Large language models (LLMs) and small language models (SLMs) are being adopted at remarkable speed, although their safety still remains a serious concern. With the advent of multilingual S/LLMs, the question now becomes a matter of scale: can we expand multilingual safety evaluations of these models with the same velocity at which they are deployed? To this end we introduce RTP-LX, a human-transcreated and human-annotated corpus of toxic prompts and outputs in 28 languages. RTP-LX follows participatory design practices, and a portion of the corpus is especially designed to detect culturally-specific toxic language. We evaluate seven S/LLMs on their ability to detect toxic content in a culturally-sensitive, multilingual scenario. We find that, although they typically score acceptably in terms of accuracy, they have low agreement with human judges when judging holistically the toxicity of a prompt, and have difficulty discerning harm in context-dependent scenarios, particularly with subtle-yet-harmful content (e.g. microagressions, bias). We release of this dataset to contribute to further reduce harmful uses of these models and improve their safe deployment.
LoRA (Low-Rank Adaptation) has emerged as a preferred method for efficiently adapting Large Language Models (LLMs) with remarkable simplicity and efficacy. This note extends the original LoRA paper by offering new perspectives that were not initially discussed and presents a series of insights for deploying LoRA at scale. Without introducing new experiments, we aim to improve the understanding and application of LoRA.
We introduce QUICK, a group of novel optimized CUDA kernels for the efficient inference of quantized Large Language Models (LLMs). QUICK addresses the shared memory bank-conflict problem of state-of-the-art mixed precision matrix multiplication kernels. Our method interleaves the quantized weight matrices of LLMs offline to skip the shared memory write-back after the dequantization. We demonstrate up to 1.91x speedup over existing kernels of AutoAWQ on larger batches and up to 1.94x throughput gain on representative LLM models on various NVIDIA GPU devices.
In MRI, researchers have long endeavored to effectively visualize myelin distribution in the brain, a pursuit with significant implications for both scientific research and clinical applications. Over time, various methods such as myelin water imaging, magnetization transfer imaging, and relaxometric imaging have been developed, each carrying distinct advantages and limitations. Recently, an innovative technique named as magnetic susceptibility source separation has emerged, introducing a novel surrogate biomarker for myelin in the form of a diamagnetic susceptibility map. This paper comprehensively reviews this cutting-edge method, providing the fundamental concepts of magnetic susceptibility, susceptibility imaging, and the validation of the diamagnetic susceptibility map as a myelin biomarker. Additionally, the paper explores essential aspects of data acquisition and processing, offering practical insights for readers. A comparison with established myelin imaging methods is also presented, and both current and prospective clinical and scientific applications are discussed to provide a holistic understanding of the technique. This work aims to serve as a foundational resource for newcomers entering this dynamic and rapidly expanding field.
Cone beam computed tomography (CBCT) is an emerging medical imaging technique to visualize the internal anatomical structures of patients. During a CBCT scan, several projection images of different angles or views are collectively utilized to reconstruct a tomographic image. However, reducing the number of projections in a CBCT scan while preserving the quality of a reconstructed image is challenging due to the nature of an ill-posed inverse problem. Recently, a neural attenuation field (NAF) method was proposed by adopting a neural radiance field algorithm as a new way for CBCT reconstruction, demonstrating fast and promising results using only 50 views. However, decreasing the number of projections is still preferable to reduce potential radiation exposure, and a faster reconstruction time is required considering a typical scan time. In this work, we propose a fast and accurate sparse-view CBCT reconstruction (FACT) method to provide better reconstruction quality and faster optimization speed in the minimal number of view acquisitions ($<$ 50 views). In the FACT method, we meta-trained a neural network and a hash-encoder using a few scans (= 15), and a new regularization technique is utilized to reconstruct the details of an anatomical structure. In conclusion, we have shown that the FACT method produced better, and faster reconstruction results over the other conventional algorithms based on CBCT scans of different body parts (chest, head, and abdomen) and CT vendors (Siemens, Phillips, and GE).
Iron and myelin are primary susceptibility sources in the human brain. These substances are essential for healthy brain, and their abnormalities are often related to various neurological disorders. Recently, an advanced susceptibility mapping technique, which is referred to as chi-separation, has been proposed successfully disentangling paramagnetic iron from diamagnetic myelin, opening a new potential for generating iron map and myelin map in the brain. Utilizing this technique, this study constructs a normative chi-separation atlas from 106 healthy human brains. The resulting atlas provides detailed anatomical structures associated with the distributions of iron and myelin, clearly delineating subcortical nuclei and white matter fiber bundles. Additionally, susceptibility values in a number of regions of interest are reported along with age-dependent changes. This atlas may have direct applications such as localization of subcortical structures for deep brain stimulation or high-intensity focused ultrasound and also serve as a valuable resource for future research.
In MRI, images of the same contrast (e.g., T1) from the same subject can show noticeable differences when acquired using different hardware, sequences, or scan parameters. These differences in images create a domain gap that needs to be bridged by a step called image harmonization, in order to process the images successfully using conventional or deep learning-based image analysis (e.g., segmentation). Several methods, including deep learning-based approaches, have been proposed to achieve image harmonization. However, they often require datasets of multiple characteristics for deep learning training and may still be unsuccessful when applied to images of an unseen domain. To address this limitation, we propose a novel concept called "Blind Harmonization," which utilizes only target domain data for training but still has the capability of harmonizing unseen domain images. For the implementation of Blind Harmonization, we developed BlindHarmony using an unconditional flow model trained on target domain data. The harmonized image is optimized to have a correlation with the input source domain image while ensuring that the latent vector of the flow model is close to the center of the Gaussian. BlindHarmony was evaluated using simulated and real datasets and compared with conventional methods. BlindHarmony achieved a noticeable performance in both datasets, highlighting its potential for future use in clinical settings.
Denoising of magnetic resonance images is beneficial in improving the quality of low signal-to-noise ratio images. Recently, denoising using deep neural networks has demonstrated promising results. Most of these networks, however, utilize supervised learning, which requires large training images of noise-corrupted and clean image pairs. Obtaining training images, particularly clean images, is expensive and time-consuming. Hence, methods such as Noise2Noise (N2N) that require only pairs of noise-corrupted images have been developed to reduce the burden of obtaining training datasets. In this study, we propose a new self-supervised denoising method, Coil2Coil (C2C), that does not require the acquisition of clean images or paired noise-corrupted images for training. Instead, the method utilizes multichannel data from phased-array coils to generate training images. First, it divides and combines multichannel coil images into two images, one for input and the other for label. Then, they are processed to impose noise independence and sensitivity normalization such that they can be used for the training images of N2N. For inference, the method inputs a coil-combined image (e.g., DICOM image), enabling a wide application of the method. When evaluated using synthetic noise-added images, C2C shows the best performance against several self-supervised methods, reporting comparable outcomes to supervised methods. When testing the DICOM images, C2C successfully denoised real noise without showing structure-dependent residuals in the error maps. Because of the significant advantage of not requiring additional scans for clean or paired images, the method can be easily utilized for various clinical applications.
To rapidly obtain high resolution T2, T2* and quantitative susceptibility mapping (QSM) source separation maps with whole-brain coverage and high geometric fidelity. We propose Blip Up-Down Acquisition for Spin And Gradient Echo imaging (BUDA-SAGE), an efficient echo-planar imaging (EPI) sequence for quantitative mapping. The acquisition includes multiple T2*-, T2'- and T2-weighted contrasts. We alternate the phase-encoding polarities across the interleaved shots in this multi-shot navigator-free acquisition. A field map estimated from interim reconstructions was incorporated into the joint multi-shot EPI reconstruction with a structured low rank constraint to eliminate geometric distortion. A self-supervised MR-Self2Self (MR-S2S) neural network (NN) was utilized to perform denoising after BUDA reconstruction to boost SNR. Employing Slider encoding allowed us to reach 1 mm isotropic resolution by performing super-resolution reconstruction on BUDA-SAGE volumes acquired with 2 mm slice thickness. Quantitative T2 and T2* maps were obtained using Bloch dictionary matching on the reconstructed echoes. QSM was estimated using nonlinear dipole inversion (NDI) on the gradient echoes. Starting from the estimated R2 and R2* maps, R2' information was derived and used in source separation QSM reconstruction, which provided additional para- and dia-magnetic susceptibility maps. In vivo results demonstrate the ability of BUDA-SAGE to provide whole-brain, distortion-free, high-resolution multi-contrast images and quantitative T2 and T2* maps, as well as yielding para- and dia-magnetic susceptibility maps. Derived quantitative maps showed comparable values to conventional mapping methods in phantom and in vivo measurements. BUDA-SAGE acquisition with self-supervised denoising and Slider encoding enabled rapid, distortion-free, whole-brain T2, T2* mapping at 1 mm3 isotropic resolution in 90 seconds.
A carefully engineered radiofrequency (RF) pulse plays a key role in a number of systems such as mobile phone, radar, and magnetic resonance imaging (MRI). The design of an RF waveform, however, is often posed as an inverse problem that has no general solution. As a result, various design methods each with a specific purpose have been developed based on the intuition of human experts. In this work, we propose an artificial intelligence-powered RF pulse design framework, DeepRF, which utilizes the self-learning characteristics of deep reinforcement learning (DRL) to generate a novel RF beyond human intuition. Additionally, the method can design various types of RF pulses via customized reward functions. The algorithm of DeepRF consists of two modules: the RF generation module, which utilizes DRL to explore new RF pulses, and the RF refinement module, which optimizes the seed RF pulses from the generation module via gradient ascent. The effectiveness of DeepRF is demonstrated using four exemplary RF pulses, slice-selective excitation pulse, slice-selective inversion pulse, B1-insensitive volume inversion pulse, and B1-insensitive selective inversion pulse, that are commonly used in MRI. The results show that the DeepRF-designed pulses successfully satisfy the design criteria while improving specific absorption rates when compared to those of the conventional RF pulses. Further analyses suggest that the DeepRF-designed pulses utilize new mechanisms of magnetization manipulation that are difficult to be explained by conventional theory, suggesting the potentials of DeepRF in discovering unseen design dimensions beyond human intuition. This work may lay the foundation for an emerging field of AI-driven RF waveform design.