English
Related papers

Related papers: CASK: Core-Aware Selective KV Compression for Reas…

200 papers

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…

Machine Learning · Computer Science 2025-09-11 Bohan Yu , Yekun Chai

Transformer-based large language models (LLMs) have demonstrated remarkable potential across a wide range of practical applications. However, long-context inference remains a significant challenge due to the substantial memory requirements…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-01-09 Bo Jiang , Taolue Yang , Youyuan Liu , Xubin He , Sheng Di , Sian Jin

Streaming video understanding with large vision-language models (VLMs) requires a compact memory that can support future reasoning over an ever-growing visual history. A common solution is to compress the key-value (KV) cache, but existing…

Computer Vision and Pattern Recognition · Computer Science 2026-05-15 Ailar Mahdizadeh , Puria Azadi , Muchen Li , Xiangteng He , Leonid Sigal

Transformer-based large language models (LLMs) rely on key-value (KV) caching to avoid redundant computation during autoregressive inference. While this mechanism greatly improves efficiency, the cache size grows linearly with the input…

Machine Learning · Computer Science 2026-03-12 Jinwoo Ahn , Ingyu Seong , Akhil Kedia , Junhan Kim , Hyemi Jang , Kangwook Lee , Yongkweon Jeon

Inference for Large Language Models (LLMs) is computationally demanding. To reduce the cost of auto-regressive decoding, Key-Value (KV) cache is used to store intermediate activations, which significantly lowers the computational overhead…

Machine Learning · Computer Science 2025-06-05 Chaoyi Jiang , Lei Gao , Hossein Entezari Zarch , Murali Annavaram

As the context length of current large language models (LLMs) rapidly increases, the memory demand for the Key-Value (KV) cache is becoming a bottleneck for LLM deployment and batch processing. Traditional KV cache compression methods…

Computation and Language · Computer Science 2025-12-23 Aomufei Yuan , Zhiming Wang , Ruijie Miao , Dayu Wang , Yuxuan Tian , Zihan Wang , Yebo Peng , Yuhan Wu , Bairen Yi , Xin Liu , Tong Yang

Serving large language models (LLMs) at scale necessitates efficient key-value (KV) cache management. KV caches can be reused across conversation turns via shared-prefix prompts that are common in iterative code editing and chat. However,…

Computation and Language · Computer Science 2026-03-12 Konrad Staniszewski , Adrian Łańcucki

Generating long sequences of tokens given a long-context input is a very compute-intensive inference scenario for large language models (LLMs). One prominent inference speed-up approach is to construct a smaller key-value (KV) cache,…

Computation and Language · Computer Science 2025-03-04 Fangyuan Xu , Tanya Goyal , Eunsol Choi

\ac{CoT} prompting improves LLM accuracy on complex tasks but often increases token usage and inference cost. Existing ``Budget Forcing'' methods reduce cost via fine-tuning with heuristic length penalties, suppressing both essential…

Machine Learning · Computer Science 2026-05-19 Fabio Valerio Massoli , Andrey Kuzmin , Arash Behboodi

Reusing KV cache is essential for high efficiency of Large Language Model (LLM) inference systems. With more LLM users, the KV cache footprint can easily exceed GPU memory capacity, so prior work has proposed to either evict KV cache to…

We present a comprehensive framework for enhancing Retrieval-Augmented Generation (RAG) systems through dynamic retrieval strategies and reinforcement fine-tuning. This approach significantly improves large language models on…

Machine Learning · Computer Science 2025-05-21 Sakhinana Sagar Srinivas , Akash Das , Shivam Gupta , Venkataramana Runkana

Several works have developed eviction policies to remove key-value (KV) pairs from the KV cache for more efficient inference. The focus has been on compressing the KV cache after the input prompt has been processed for faster token…

Computation and Language · Computer Science 2025-07-04 Michael R. Metel , Boxing Chen , Mehdi Rezagholizadeh

Large Language Models (LLMs) often rely on long chain-of-thought (CoT) reasoning to solve complex tasks. While effective, these trajectories are frequently inefficient, leading to high latency from excessive token generation, or unstable…

Efficient key-value (KV) cache compression is critical for scaling transformer-based Large Language Models (LLMs) in long sequences and resource-limited settings. Existing methods evict tokens based on their positions or importance scores,…

Computation and Language · Computer Science 2025-05-19 Ziwei He , Jian Yuan , Haoli Bai , Jingwen Leng , Bo Jiang

Multi-hop retrieval-augmented generation (RAG) is a promising strategy for complex reasoning, yet existing iterative prompting approaches remain inefficient. They often regenerate predictable token sequences at every step and rely on…

Computation and Language · Computer Science 2025-10-23 Jihwan Bang , Juntae Lee , Seunghan Yang , Sungha Choi

We investigate learned KV cache compression through Speculative Importance Prediction (SIP), a 1.7M parameter non-query-aware scorer that predicts token importance from KV representations alone. Despite architectural sophistication…

Machine Learning · Computer Science 2026-01-22 Brady Steele

Long chains of thought (Long CoTs) are widely employed in multimodal reasoning models to tackle complex tasks by capturing detailed visual information. However, these Long CoTs are often excessively lengthy and contain redundant reasoning…

Artificial Intelligence · Computer Science 2026-02-11 Yizhi Wang , Linan Yue , Min-Ling Zhang

Long-horizon LLM inference turns the key--value (KV) cache into the dominant GPU memory consumer and makes per-token attention increasingly expensive. Many common eviction policies use static recency windows or historical attention, leaving…

Machine Learning · Computer Science 2026-05-26 Yubo Li , Yidi Miao

Through systematic experiments on long-context generation, we observe a damaging failure mode in which decoding can collapse into persistent repetition loops. We find that this degeneration is driven by collapsed attention patterns, where a…

Artificial Intelligence · Computer Science 2026-04-14 Dongjie Xu , Hao Wu , Weijie Shi , Yue Cui , Yuanjun Liu , Jiawei Li , Haolun Ma , An Liu , Jia Zhu , Jiajie Xu

Large Language Models (LLMs) are increasingly deployed in multi-turn conversational applications, where the management of the Key-Value (KV) Cache presents a significant bottleneck. The linear growth of the KV Cache with dialogue history…

Computation and Language · Computer Science 2025-10-09 Xiang Liu , Hong Chen , Xuming Hu , Xiaowen Chu
‹ Prev 1 8 9 10 Next ›