Related papers: A Simple Plug-in for Improving Eviction-Based KV C…

Key-Value (KV) cache has become a bottleneck of LLMs for long-context generation. Despite the numerous efforts in this area, the optimization for the decoding phase is generally ignored. However, we believe such optimization is crucial,…

Quantization methods have been introduced to perform large scale approximate nearest search tasks. Residual Vector Quantization (RVQ) is one of the effective quantization methods. RVQ uses a multi-stage codebook learning scheme to lower the…

Multi-vector models dominate Visual Document Retrieval (VDR) due to their fine-grained matching capabilities, but their high storage and computational costs present a major barrier to practical deployment. In this paper, we propose…

Numerical simulations of geological CO2 storage in deep saline aquifers have demonstrated that vertical equilibrium (VE) models offer a robust and computationally efficient framework for reservoir optimization and upscaling. These studies…

Large reasoning models (LRMs) often incur significant key-value (KV) cache overhead, due to their linear growth with the verbose chain-of-thought (CoT) reasoning. This incurs both memory overhead and throughput bottlenecks, limiting…

Token compression expedites the training and inference of Vision Transformers (ViTs) by reducing the number of the redundant tokens, e.g., pruning inattentive tokens or merging similar tokens. However, when applied to downstream tasks,…

The computational expense of redundant vision tokens in Large Vision-Language Models (LVLMs) has led many existing methods to compress them via a vision projector. However, this compression may lose visual information that is crucial for…

Large Language Models (LLMs) have significantly advanced the field of Artificial Intelligence. However, their deployment is resource-intensive, not only due to the large number of model parameters but also because the (Key-Value) KV cache…

Modern retrieval-augmented generation (RAG) systems treat vector embeddings as static, context-free artifacts: an embedding has no notion of when it was created, how trustworthy its source is, or which other embeddings depend on it. This…

Recent advances in long-text understanding have pushed the context length of large language models (LLMs) up to one million tokens. It boosts LLMs's accuracy and reasoning capacity but causes exorbitant computational costs and…

Visual Document Retrieval (VDR), the task of retrieving visually-rich document pages using queries that combine visual and textual cues, is crucial for numerous real-world applications. Recent state-of-the-art methods leverage Large…

Scaling the context size of large language models (LLMs) enables them to perform various new tasks, e.g., book summarization. However, the memory cost of the Key and Value (KV) cache in attention significantly limits the practical…

Key-value (KV) caching has become the de-facto to accelerate generation speed for large language models (LLMs) inference. However, the growing cache demand with increasing sequence length has transformed LLM inference to be a memory bound…

KV cache compression is critical for efficient long-context LLM inference. Approaches that reduce the per-pair footprint -- quantization and low-rank decomposition -- are orthogonal to those that reduce the sequence length of the cache.…

Active Visual Exploration (AVE) optimizes the utilization of robotic resources in real-world scenarios by sequentially selecting the most informative observations. However, modern methods require a high computational budget due to…

Enhancing the adaptive capabilities of large language models is a critical pursuit in both research and application. Traditional fine-tuning methods require substantial data and computational resources, especially for enhancing specific…

The Key-Value (KV) cache reading latency increases significantly with context lengths, hindering the efficiency of long-context LLM inference. To address this, previous works propose retaining a small fraction of KV cache based on token…

Vector Quantisation (VQ) is experiencing a comeback in machine learning, where it is increasingly used in representation learning. However, optimizing the codevectors in existing VQ-VAE is not entirely trivial. A problem is codebook…

Transformer-based document cross-encoder rerankers are a central component of modern information retrieval systems. Despite their success, these models suffer from high computational costs due to processing long query-document sequences at…

With the skyrocketing costs of GPUs and their virtual instances in the cloud, there is a significant desire to use CPUs for large language model (LLM) inference. KV cache update, often implemented as allocation, copying, and in-place…