Data Augmentation (DA) technique has been widely implemented in the computer vision field to relieve the data shortage, while the DA in Medical Image Analysis (MIA) is still mostly experience-driven. Here, we develop a plug-and-use DA method, named MedAugment, to introduce the automatic DA argumentation to the MIA field. To settle the difference between natural images and medical images, we divide the augmentation space into pixel augmentation space and spatial augmentation space. A novel operation sampling strategy is also proposed when sampling DA operations from the spaces. To demonstrate the performance and universality of MedAugment, we implement extensive experiments on four classification datasets and three segmentation datasets. The results show that our MedAugment outperforms most state-of-the-art DA methods. This work shows that the plug-and-use MedAugment may benefit the MIA community. Code is available at https://github.com/NUS-Tim/MedAugment_Pytorch.
Inspired by the superior language abilities of large language models (LLM), large vision-language models (LVLM) have been recently explored by integrating powerful LLMs for improving the performance on complex multimodal tasks. Despite the promising progress on LVLMs, we find that LVLMs suffer from the hallucination problem, i.e. they tend to generate objects that are inconsistent with the target images in the descriptions. To investigate it, this work presents the first systematic study on object hallucination of LVLMs. We conduct the evaluation experiments on several representative LVLMs, and show that they mostly suffer from severe object hallucination issue. We further discuss that the visual instructions may influence the hallucination, and find that: objects that frequently occur in the visual instructions or co-occur with the image objects, are obviously prone to be hallucinated by LVLMs. Besides, we find that existing evaluation methods might be affected by the input instructions and generation styles of LVLMs. Thus, we further design an improved evaluation method for object hallucination by proposing a polling-based query method called POPE. Experiment results demonstrate that our POPE can evaluate the object hallucination in a more stable and flexible way. Our codes and data are publicly available at https://github.com/RUCAIBox/POPE.
Recently, continuous diffusion models (CDM) have been introduced into non-autoregressive (NAR) text-to-text generation. However, the discrete nature of text increases the difficulty of CDM to generate coherent and fluent texts, and also causes the incompatibility problem between CDM and advanced NLP techniques, especially the popular pre-trained language models~(PLMs). To solve it, we propose Diffusion-NAT, which introduces discrete diffusion models~(DDM) into NAR text-to-text generation and integrates BART to improve the performance. By revising the decoding process of BART and the typical settings of DDM, we unify the inference process of BART and the denoising process of DDM into the same NAR masked tokens recovering task. In this way, DDM can rely on BART to perform denoising, which can benefit from both the rich pre-learned knowledge of BART and the iterative refining paradigm of DDM. Besides, we also propose the iterative self-prompting strategy to further improve the generation quality. Experimental results on 7 datasets show that our approach can outperform competitive NAR methods, and even surpass autoregressive methods. Our code and data will be publicly released.
Language is essentially a complex, intricate system of human expressions governed by grammatical rules. It poses a significant challenge to develop capable AI algorithms for comprehending and grasping a language. As a major approach, language modeling has been widely studied for language understanding and generation in the past two decades, evolving from statistical language models to neural language models. Recently, pre-trained language models (PLMs) have been proposed by pre-training Transformer models over large-scale corpora, showing strong capabilities in solving various NLP tasks. Since researchers have found that model scaling can lead to performance improvement, they further study the scaling effect by increasing the model size to an even larger size. Interestingly, when the parameter scale exceeds a certain level, these enlarged language models not only achieve a significant performance improvement but also show some special abilities that are not present in small-scale language models. To discriminate the difference in parameter scale, the research community has coined the term large language models (LLM) for the PLMs of significant size. Recently, the research on LLMs has been largely advanced by both academia and industry, and a remarkable progress is the launch of ChatGPT, which has attracted widespread attention from society. The technical evolution of LLMs has been making an important impact on the entire AI community, which would revolutionize the way how we develop and use AI algorithms. In this survey, we review the recent advances of LLMs by introducing the background, key findings, and mainstream techniques. In particular, we focus on four major aspects of LLMs, namely pre-training, adaptation tuning, utilization, and capacity evaluation. Besides, we also summarize the available resources for developing LLMs and discuss the remaining issues for future directions.
Non-autoregressive (NAR) text generation has attracted much attention in the field of natural language processing, which greatly reduces the inference latency but has to sacrifice the generation accuracy. Recently, diffusion models, a class of latent variable generative models, have been introduced into NAR text generation, showing improved generation quality. In this survey, we review the recent progress in diffusion models for NAR text generation. As the background, we first present the general definition of diffusion models and the text diffusion models, and then discuss their merits for NAR generation. As the core content, we further introduce two mainstream diffusion models in existing text diffusion works, and review the key designs of the diffusion process. Moreover, we discuss the utilization of pre-trained language models (PLMs) for text diffusion models and introduce optimization techniques for text data. Finally, we discuss several promising directions and conclude this paper. Our survey aims to provide researchers with a systematic reference of related research on text diffusion models for NAR generation.
Angle of arrival (AOA) is widely used to locate a wireless signal emitter. Compared with received signal strength (RSS) and time of arrival (TOA), it has higher accuracy and is not sensitive to time synchronization of the distributed sensors. However, there are few works focused on three-dimensional (3-D) scenario. Furthermore, although maximum likelihood estimator (MLE) has a relatively high performance, its computational complexity is ultra high. It is hard to employ it in practical applications. This paper proposed two multiplane geometric center based methods for 3-D AOA positioning. The first method could estimate the source position and angle measurement noise at the same time by seeking a center of the inscribed sphere, called CIS. Firstly, every sensor could measure two angles, azimuth angle and elevation angle. Based on that, two planes are constructed. Then, the estimated values of source position and angle noise are achieved by seeking the center and radius of the corresponding inscribed sphere. Deleting the estimation of the radius, the second algorithm, called MSD-LS, is born. It is not able to estimate angle noise but has lower computational complexity. Theoretical analysis and simulation results show that proposed methods could approach the Cramer-Rao lower bound (CRLB) and have lower complexity than MLE.