Reasoning models have demonstrated impressive performance in self-reflection and chain-of-thought reasoning. However, they often produce excessively long outputs, leading to prohibitively large key-value (KV) caches during inference. While chain-of-thought inference significantly improves performance on complex reasoning tasks, it can also lead to reasoning failures when deployed with existing KV cache compression approaches. To address this, we propose Redundancyaware KVCache Compression for Reasoning models (R-KV), a novel method specifically targeting redundant tokens in reasoning models. Our method preserves nearly 100% of the full KV cache performance using only 10% of the KV cache, substantially outperforming existing KV cache baselines, which reaches only 60% of the performance. Remarkably, R-KV even achieves 105% of full KV cache performance with 16% of the KV cache. This KV-cache reduction also leads to a 90% memory saving and a 6.6 throughput over standard chain-ofthought reasoning inference. Experimental results show that R-KV consistently outperforms existing KV cache compression baselines across two mathematical reasoning datasets.

KV cache eviction has emerged as an effective solution to alleviate resource constraints faced by LLMs in long-context scenarios. However, existing token-level eviction methods often overlook two critical aspects: (1) their irreversible eviction strategy fails to adapt to dynamic attention patterns during decoding (the saliency shift problem), and (2) they treat both marginally important tokens and truly unimportant tokens equally, despite the collective significance of marginal tokens to model performance (the marginal information over-compression problem). To address these issues, we design two compensation mechanisms based on the high similarity of attention matrices between LLMs of different scales. We propose SmallKV, a small model assisted compensation method for KV cache compression. SmallKV can maintain attention matching between different-scale LLMs to: 1) assist the larger model in perceiving globally important information of attention; and 2) use the smaller model's attention scores to approximate those of marginal tokens in the larger model. Extensive experiments on benchmarks including GSM8K, BBH, MT-Bench, and LongBench demonstrate the effectiveness of SmallKV. Moreover, efficiency evaluations show that SmallKV achieves 1.75 2.56 times higher throughput than baseline methods, highlighting its potential for efficient and performant LLM inference in resource constrained environments.

The growing memory footprint of the Key-Value (KV) cache poses a severe scalability bottleneck for long-context Large Language Model (LLM) inference. While KV cache eviction has emerged as an effective solution by discarding less critical tokens, existing token-, block-, and sentence-level compression methods struggle to balance semantic coherence and memory efficiency. To this end, we introduce SABlock, a \underline{s}emantic-aware KV cache eviction framework with \underline{a}daptive \underline{block} sizes. Specifically, SABlock first performs semantic segmentation to align compression boundaries with linguistic structures, then applies segment-guided token scoring to refine token importance estimation. Finally, for each segment, a budget-driven search strategy adaptively determines the optimal block size that preserves semantic integrity while improving compression efficiency under a given cache budget. Extensive experiments on long-context benchmarks demonstrate that SABlock consistently outperforms state-of-the-art baselines under the same memory budgets. For instance, on Needle-in-a-Haystack (NIAH), SABlock achieves 99.9% retrieval accuracy with only 96 KV entries, nearly matching the performance of the full-cache baseline that retains up to 8K entries. Under a fixed cache budget of 1,024, SABlock further reduces peak memory usage by 46.28% and achieves up to 9.5x faster decoding on a 128K context length.

Existing key-value (KV) cache compression methods typically rely on heuristics, such as uniform cache allocation across layers or static eviction policies, however, they ignore the critical interplays among layer-specific feature patterns and task performance, which can lead to degraded generalization. In this paper, we propose EvolKV, an adaptive framework for layer-wise, task-driven KV cache compression that jointly optimizes the memory efficiency and task performance. By reformulating cache allocation as a multi-objective optimization problem, EvolKV leverages evolutionary search to dynamically configure layer budgets while directly maximizing downstream performance. Extensive experiments on 11 tasks demonstrate that our approach outperforms all baseline methods across a wide range of KV cache budgets on long-context tasks and surpasses heuristic baselines by up to 7 percentage points on GSM8K. Notably, EvolKV achieves superior performance over the full KV cache setting on code completion while utilizing only 1.5% of the original budget, suggesting the untapped potential in learned compression strategies for KV cache budget allocation.

Large Language Models (LLMs) are increasingly deployed in long-context tasks such as reasoning, code generation, and multi-turn dialogue. However, inference over extended contexts is bottlenecked by the Key-Value (KV) cache, whose memory footprint grows linearly with sequence length and dominates latency at each decoding step. While recent KV cache compression methods identify and load important tokens, they focus predominantly on input contexts and fail to address the cumulative attention errors that arise during long decoding. In this paper, we introduce RetroAttention, a novel KV cache update technique that retrospectively revises past attention outputs using newly arrived KV entries from subsequent decoding steps. By maintaining a lightweight output cache, RetroAttention enables past queries to efficiently access more relevant context, while incurring minimal latency overhead. This breaks the fixed-attention-output paradigm and allows continual correction of prior approximations. Extensive experiments on long-generation benchmarks show that RetroAttention consistently outperforms state-of-the-art (SOTA) KV compression methods, increasing effective KV exposure by up to 1.6$\times$ and accuracy by up to 21.9\%.

Recent advances in large language models (LLMs) have significantly boosted long-context processing. However, the increasing key-value (KV) cache size poses critical challenges to memory and execution efficiency. Most KV cache compression methods rely on heuristic token eviction using all attention heads in Grouped Query Attention (GQA)-based LLMs. This method ignores the different functionalities of attention heads, leading to the eviction of critical tokens and thus degrades the performance of LLMs. To address the issue above, instead of using all the attention heads in GQA-based LLMs to determine important tokens as in the previous work, we first identify the attention heads in each layer that are not only capable of retrieving the initial and final tokens of a prompt, but also capable of retrieving important tokens within the text and attending to their surrounding semantic context. Afterwards, we exploit such heads to determine the important tokens and retain their corresponding KV cache pairs. Furthermore, we analyze the cache eviction error of each layer individually and introduce a layer-adaptive KV cache allocation strategy. Experimental results demonstrate the proposed CompressKV consistently outperforms state-of-the-art approaches under various memory budgets on LongBench and Needle-in-a-Haystack benchmarks. Our code is publicly available at: https://github.com/TUDa-HWAI/CompressKV.git.

Large language models (LLMs) with extended context windows have become increasingly prevalent for tackling complex tasks. However, the substantial Key-Value (KV) cache required for long-context LLMs poses significant deployment challenges. Existing approaches either discard potentially critical information needed for future generations or offer limited efficiency gains due to high computational overhead. In this paper, we introduce Chelsea, a simple yet effective framework for online KV cache clustering. Our approach is based on the observation that key states exhibit high similarity along the sequence dimension. To enable efficient clustering, we divide the sequence into chunks and propose Chunked Soft Matching, which employs an alternating partition strategy within each chunk and identifies clusters based on similarity. Chelsea then merges the KV cache within each cluster into a single centroid. Additionally, we provide a theoretical analysis of the computational complexity and the optimality of the intra-chunk partitioning strategy. Extensive experiments across various models and long-context benchmarks demonstrate that Chelsea achieves up to 80% reduction in KV cache memory usage while maintaining comparable model performance. Moreover, with minimal computational overhead, Chelsea accelerates the decoding stage of inference by up to 3.19$\times$ and reduces end-to-end latency by up to 2.72$\times$.

KV cache techniques in Transformer models aim to reduce redundant computations at the expense of substantially increased memory usage, making KV cache compression an important and popular research topic. Recently, state-of-the-art KV cache compression methods implement imbalanced, per-head allocation algorithms that dynamically adjust the KV cache budget for each attention head, achieving excellent performance in single-GPU scenarios. However, we observe that such imbalanced compression leads to significant load imbalance when deploying multi-GPU inference, as some GPUs become overburdened while others remain underutilized. In this paper, we propose FairKV, a method designed to ensure fair memory usage among attention heads in systems employing imbalanced KV cache compression. The core technique of FairKV is Fair-Copying, which replicates a small subset of memory-intensive attention heads across GPUs using data parallelism to mitigate load imbalance. Our experiments on popular models, including LLaMA 70b and Mistral 24b model, demonstrate that FairKV increases throughput by 1.66x compared to standard tensor parallelism inference. Our code will be released as open source upon acceptance.

KV cache is a widely used acceleration technique for large language models (LLMs) inference. However, its memory requirement grows rapidly with input length. Previous studies have reduced the size of KV cache by either removing the same number of unimportant tokens for all attention heads or by allocating differentiated KV cache budgets for pre-identified attention heads. However, due to the importance of attention heads varies across different tasks, the pre-identified attention heads fail to adapt effectively to various downstream tasks. To address this issue, we propose Task-KV, a method that leverages the semantic differentiation of attention heads to allocate differentiated KV cache budgets across various tasks. We demonstrate that attention heads far from the semantic center (called heterogeneous heads) make an significant contribution to task outputs and semantic understanding. In contrast, other attention heads play the role of aggregating important information and focusing reasoning. Task-KV allocates full KV cache budget to heterogeneous heads to preserve comprehensive semantic information, while reserving a small number of recent tokens and attention sinks for non-heterogeneous heads. Furthermore, we innovatively introduce middle activations to preserve key contextual information aggregated from non-heterogeneous heads. To dynamically perceive semantic differences among attention heads, we design a semantic separator to distinguish heterogeneous heads from non-heterogeneous ones based on their distances from the semantic center. Experimental results on multiple benchmarks and different model architectures demonstrate that Task-KV significantly outperforms existing baseline methods.