GMSA: Enhancing Context Compression via Group Merging and Layer Semantic Alignment

• URL: http://arxiv.org/abs/2505.12215v1
• Date: Sun, 18 May 2025 03:21:30 GMT
• Title: GMSA: Enhancing Context Compression via Group Merging and Layer Semantic Alignment
• Abstract summary: This paper introduces GMSA, a context compression framework based on the encoder-decoder architecture.<n>GMSA reduces input sequence length and redundant information.<n>It can achieve approximately a 2x speedup in end-to-end inference.
• Score: 18.256369876037883
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models (LLMs) have achieved impressive performance in a variety of natural language processing (NLP) tasks. However, when applied to long-context scenarios, they face two challenges, i.e., low computational efficiency and much redundant information. This paper introduces GMSA, a context compression framework based on the encoder-decoder architecture, which addresses these challenges by reducing input sequence length and redundant information. Structurally, GMSA has two key components: Group Merging and Layer Semantic Alignment (LSA). Group merging is used to effectively and efficiently extract summary vectors from the original context. Layer semantic alignment, on the other hand, aligns the high-level summary vectors with the low-level primary input semantics, thus bridging the semantic gap between different layers. In the training process, GMSA first learns soft tokens that contain complete semantics through autoencoder training. To furtherly adapt GMSA to downstream tasks, we propose Knowledge Extraction Fine-tuning (KEFT) to extract knowledge from the soft tokens for downstream tasks. We train GMSA by randomly sampling the compression rate for each sample in the dataset. Under this condition, GMSA not only significantly outperforms the traditional compression paradigm in context restoration but also achieves stable and significantly faster convergence with only a few encoder layers. In downstream question-answering (QA) tasks, GMSA can achieve approximately a 2x speedup in end-to-end inference while outperforming both the original input prompts and various state-of-the-art (SOTA) methods by a large margin.

Related papers

• No More K-means: Single-Stage Sparse Coding for Efficient Multi-Vector Retrieval [51.43543998583709]
  Single-stage Sparse Retrieval (SSR) is a paradigm shift that replaces expensive clustering with efficient sparse coding.<n>It reduces indexing time by 15x compared to ColBERTv2, halves retrieval latency, and simultaneously improves retrieval performance.
  arXiv  Detail & Related papers  (2026-05-28T15:53:34Z)
• From I/O to Code with Discovery Agent [103.88427301265669]
  We propose DIO-Agent, a discovery agent for IO2Code.<n>Our method frames IO2Code as an evolutionary search over discrete program space.<n>Extensive experiments show that DIO-Agent consistently outperforms both traditional program-by-example method and SOTA evolution-agent baselines.
  arXiv  Detail & Related papers  (2026-05-14T18:57:32Z)
• Context Pruning for Coding Agents via Multi-Rubric Latent Reasoning [14.508049757681862]
  LLM-powered coding agents spend the majority of their token budget reading repository files.<n>We show that this formulation creates a modeling bottleneck.<n>We propose LaMR, a structured pruning framework that decomposes code relevance into two interpretable quality dimensions.
  arXiv  Detail & Related papers  (2026-05-14T18:30:10Z)
• Forget, Then Recall: Learnable Compression and Selective Unfolding via Gist Sparse Attention [7.694795562178033]
  Scaling large language models to long contexts is challenging due to the quadratic computational cost of full attention.<n>We introduce selective unfolding via GSA, which first compresses the context into gist tokens, then selects the most relevant gists.<n>This yields a simple coarse-to-fine mechanism that combines compact global representations with targeted access to fine-leaved evidence.
  arXiv  Detail & Related papers  (2026-04-22T04:22:32Z)
• Context Compression via Explicit Information Transmission [25.078241611630585]
  Long-context inference with Large Language Models (LLMs) is costly due to quadratic attention and growing key-value caches.<n>We propose ComprExIT, a lightweight framework that formulates soft compression into a new paradigm.
  arXiv  Detail & Related papers  (2026-02-03T17:44:12Z)
• COMI: Coarse-to-fine Context Compression via Marginal Information Gain [28.75695496228666]
  Large Language Models (LLMs) have demonstrated exceptional capabilities across diverse tasks.<n>Context compression methods address these challenges by significantly reducing input length and eliminating redundancy.<n>We propose COMI, a coarse-to-fine adaptive context compression framework.
  arXiv  Detail & Related papers  (2026-02-02T06:57:22Z)
• Autoencoding-Free Context Compression for LLMs via Contextual Semantic Anchors [43.02557489472655]
  Current context compression methods rely on autoencoding tasks to train context-agnostic compression tokens to compress contextual semantics.<n>We propose Semantic-Anchor Compression (SAC), a novel method that shifts from autoencoding task based compression to an architecture that is equipped with this compression capability.<n>SAC consistently outperforms existing context compression methods across various compression ratios.
  arXiv  Detail & Related papers  (2025-10-10T01:42:14Z)
• UniGist: Towards General and Hardware-aligned Sequence-level Long Context Compression [86.33995240043936]
  UniGist is a sequence-level long-context compression framework for large language models.<n>It efficiently preserves context information by replacing raw tokens with special compression tokens (gists) in a fine-grained manner.<n>Our scheme also supports flexible inference by allowing the actual removal of compressed tokens, resulting in real-time memory savings.
  arXiv  Detail & Related papers  (2025-09-19T08:47:37Z)
• Low-Complexity Semantic Packet Aggregation for Token Communication via Lookahead Search [32.63323958382152]
  This paper focuses on token packetization to maximize the average token similarity (ATS) between the original and received token channels.<n>To address this, we propose a novel framework of semantic aggregation with lookahead search (SemPA-Look)<n>SemPA-Look applies a lookahead search-inspired algorithm that samples intra-packet token candidates without replacement.
  arXiv  Detail & Related papers  (2025-06-24T09:25:44Z)
• 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)
• EXIT: Context-Aware Extractive Compression for Enhancing Retrieval-Augmented Generation [8.757777529568383]
  Current RAG systems often struggle when retrieval models fail to rank the most relevant documents.<n>We introduce EXIT, an extractive context compression framework.<n>Our evaluations show that EXIT consistently surpasses existing compression methods.
  arXiv  Detail & Related papers  (2024-12-17T05:38:27Z)
• Position IDs Matter: An Enhanced Position Layout for Efficient Context Compression in Large Language Models [50.637714223178456]
  We propose Enhanced Position Layout (EPL) to improve the context compression capability of large language models (LLMs)<n>EPL minimizes the distance between context tokens and their corresponding special tokens and at the same time maintains the sequence order in position IDs.<n>When extended to multimodal scenarios, EPL brings an average accuracy gain of 2.6 to vision compression LLMs.
  arXiv  Detail & Related papers  (2024-09-22T08:51:18Z)
• Token-level Correlation-guided Compression for Efficient Multimodal Document Understanding [54.532578213126065]
  Most document understanding methods preserve all tokens within sub-images and treat them equally. This neglects their different informativeness and leads to a significant increase in the number of image tokens. We propose Token-level Correlation-guided Compression, a parameter-free and plug-and-play methodology to optimize token processing.
  arXiv  Detail & Related papers  (2024-07-19T16:11:15Z)
• Hierarchical Context Merging: Better Long Context Understanding for Pre-trained LLMs [61.40047491337793]
  We present Hierarchical cOntext MERging (HOMER), a new training-free scheme designed to overcome the limitations of large language models. HomeR uses a divide-and-conquer algorithm, dividing long inputs into manageable chunks. A token reduction technique precedes each merging, ensuring memory usage efficiency.
  arXiv  Detail & Related papers  (2024-04-16T06:34:08Z)
• Adapting Language Models to Compress Contexts [71.98287002918941]
  Transformer-based language models (LMs) are powerful and widely-applicable tools, but their usefulness is constrained by a finite context window. We propose to adapt pre-trained LMs into AutoCompressors, which are capable of compressing long contexts into compact summary vectors. We fine-tune OPT and Llama-2 models on sequences of up to 30,720 tokens and show that AutoCompressors can utilize long contexts to improve perplexity.
  arXiv  Detail & Related papers  (2023-05-24T06:42:44Z)
• MASTER: Multi-task Pre-trained Bottlenecked Masked Autoencoders are Better Dense Retrievers [140.0479479231558]
  In this work, we aim to unify a variety of pre-training tasks into a multi-task pre-trained model, namely MASTER. MASTER utilizes a shared-encoder multi-decoder architecture that can construct a representation bottleneck to compress the abundant semantic information across tasks into dense vectors.
  arXiv  Detail & Related papers  (2022-12-15T13:57:07Z)
• RetroMAE v2: Duplex Masked Auto-Encoder For Pre-Training Retrieval-Oriented Language Models [3.4523793651427113]
  We propose duplex masked auto-encoder, a.k.a. DupMAE, which targets on improving the semantic representation capacity for contextualized embeddings of both [] and ordinary tokens. DupMAE is simple but empirically competitive: with a small decoding cost, it substantially contributes to the model's representation capability and transferability.
  arXiv  Detail & Related papers  (2022-11-16T08:57:55Z)
• Efficient Long Sequence Encoding via Synchronization [29.075962393432857]
  We propose a synchronization mechanism for hierarchical encoding. Our approach first identifies anchor tokens across segments and groups them by their roles in the original input sequence. Our approach is able to improve the global information exchange among segments while maintaining efficiency.
  arXiv  Detail & Related papers  (2022-03-15T04:37:02Z)
• Learning to Ask Conversational Questions by Optimizing Levenshtein Distance [83.53855889592734]
  We introduce a Reinforcement Iterative Sequence Editing (RISE) framework that optimize the minimum Levenshtein distance (MLD) through explicit editing actions. RISE is able to pay attention to tokens that are related to conversational characteristics. Experimental results on two benchmark datasets show that RISE significantly outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2021-06-30T08:44:19Z)

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.