Optimizing KV Cache Eviction in LLMs: Adaptive Allocation for Enhanced Budget Utilization

Large Language Models have excelled in various fields but encounter efficiency limitations due to the extensive KV cache required for long sequences inference. Many efforts try to evict non-critical cache elements during runtime, thereby reducing cache size within a given memory budget while preserving generation quality. Our reexamination of their underlying principles discerns that prevailing strategies essentially aim to minimize an upper bound of eviction loss within a specific budget allocation. However, we observe that the current practice of uniformly allocating budgets across different attention heads during the eviction procedure tends to degrade the quality of generation posten-eviction. In light of these findings, we propose a simple yet effective adaptive allocation algorithm that not only theoretically ensures its loss upper bound does not exceed that of previous uniform allocation methods, but also effectively aligns with the characteristics of the self-attention mechanism, thus practically reducing the upper bound. Further, integrating this algorithm with two of the most advanced methods yields Ada-SnapKV and Ada-Pyramid. Extensive experimental validation across 16 datasets and the Needle-in-a-Haystack test confirm that Ada-SnapKV and Ada-Pyramid achieve further enhancements, establishing new benchmarks in state-of-the-art performance.

Learn to Explore: on Bootstrapping Interactive Data Exploration with Meta-learning

Interactive data exploration (IDE) is an effective way of comprehending big data, whose volume and complexity are beyond human abilities. The main goal of IDE is to discover user interest regions from a database through multi-rounds of user labelling. Existing IDEs adopt active-learning framework, where users iteratively discriminate or label the interestingness of selected tuples. The process of data exploration can be viewed as the process of training a classifier, which determines whether a database tuple is interesting to a user. An efficient exploration thus takes very few iterations of user labelling to reach the data region of interest. In this work, we consider the data exploration as the process of few-shot learning, where the classifier is learned with only a few training examples, or exploration iterations. To this end, we propose a learning-to-explore framework, based on meta-learning, which learns how to learn a classifier with automatically generated meta-tasks, so that the exploration process can be much shortened. Extensive experiments on real datasets show that our proposal outperforms existing explore-by-example solutions in terms of accuracy and efficiency.

Isomer: Transfer enhanced Dual-Channel Heterogeneous Dependency Attention Network for Aspect-based Sentiment Classification

Aspect-based sentiment classification aims to predict the sentiment polarity of a specific aspect in a sentence. However, most existing methods attempt to construct dependency relations into a homogeneous dependency graph with the sparsity and ambiguity, which cannot cover the comprehensive contextualized features of short texts or consider any additional node types or semantic relation information. To solve those issues, we present a sentiment analysis model named Isomer, which performs a dual-channel attention on heterogeneous dependency graphs incorporating external knowledge, to effectively integrate other additional information. Specifically, a transfer-enhanced dual-channel heterogeneous dependency attention network is devised in Isomer to model short texts using heterogeneous dependency graphs. These heterogeneous dependency graphs not only consider different types of information but also incorporate external knowledge. Experiments studies show that our model outperforms recent models on benchmark datasets. Furthermore, the results suggest that our method captures the importance of various information features to focus on informative contextual words.

GASNet: Weakly-supervised Framework for COVID-19 Lesion Segmentation

Segmentation of infected areas in chest CT volumes is of great significance for further diagnosis and treatment of COVID-19 patients. Due to the complex shapes and varied appearances of lesions, a large number of voxel-level labeled samples are generally required to train a lesion segmentation network, which is a main bottleneck for developing deep learning based medical image segmentation algorithms. In this paper, we propose a weakly-supervised lesion segmentation framework by embedding the Generative Adversarial training process into the Segmentation Network, which is called GASNet. GASNet is optimized to segment the lesion areas of a COVID-19 CT by the segmenter, and to replace the abnormal appearance with a generated normal appearance by the generator, so that the restored CT volumes are indistinguishable from healthy CT volumes by the discriminator. GASNet is supervised by chest CT volumes of many healthy and COVID-19 subjects without voxel-level annotations. Experiments on three public databases show that when using as few as one voxel-level labeled sample, the performance of GASNet is comparable to fully-supervised segmentation algorithms trained on dozens of voxel-level labeled samples.

Learning Directional Feature Maps for Cardiac MRI Segmentation

Cardiac MRI segmentation plays a crucial role in clinical diagnosis for evaluating personalized cardiac performance parameters. Due to the indistinct boundaries and heterogeneous intensity distributions in the cardiac MRI, most existing methods still suffer from two aspects of challenges: inter-class indistinction and intra-class inconsistency. To tackle these two problems, we propose a novel method to exploit the directional feature maps, which can simultaneously strengthen the differences between classes and the similarities within classes. Specifically, we perform cardiac segmentation and learn a direction field pointing away from the nearest cardiac tissue boundary to each pixel via a direction field (DF) module. Based on the learned direction field, we then propose a feature rectification and fusion (FRF) module to improve the original segmentation features, and obtain the final segmentation. The proposed modules are simple yet effective and can be flexibly added to any existing segmentation network without excessively increasing time and space complexity. We evaluate the proposed method on the 2017 MICCAI Automated Cardiac Diagnosis Challenge (ACDC) dataset and a large-scale self-collected dataset, showing good segmentation performance and robust generalization ability of the proposed method.