LLavaCode: Compressed Code Representations for Retrieval-Augmented Code Generation

• URL: http://arxiv.org/abs/2510.19644v1
• Date: Wed, 22 Oct 2025 14:49:21 GMT
• Title: LLavaCode: Compressed Code Representations for Retrieval-Augmented Code Generation
• Authors: Daria Cherniuk, Nikita Sukhorukov, Nikita Sushko, Daniil Gusak, Danil Sivtsov, Elena Tutubalina, Evgeny Frolov,
• Abstract summary: We introduce LlavaCode, a framework that compresses code into compact, semantically rich representations interpretable by code LLM.<n>Our experiments demonstrate that compressed context enables 20-38% reduction in Time-to-First-Token (TTFT) on line completion tasks.
• Score: 8.868449925993994
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Retrieval-augmented generation has emerged as one of the most effective approaches for code completion, particularly when context from a surrounding repository is essential. However, incorporating context significantly extends sequence length, leading to slower inference - a critical limitation for interactive settings such as IDEs. In this work, we introduce LlavaCode, a framework that compresses code into compact, semantically rich representations interpretable by code LLM, enhancing generation quality while reducing the retrieved context to only a few compressed single-token vectors. Using a small projector module we can significantly increase the EM and ES metrics of coding model with negligible latency increase. Our experiments demonstrate that compressed context enables 20-38% reduction in Time-to-First-Token (TTFT) on line completion tasks compared to full-RAG pipelines.

Related papers

• LongCodeZip: Compress Long Context for Code Language Models [16.940525379087326]
  LongCodeZip is a novel plug-and-play code compression framework designed specifically for Large Language Models (LLMs)<n>By effectively reducing context size while preserving essential information, LongCodeZip enables LLMs to better scale to real-world, large-scale code scenarios.
  arXiv  Detail & Related papers  (2025-10-01T02:54:57Z)
• REFRAG: Rethinking RAG based Decoding [67.4862300145604]
  REFRAG is an efficient decoding framework that compresses, senses, and expands to improve latency in RAG applications.<n>We provide rigorous validation of REFRAG across diverse long-context tasks, including RAG, multi-turn conversations, and long document summarization.
  arXiv  Detail & Related papers  (2025-09-01T03:31:44Z)
• Is Compression Really Linear with Code Intelligence? [60.123628177110206]
  textitFormat Annealing is a lightweight, transparent training methodology designed to assess the intrinsic capabilities of pre-trained models equitably.<n>Our empirical results reveal a fundamental logarithmic relationship between measured code intelligence and bits-per-character (BPC)<n>Our work provides a more nuanced understanding of compression's role in developing code intelligence and contributes a robust evaluation framework in the code domain.
  arXiv  Detail & Related papers  (2025-05-16T16:59:14Z)
• CodeRAG: Supportive Code Retrieval on Bigraph for Real-World Code Generation [69.684886175768]
  Large language models (LLMs) have shown promising performance in automated code generation.<n>In this paper, we propose CodeRAG, a retrieval-augmented code generation framework.<n> Experiments show that CodeRAG achieves significant improvements compared to no RAG scenarios.
  arXiv  Detail & Related papers  (2025-04-14T09:51:23Z)
• LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and Verification [42.54363549922909]
  LongSpec is a framework that addresses the challenges of efficient inference over long contexts.<n>LongSpec achieves up to a 3.26x speedup over strong Flash Attention baselines.<n>The code is available at https://github.com/sail-sg/LongSpec.
  arXiv  Detail & Related papers  (2025-02-24T18:53:31Z)
• FastCoder: Accelerating Repository-level Code Generation via Efficient Retrieval and Verification [10.286072352686874]
  We propose FastCoder, an inference acceleration approach specifically designed for code generation.<n>FastCoder constructs a multi-source datastore, providing access to both general and project-specific knowledge.<n>It can reach up to 2.53x and 2.54x speedup compared to autoregressive decoding in repository-level and standalone code generation tasks.
  arXiv  Detail & Related papers  (2025-02-24T13:30:30Z)
• CODEPROMPTZIP: Code-specific Prompt Compression for Retrieval-Augmented Generation in Coding Tasks with LMs [6.936336826531964]
  Retrieval-Augmented Generation (RAG) enhances coding tasks by incorporating retrieved code examples into prompts.<n>Existing prompt compression techniques focus on natural language, lacking tailored solutions for code.<n>We propose CodePromptZip, a framework that compresses code examples before integrating into RAG.
  arXiv  Detail & Related papers  (2025-02-19T23:15:23Z)
• Hierarchical Context Pruning: Optimizing Real-World Code Completion with Repository-Level Pretrained Code LLMs [24.00351065427465]
  We propose a strategy named Hierarchical Context Pruning (HCP) to construct completion prompts with high informational code content. The HCP models the code repository at the function level, maintaining the topological dependencies between code files while removing a large amount of irrelevant code content.
  arXiv  Detail & Related papers  (2024-06-26T12:26:16Z)
• REPOFUSE: Repository-Level Code Completion with Fused Dual Context [11.531678717514724]
  This paper introduces REPOFUSE, a pioneering solution designed to enhance repository-level code completion without the latency trade-off. We propose a novel rank truncated generation (RTG) technique that efficiently condenses two types of context into prompts with restricted size. REPOFUSE has demonstrated a significant leap over existing models, achieving a 40.90% to 59.75% increase in exact match (EM) accuracy for code completions and a 26.8% enhancement in inference speed.
  arXiv  Detail & Related papers  (2024-02-22T06:34:50Z)
• StepCoder: Improve Code Generation with Reinforcement Learning from Compiler Feedback [58.20547418182074]
  We introduce StepCoder, a novel framework for code generation, consisting of two main components. CCCS addresses the exploration challenge by breaking the long sequences code generation task into a Curriculum of Code Completion Subtasks. FGO only optimize the model by masking the unexecuted code segments to provide Fine-Grained Optimization. Our method improves the ability to explore the output space and outperforms state-of-the-art approaches in corresponding benchmarks.
  arXiv  Detail & Related papers  (2024-02-02T13:14:31Z)
• Fast Chain-of-Thought: A Glance of Future from Parallel Decoding Leads to Answers Faster [61.83949316226113]
  FastCoT is a model-agnostic framework based on parallel decoding. We show that FastCoT saves inference time by nearly 20% with only a negligible performance drop compared to the regular approach.
  arXiv  Detail & Related papers  (2023-11-14T15:56:18Z)
• CodeT5+: Open Code Large Language Models for Code Understanding and Generation [72.1638273937025]
  Large language models (LLMs) pretrained on vast source code have achieved prominent progress in code intelligence. CodeT5+ is a family of encoder-decoder LLMs for code in which component modules can be flexibly combined to suit a wide range of downstream code tasks. We extensively evaluate CodeT5+ on over 20 code-related benchmarks in different settings, including zero-shot, finetuning, and instruction-tuning.
  arXiv  Detail & Related papers  (2023-05-13T14:23:07Z)

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.