| Challenge: | Existing methods for generating long sequences of tokens are expensive and require memory and computation resources. |
| Approach: | They propose a method that alternates between full context attention and attention over a subset of input tokens during generation. |
| Outcome: | The proposed method achieves comparable speedup to eviction-based methods while improving performance for various long-form generation tasks. |
Similar Papers
KV Cache Compression, But What Must We Give in Return? A Comprehensive Benchmark of Long Context Capable Approaches (2024.findings-emnlp)
Copied to clipboard
Jiayi Yuan, Hongyi Liu, Shaochen Zhong, Yu-Neng Chuang, Songchen Li, Guanchu Wang, Duy Le, Hongye Jin, Vipin Chaudhary, Zhaozhuo Xu, Zirui Liu, Xia Hu
| Challenge: | Long context capability is a crucial competency for large language models as it mitigates the human struggle to digest long-form texts. |
| Approach: | They propose to evaluate 10+ state-of-the-art approaches for long context-capable LLMs. |
| Outcome: | The proposed methods are compared against 10+ state-of-the-art approaches across seven categories of long context tasks. |
Rectified Sparse Attention for Efficient Long-Sequence Generation (2026.findings-acl)
Copied to clipboard
Yutao Sun, Tianzhu Ye, Li Dong, Yuqing Xia, Jian Chen, Yizhao Gao, Shijie Cao, Jianyong Wang, Furu Wei
| Challenge: | Recent sparse decoding methods improve efficiency but suffer from KV cache misalignment, resulting in performance degradation. |
| Approach: | They propose a method that combines block-sparse attention with periodic dense rectification to bound error accumulation and preserve alignment with the pretraining distribution. |
| Outcome: | Experiments on math reasoning, language modeling, and retrieval tasks show that ReSA achieves near-lossless generation quality with significantly improved efficiency. |
Question Tells You Where the Answer Is: Intention-aware Long-Context KV Cache Compression (2026.acl-long)
Copied to clipboard
Liang Zhao, Xiaocheng Feng, Weihong Zhong, Lei Huang, Kun Zhu, Baoxin Wang, Dayong Wu, Guoping Hu, Ting Liu, Bing Qin
| Challenge: | Recent methods to reduce the KV cache size fail to identify crucial KVs for generation while excluding others accurately, resulting in severe information loss. |
| Approach: | They propose an intention-aware KV cache eviction method that identifies and retains crucial KVs according to the attention distribution of intention, which semantically reflects the user’s goal and determines which part of the context is relevant. |
| Outcome: | The proposed method can maintain the model performance while reducing the KV cache size from 128K to 2K, leading to a 6.3x increase in decoding speed and 7.8x enhancement in memory efficiency compared to the default setting. |
TokenSelect: Efficient Long-Context Inference and Length Extrapolation for LLMs via Dynamic Token-Level KV Cache Selection (2025.emnlp-main)
Copied to clipboard
| Challenge: | Rapid advances in Large Language Models have spurred demand for processing extended context sequences . however, performance degradation due to sequence lengths out-of-distribution and excessively long inference times are limiting LLMs in long-context scenarios. |
| Approach: | They propose a training-free method for efficient and accurate long-context inference . they selectively involves a few critical KV cache tokens in attention calculation . |
| Outcome: | The proposed method speeds up attention computation and accelerates inference time while reducing selection overhead. |
ZigZagKV: Dynamic KV Cache Compression for Long-context Modeling based on Layer Uncertainty (2025.coling-main)
Copied to clipboard
| Challenge: | Existing methods to accelerate inference of Large Language models (LLMs) are limited in their ability to retain key tokens as input length increases. |
| Approach: | They propose a method that leverages layer uncertainty to allocate budget size for each layer to reduce memory usage. |
| Outcome: | The proposed method reduces memory usage of the KV caches to only 20% when compared to full KV inference while achieving nearly lossless performance. |
Layer-Condensed KV Cache for Efficient Inference of Large Language Models (2024.acl-long)
Copied to clipboard
| Challenge: | Using a key-value cache, memory consumption is a bottleneck for high-throughput language models. |
| Approach: | They propose a method that only computes and caches the KVs of a small number of layers, thus saving memory consumption and improving inference throughput. |
| Outcome: | The proposed method achieves higher throughput and competitive performance than standard transformers and is orthogonal to existing transformer memory-saving techniques. |
FIER: Fine-Grained and Efficient KV Cache Retrieval for Long-context LLM Inference (2025.findings-emnlp)
Copied to clipboard
Dongwei Wang, Zijie Liu, Song Wang, Yuxin Ren, Jianing Deng, Jingtong Hu, Tianlong Chen, Huanrui Yang
| Challenge: | Key-Value (KV) cache reading latency increases with context lengths hindering LLM inference . important tokens are sparsely distributed across the long context, making existing retrieval inaccurate . |
| Approach: | They propose a method to retain a small fraction of KV cache based on token importance . important tokens are often sparsely distributed across the long context . |
| Outcome: | The proposed method reduces decoding latency by 1.2 to 1.5. |
FastKV: Decoupling of Context Reduction and KV Cache Compression for Prefill-Decoding Acceleration (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) excel at handling long-context sequences, but require substantial prefill computation and key-value (KV) cache. |
| Approach: | They propose a KV cache compression framework that decouples prefill computation from decoding KV budget. |
| Outcome: | The proposed framework reduces latency in prefill and decoding by leveraging the stabilization of token importance in later layers. |
MadaKV: Adaptive Modality-Perception KV Cache Eviction for Efficient Multimodal Long-Context Inference (2025.acl-long)
Copied to clipboard
Kunxi Li, Zhonghua Jiang, Zhouzhou Shen, ZhaodeWang ZhaodeWang, Chengfei Lv, Shengyu Zhang, Fan Wu, Fei Wu
| Challenge: | Existing KV cache eviction methods fail to capture modality-specific information, resulting in suboptimal performance. |
| Approach: | They propose a modality-adaptive key-value (KV) cache eviction strategy to enhance the efficiency of multimodal large language models in long-context inference. |
| Outcome: | The proposed method reduces the KV cache memory footprint and model inference latency while maintaining high accuracy across multimodal long-context tasks. |
TailorKV: A Hybrid Framework for Long-Context Inference via Tailored KV Cache Optimization (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing work mitigates memory overhead by offloading or compressing the Key-Value cache. |
| Approach: | They propose a method that integrates quantization and offloading into a generative large language model by using a hybrid compression method. |
| Outcome: | The proposed method outperforms the state-of-the-art in long-context evaluations. |