Long-Context Inference with Retrieval-Augmented Speculative Decoding

• URL: http://arxiv.org/abs/2502.20330v1
• Date: Thu, 27 Feb 2025 17:59:36 GMT
• Title: Long-Context Inference with Retrieval-Augmented Speculative Decoding
• Authors: Guanzheng Chen, Qilong Feng, Jinjie Ni, Xin Li, Michael Qizhe Shieh,
• Abstract summary: Long-context large language models (LLMs) offer a promising alternative to traditional retrieval-augmented generation (RAG)<n>We present Retrieval-Augmented Speculative Decoding (RAPID), which leverages RAG for both accelerating and enhancing generation quality.<n>Our approach enables a new paradigm where same-scale or even larger LLMs can serve as RAG drafters while maintaining computational efficiency.
• Score: 7.785459677641105
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The emergence of long-context large language models (LLMs) offers a promising alternative to traditional retrieval-augmented generation (RAG) for processing extensive documents. However, the computational overhead of long-context inference, particularly in managing key-value (KV) caches, presents significant efficiency challenges. While Speculative Decoding (SD) traditionally accelerates inference using smaller draft models, its effectiveness diminishes substantially in long-context scenarios due to memory-bound KV cache operations. We present Retrieval-Augmented Speculative Decoding (RAPID), which leverages RAG for both accelerating and enhancing generation quality in long-context inference. RAPID introduces the RAG drafter-a draft LLM operating on shortened retrieval contexts-to speculate on the generation of long-context target LLMs. Our approach enables a new paradigm where same-scale or even larger LLMs can serve as RAG drafters while maintaining computational efficiency. To fully leverage the potentially superior capabilities from stronger RAG drafters, we develop an inference-time knowledge transfer dynamic that enriches the target distribution by RAG. Extensive experiments on the LLaMA-3.1 and Qwen2.5 backbones demonstrate that RAPID effectively integrates the strengths of both approaches, achieving significant performance improvements (e.g., from 39.33 to 42.83 on InfiniteBench for LLaMA-3.1-8B) with more than 2x speedups. Our analyses reveal that RAPID achieves robust acceleration beyond 32K context length and demonstrates superior generation quality in real-world applications.

Related papers

• RAGO: Systematic Performance Optimization for Retrieval-Augmented Generation Serving [9.962031642362813]
  Retrieval-augmented generation (RAG) is emerging as a popular approach for reliable LLM serving. RAG is a structured abstraction that captures the wide range of RAG algorithms. RAGO is a system optimization framework for efficient RAG serving.
  arXiv  Detail & Related papers  (2025-03-18T18:58:13Z)
• PowerAttention: Exponentially Scaling of Receptive Fields for Effective Sparse Attention [73.26995918610669]
  Large Language Models (LLMs) face efficiency bottlenecks due to the quadratic complexity of the attention mechanism when processing long contexts. We introduce PowerAttention, a novel sparse attention design that facilitates effective and complete context extension. Experiments demonstrate that PowerAttention outperforms existing static sparse attention methods by $5sim 40%$.
  arXiv  Detail & Related papers  (2025-03-05T15:24:11Z)
• LongSpec: Long-Context Speculative Decoding with Efficient Drafting and Verification [42.54363549922909]
  Speculative decoding has become a promising technique to mitigate the high inference latency of autoregressive decoding in Large Language Models.<n>Despite its promise, the effective application of speculative decoding in LLMs still confronts three key challenges.<n>We enhance the performance of speculative decoding in long-context settings by addressing these challenges.
  arXiv  Detail & Related papers  (2025-02-24T18:53:31Z)
• Does RAG Really Perform Bad For Long-Context Processing? [15.889864680212147]
  RetroLM is a novel framework for long-context processing.<n>Unlike traditional methods, RetroLM employs KV-level retrieval augmentation.<n>Building on this framework, we develop a specialized retriever for precise retrieval of critical pages.
  arXiv  Detail & Related papers  (2025-02-17T05:02:25Z)
• RoseRAG: Robust Retrieval-augmented Generation with Small-scale LLMs via Margin-aware Preference Optimization [53.63439735067081]
  Large language models (LLMs) have achieved impressive performance but face high computational costs and latency. Retrieval-augmented generation (RAG) helps by integrating external knowledge, but imperfect retrieval can introduce distracting noise that misleads SLMs. We propose RoseRAG, a robust RAG framework for SLMs via Margin-aware Preference Optimization.
  arXiv  Detail & Related papers  (2025-02-16T04:56:53Z)
• LaRA: Benchmarking Retrieval-Augmented Generation and Long-Context LLMs - No Silver Bullet for LC or RAG Routing [70.35888047551643]
  We present LaRA, a novel benchmark specifically designed to rigorously compare RAG and LC LLMs.<n>LaRA encompasses 2,326 test cases across four practical QA task categories and three types of naturally occurring long texts.<n>We find that the optimal choice between RAG and LC depends on a complex interplay of factors, including the model's parameter size, long-text capabilities, context length, task type, and the characteristics of the retrieved chunks.
  arXiv  Detail & Related papers  (2025-02-14T08:04:22Z)
• Reward-Guided Speculative Decoding for Efficient LLM Reasoning [80.55186052123196]
  We introduce Reward-Guided Speculative Decoding (RSD), a novel framework aimed at improving the efficiency of inference in large language models (LLMs)<n>RSD incorporates a controlled bias to prioritize high-reward outputs, in contrast to existing speculative decoding methods that enforce strict unbiasedness.<n>RSD delivers significant efficiency gains against decoding with the target model only, while achieving significant better accuracy than parallel decoding method on average.
  arXiv  Detail & Related papers  (2025-01-31T17:19:57Z)
• Don't Do RAG: When Cache-Augmented Generation is All You Need for Knowledge Tasks [11.053340674721005]
  Retrieval-augmented generation (RAG) has gained traction as a powerful approach for enhancing language models by integrating external knowledge sources.<n>This paper proposes an alternative paradigm, cache-augmented generation (CAG) that bypasses real-time retrieval.
  arXiv  Detail & Related papers  (2024-12-20T06:58:32Z)
• ChatQA 2: Bridging the Gap to Proprietary LLMs in Long Context and RAG Capabilities [53.97515452727115]
  ChatQA 2 is a Llama 3.0-based model with a 128K context window.<n>We present a training recipe to extend the context window of Llama3-70B-base from 8K to 128K tokens.<n>We find that the performance of strong long-context LLMs using RAG improves when retrieving a larger number of chunks.
  arXiv  Detail & Related papers  (2024-07-19T17:35:47Z)
• Speculative RAG: Enhancing Retrieval Augmented Generation through Drafting [68.90949377014742]
  Speculative RAG is a framework that leverages a larger generalist LM to efficiently verify multiple RAG drafts produced in parallel by a smaller, distilled specialist LM. Our method accelerates RAG by delegating drafting to the smaller specialist LM, with the larger generalist LM performing a single verification pass over the drafts. It notably enhances accuracy by up to 12.97% while reducing latency by 51% compared to conventional RAG systems on PubHealth.
  arXiv  Detail & Related papers  (2024-07-11T06:50:19Z)
• 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)

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.