LeanK: Learnable K Cache Channel Pruning for Efficient Decoding (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing efforts to optimize the key-value (KV) cache include: (1) Eviction, which discards cache of less important tokens; (2) Selection, which retains the full KV cache but selectively reads relevant entries. |
| Approach: | They propose a learning-based method that prunes unimportant key (K) cache channels by leveraging static channel sparsity. |
| Outcome: | Experiments show that LeanK reduces GPU memory and accelerates decoding without sacrificing accuracy. |
Similar Papers
Efficient Sparse Attention needs Adaptive Token Release (2024.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated remarkable capabilities across a wide array of text-centric tasks, however, their ‘large’ scale introduces significant computational and storage challenges, particularly in managing the key-value states of the transformer, which limits their wider applicability. |
| Approach: | They propose to release resources from caches and rebuild key-value states by a lightweight controller module to approximate an ideal top-K sparse attention. |
| Outcome: | The proposed method achieves a significant throughput improvement of 221.8% over full attention and a model with 7 billion tokens. |
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. |
HATA: Trainable and Hardware-Efficient Hash-Aware Top-k Attention for Scalable Large Model Inference (2025.findings-acl)
Copied to clipboard
Ping Gong, Jiawei Yi, Shengnan Wang, Juncheng Zhang, Zewen Jin, Ouxiang Zhou, Ruibo Liu, Guanbin Xu, Youhui Bai, Bowen Ye, Kun Yuan, Tong Yang, Gong Zhang, Renhai Chen, Feng Wu, Cheng Li
| Challenge: | Existing top-k attention methods struggle to strike a balance between efficiency and accuracy. |
| Approach: | They propose a top-k attention approach that integrates low-overhead techniques into the Top-k Attention process to achieve 7.2 speedup compared to vanilla full attention. |
| Outcome: | The proposed approach achieves 7.2 speedup compared to current top-k attention methods while maintaining model accuracy. |
RefreshKV: Updating Small KV Cache During Long-form Generation (2025.acl-long)
Copied to clipboard
| 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. |
KVPR: Efficient LLM Inference with I/O-Aware KV Cache Partial Recomputation (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to inference large language models are limited by CPU capabilities and memory constraints. |
| Approach: | They propose an efficient I/O-aware LLM inference method that overlaps GPU computation with KV cache transfer to minimize idle GPU time. |
| Outcome: | The proposed method reduces the cost of auto-regressive decoding by 35.8% . it also achieves 46.2% higher throughput during decoding compared to state-of-the-art methods. |
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. |
SpindleKV: A Novel KV Cache Reduction Method Balancing Both Shallow and Deep Layers (2025.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) have impressive capabilities across various fields, but their widespread use is facing a severe and realistic challenge, which is their high demand for GPU memory. |
| Approach: | They propose a KV cache reduction method which balances both shallow and deep layers by using an attention weight based eviction method and a codebook based replacement approach. |
| Outcome: | The proposed method reduces the KV cache for shallower layers while preserving similar or even better model 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. |
Sparse-to-Dense: A Free Lunch for Lossless Acceleration of Video Understanding in LLMs (2025.acl-short)
Copied to clipboard
| Challenge: | Recent advances in Video Large Language Models (Video-LLMs) have achieved exceptional performance on tasks like video question answering and captioning. |
| Approach: | They propose a decoding strategy that leverages sparse top-K attention and dense full attention to accelerate Video-LLMs without loss. |
| Outcome: | The proposed approach achieves a 1.94 walltime speedup in video processing. |
Less is KEN: a Universal and Simple Non-Parametric Pruning Algorithm for Large Language Models (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing pruning algorithms suffer from limitations such as architecture specificity and reliance on demanding calculations. |
| Approach: | They propose a pruning algorithm based on Kernel Density Estimation . it preserves most significant parameters while restoring others to their pre-training state . |
| Outcome: | The proposed pruning algorithm achieves better performance than the original unpruned version. |