models (LLMs)<n>We propose an efficient, training-free prompt compression method that retains key information within compressed prompts.<n>We develop Evaluator Head-based Prompt Compression, which enables LLMs to rapidly "skim through" input prompts.
• Score: 6.699062502635993
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Although applications involving long-context inputs are crucial for the effective utilization of large language models (LLMs), they also result in increased computational costs and reduced performance. To address this challenge, we propose an efficient, training-free prompt compression method that retains key information within compressed prompts. We identify specific attention heads in transformer-based LLMs, which we designate as evaluator heads, that are capable of selecting tokens in long inputs that are most significant for inference. Building on this discovery, we develop EHPC, an Evaluator Head-based Prompt Compression method, which enables LLMs to rapidly "skim through" input prompts by leveraging only the first few layers with evaluator heads during the pre-filling stage, subsequently passing only the important tokens to the model for inference. EHPC achieves state-of-the-art results across two mainstream benchmarks: prompt compression and long-context inference acceleration. Consequently, it effectively reduces the complexity and costs associated with commercial API calls. We further demonstrate that EHPC attains competitive results compared to key-value cache-based acceleration methods, thereby highlighting its potential to enhance the efficiency of LLMs for long-context tasks.

Related papers

• PIS: Linking Importance Sampling and Attention Mechanisms for Efficient Prompt Compression [3.6268731121741067]
  Large language models (LLMs) have achieved remarkable progress, demonstrating unprecedented capabilities across various natural language processing tasks. Existing prompt compression methods rely on truncation or abstractive summarization techniques. We introduce Prompt Importance Sampling (PIS), a novel compression framework that dynamically compresses prompts by sampling important tokens.
  arXiv  Detail & Related papers  (2025-04-23T09:53:01Z)
• PromptDistill: Query-based Selective Token Retention in Intermediate Layers for Efficient Large Language Model Inference [23.828853446344663]
  We propose PromptDistill, a training-free method that improves inference efficiency while preserving generation quality. PromptDistill identifies and retains the most informative tokens by leveraging attention interactions in early layers, preserving their hidden states while reducing the computational burden in later layers.
  arXiv  Detail & Related papers  (2025-03-30T01:47:23Z)
• Understanding and Improving Information Preservation in Prompt Compression for LLMs [10.912320980464571]
  In information-intensive tasks, the prompt length can grow fast, leading to increased computational requirements, performance degradation, and induced biases from irrelevant or redundant information. We propose a holistic evaluation framework that allows for in-depth analysis of prompt compression methods.
  arXiv  Detail & Related papers  (2025-03-24T20:06:11Z)
• Task-Oriented Feature Compression for Multimodal Understanding via Device-Edge Co-Inference [49.77734021302196]
  We propose a task-oriented feature compression (TOFC) method for multimodal understanding in a device-edge co-inference framework. To enhance compression efficiency, multiple entropy models are adaptively selected based on the characteristics of the visual features. Results show that TOFC achieves up to 60% reduction in data transmission overhead and 50% reduction in system latency.
  arXiv  Detail & Related papers  (2025-03-17T08:37:22Z)
• Achieving More with Less: Additive Prompt Tuning for Rehearsal-Free Class-Incremental Learning [76.32953653161417]
  Class-incremental learning enables models to learn new classes progressively while preserving knowledge of previously learned ones. Recent advances in this field have shifted towards parameter-efficient fine-tuning techniques. We present a novel prompt-based approach that addresses the limitation of current approaches.
  arXiv  Detail & Related papers  (2025-03-11T02:27:37Z)
• Efficient and Effective Prompt Tuning via Prompt Decomposition and Compressed Outer Product [8.014705094248589]
  Low- parameters prompt tuning method outperforms state-of-the-art PT-based and LoRA-based methods in performance and efficiency. Experiments across six architectures and eight datasets demonstrate that LAMP outperforms state-of-the-art PT-based and LoRA-based methods in performance and efficiency.
  arXiv  Detail & Related papers  (2025-02-16T05:50:12Z)
• Squeezed Attention: Accelerating Long Context Length LLM Inference [64.11145320159126]
  We propose Squeezed Attention as a mechanism to accelerate LLM applications where a large portion of the input prompt is fixed. We use K-means clustering offline to group the keys for the fixed context based on semantic similarity and represent each cluster with a single centroid value. We then compute exact attention using only these important keys from the fixed context, thereby reducing bandwidth and computational costs.
  arXiv  Detail & Related papers  (2024-11-14T18:54:19Z)
• Perception Compressor:A training-free prompt compression method in long context scenarios [17.720102137585503]
  Perception is a training-free prompt compression method for large language models. It outperforms existing methods by a large margin, achieving state-of-the-art performance.
  arXiv  Detail & Related papers  (2024-09-28T07:13:33Z)
• QUITO-X: A New Perspective on Context Compression from the Information Bottleneck Theory [66.01597794579568]
  We introduce information bottleneck theory (IB) to model the problem.<n>We propose a cross-attention-based approach to approximate mutual information in IB.<n>Our method achieves a 25% increase in compression rate compared to the state-of-the-art.
  arXiv  Detail & Related papers  (2024-08-20T02:44:45Z)
• FactorLLM: Factorizing Knowledge via Mixture of Experts for Large Language Models [50.331708897857574]
  We introduce FactorLLM, a novel approach that decomposes well-trained dense FFNs into sparse sub-networks without requiring any further modifications. FactorLLM achieves comparable performance to the source model securing up to 85% model performance while obtaining over a 30% increase in inference speed.
  arXiv  Detail & Related papers  (2024-08-15T16:45:16Z)
• A Thorough Performance Benchmarking on Lightweight Embedding-based Recommender Systems [67.52782366565658]
  State-of-the-art recommender systems (RSs) depend on categorical features, which ecoded by embedding vectors, resulting in excessively large embedding tables. Despite the prosperity of lightweight embedding-based RSs, a wide diversity is seen in evaluation protocols. This study investigates various LERS' performance, efficiency, and cross-task transferability via a thorough benchmarking process.
  arXiv  Detail & Related papers  (2024-06-25T07:45:00Z)
• Anchor-based Large Language Models [33.86392289481657]
  This study introduces Anchor-based LLMs (AnLLMs), which utilize an anchor-based self-attention network (AnSAN) and also an anchor-based inference strategy. AnLLMs maintain similar accuracy levels while achieving up to 99% keys/values cache reduction and up to 3.5 times faster inference.
  arXiv  Detail & Related papers  (2024-02-12T12:48:02Z)
• Accelerating LLaMA Inference by Enabling Intermediate Layer Decoding via Instruction Tuning with LITE [62.13435256279566]
  Large Language Models (LLMs) have achieved remarkable performance across a wide variety of natural language tasks. However, their large size makes their inference slow and computationally expensive. We show that it enables these layers to acquire 'good' generation ability without affecting the generation ability of the final layer.
  arXiv  Detail & Related papers  (2023-10-28T04:07:58Z)
• Do Compressed LLMs Forget Knowledge? An Experimental Study with Practical Implications [63.29358103217275]
  Large Language Models (LLMs) often leads to reduced performance, especially for knowledge-intensive tasks. We propose two conjectures on the nature of the damage: one is certain knowledge being forgotten (or erased) after compression. We introduce a variant called Inference-time Dynamic Prompting (IDP) that can effectively increase prompt diversity without incurring any inference overhead.
  arXiv  Detail & Related papers  (2023-10-02T03:12:06Z)
• Approximated Prompt Tuning for Vision-Language Pre-trained Models [54.326232586461614]
  In vision-language pre-trained models, prompt tuning often requires a large number of learnable tokens to bridge the gap between the pre-training and downstream tasks. We propose a novel Approximated Prompt Tuning (APT) approach towards efficient VL transfer learning.
  arXiv  Detail & Related papers  (2023-06-27T05:43:47Z)
• Compress, Then Prompt: Improving Accuracy-Efficiency Trade-off of LLM Inference with Transferable Prompt [96.24800696597707]
  We introduce a new perspective to optimize this trade-off by prompting compressed models. We propose a soft prompt learning method where we expose the compressed model to the prompt learning process. Our experimental analysis suggests our soft prompt strategy greatly improves the performance of the 8x compressed LLaMA-7B model.
  arXiv  Detail & Related papers  (2023-05-17T20:45:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.