Searcharxiv⌕ Search

arXiv subjects

Yan Liu

Publications and source records attributed to Yan Liu.

At least 253 records · Page 14Linked to original sources
arXiv2023

Evaluating Large Language Models: A Comprehensive Survey

Large language models (LLMs) have demonstrated remarkable capabilities across a broad spectrum of tasks. They have attracted significant attention and been deployed in numerous downstream applications. Nevertheless, akin to a double-edged sword, LLMs also present potential risks. They could suffer from private data leaks or yield inappropriate, harmful, or misleading content. Additionally, the rapid progress of LLMs raises concerns about the potential emergence of superintelligent systems without adequate safeguards. To effectively capitalize on LLM capacities as well as ensure their safe and beneficial development, it is critical to conduct a rigorous and comprehensive evaluation of LLMs. This survey endeavors to offer a panoramic perspective on the evaluation of LLMs. We categorize the evaluation of LLMs into three major groups: knowledge and capability evaluation, alignment evaluation and safety evaluation. In addition to the comprehensive review on the evaluation methodologies and benchmarks on these three aspects, we collate a compendium of evaluations pertaining to LLMs' performance in specialized domains, and discuss the construction of comprehensive evaluation platforms that cover LLM evaluations on capabilities, alignment, safety, and applicability. We hope that this comprehensive overview will stimulate further research interests in the evaluation of LLMs, with the ultimate goal of making evaluation serve as a cornerstone in guiding the responsible development of LLMs. We envision that this will channel their evolution into a direction that maximizes societal benefit while minimizing potential risks. A curated list of related papers has been publicly available at https://github.com/tjunlp-lab/Awesome-LLMs-Evaluation-Papers.

cs.CL↗
arXiv2023

Sparse Interaction Additive Networks via Feature Interaction Detection and Sparse Selection

There is currently a large gap in performance between the statistically rigorous methods like linear regression or additive splines and the powerful deep methods using neural networks. Previous works attempting to close this gap have failed to fully investigate the exponentially growing number of feature combinations which deep networks consider automatically during training. In this work, we develop a tractable selection algorithm to efficiently identify the necessary feature combinations by leveraging techniques in feature interaction detection. Our proposed Sparse Interaction Additive Networks (SIAN) construct a bridge from these simple and interpretable models to fully connected neural networks. SIAN achieves competitive performance against state-of-the-art methods across multiple large-scale tabular datasets and consistently finds an optimal tradeoff between the modeling capacity of neural networks and the generalizability of simpler methods.

cs.LG↗
arXiv2023

SIAD: Self-supervised Image Anomaly Detection System

Recent trends in AIGC effectively boosted the application of visual inspection. However, most of the available systems work in a human-in-the-loop manner and can not provide long-term support to the online application. To make a step forward, this paper outlines an automatic annotation system called SsaA, working in a self-supervised learning manner, for continuously making the online visual inspection in the manufacturing automation scenarios. Benefit from the self-supervised learning, SsaA is effective to establish a visual inspection application for the whole life-cycle of manufacturing. In the early stage, with only the anomaly-free data, the unsupervised algorithms are adopted to process the pretext task and generate coarse labels for the following data. Then supervised algorithms are trained for the downstream task. With user-friendly web-based interfaces, SsaA is very convenient to integrate and deploy both of the unsupervised and supervised algorithms. So far, the SsaA system has been adopted for some real-life industrial applications.

cs.CV↗
arXiv2023

Large Language Model Alignment: A Survey

Recent years have witnessed remarkable progress made in large language models (LLMs). Such advancements, while garnering significant attention, have concurrently elicited various concerns. The potential of these models is undeniably vast; however, they may yield texts that are imprecise, misleading, or even detrimental. Consequently, it becomes paramount to employ alignment techniques to ensure these models to exhibit behaviors consistent with human values. This survey endeavors to furnish an extensive exploration of alignment methodologies designed for LLMs, in conjunction with the extant capability research in this domain. Adopting the lens of AI alignment, we categorize the prevailing methods and emergent proposals for the alignment of LLMs into outer and inner alignment. We also probe into salient issues including the models' interpretability, and potential vulnerabilities to adversarial attacks. To assess LLM alignment, we present a wide variety of benchmarks and evaluation methodologies. After discussing the state of alignment research for LLMs, we finally cast a vision toward the future, contemplating the promising avenues of research that lie ahead. Our aspiration for this survey extends beyond merely spurring research interests in this realm. We also envision bridging the gap between the AI alignment research community and the researchers engrossed in the capability exploration of LLMs for both capable and safe LLMs.

cs.CL↗
arXiv2023

Holographic superfluid with gauge-axion coupling

We have constructed a holographic superfluid with gauge-axion coupling. Depending on whether the coupling is positive or negative, the system displays metallic or insulating behavior in its normal state. A significant feature of the system is the appearance of a mid-IR peak in the alternating current (AC) conductivity in a certain range of parameters. This peak arises due to competition between explicit symmetry breaking (ESB) and spontaneous symmetry breaking (SSB), which results in the presence of a pseudo-Goldstone mode. Moreover, a dip in low-frequency AC conductivity is observed, stemming from the excitation of the SSB Goldstone mode. In the superfluid phase, the effect of gauge-axion coupling on the condensation or superfluid energy gap is only amplified in the presence of strong momentum dissipation. Notably, for the case with negative gauge-axion coupling, a hard-gap-like behavior at low frequency and a pronounced peak at intermediate frequency are observed, indicating that the evolution of the superfluid component is distinct from that of positive coupling.

hep-th↗
arXiv2023

Multi-level Multiple Instance Learning with Transformer for Whole Slide Image Classification

Whole slide image (WSI) refers to a type of high-resolution scanned tissue image, which is extensively employed in computer-assisted diagnosis (CAD). The extremely high resolution and limited availability of region-level annotations make employing deep learning methods for WSI-based digital diagnosis challenging. Recently integrating multiple instance learning (MIL) and Transformer for WSI analysis shows very promising results. However, designing effective Transformers for this weakly-supervised high-resolution image analysis is an underexplored yet important problem. In this paper, we propose a Multi-level MIL (MMIL) scheme by introducing a hierarchical structure to MIL, which enables efficient handling of MIL tasks involving a large number of instances. Based on MMIL, we instantiated MMIL-Transformer, an efficient Transformer model with windowed exact self-attention for large-scale MIL tasks. To validate its effectiveness, we conducted a set of experiments on WSI classification tasks, where MMIL-Transformer demonstrate superior performance compared to existing state-of-the-art methods, i.e., 96.80% test AUC and 97.67% test accuracy on the CAMELYON16 dataset, 99.04% test AUC and 94.37% test accuracy on the TCGA-NSCLC dataset, respectively. All code and pre-trained models are available at: https://github.com/hustvl/MMIL-Transformer

cs.CV↗
arXiv2023

Measuring, Interpreting, and Improving Fairness of Algorithms using Causal Inference and Randomized Experiments

Algorithm fairness has become a central problem for the broad adoption of artificial intelligence. Although the past decade has witnessed an explosion of excellent work studying algorithm biases, achieving fairness in real-world AI production systems has remained a challenging task. Most existing works fail to excel in practical applications since either they have conflicting measurement techniques and/ or heavy assumptions, or require code-access of the production models, whereas real systems demand an easy-to-implement measurement framework and a systematic way to correct the detected sources of bias. In this paper, we leverage recent advances in causal inference and interpretable machine learning to present an algorithm-agnostic framework (MIIF) to Measure, Interpret, and Improve the Fairness of an algorithmic decision. We measure the algorithm bias using randomized experiments, which enables the simultaneous measurement of disparate treatment, disparate impact, and economic value. Furthermore, using modern interpretability techniques, we develop an explainable machine learning model which accurately interprets and distills the beliefs of a blackbox algorithm. Altogether, these techniques create a simple and powerful toolset for studying algorithm fairness, especially for understanding the cost of fairness in practical applications like e-commerce and targeted advertising, where industry A/B testing is already abundant.

cs.LG↗
arXiv2023

AATCT-IDS: A Benchmark Abdominal Adipose Tissue CT Image Dataset for Image Denoising, Semantic Segmentation, and Radiomics Evaluation

Methods: In this study, a benchmark \emph{Abdominal Adipose Tissue CT Image Dataset} (AATTCT-IDS) containing 300 subjects is prepared and published. AATTCT-IDS publics 13,732 raw CT slices, and the researchers individually annotate the subcutaneous and visceral adipose tissue regions of 3,213 of those slices that have the same slice distance to validate denoising methods, train semantic segmentation models, and study radiomics. For different tasks, this paper compares and analyzes the performance of various methods on AATTCT-IDS by combining the visualization results and evaluation data. Thus, verify the research potential of this data set in the above three types of tasks. Results: In the comparative study of image denoising, algorithms using a smoothing strategy suppress mixed noise at the expense of image details and obtain better evaluation data. Methods such as BM3D preserve the original image structure better, although the evaluation data are slightly lower. The results show significant differences among them. In the comparative study of semantic segmentation of abdominal adipose tissue, the segmentation results of adipose tissue by each model show different structural characteristics. Among them, BiSeNet obtains segmentation results only slightly inferior to U-Net with the shortest training time and effectively separates small and isolated adipose tissue. In addition, the radiomics study based on AATTCT-IDS reveals three adipose distributions in the subject population. Conclusion: AATTCT-IDS contains the ground truth of adipose tissue regions in abdominal CT slices. This open-source dataset can attract researchers to explore the multi-dimensional characteristics of abdominal adipose tissue and thus help physicians and patients in clinical practice. AATCT-IDS is freely published for non-commercial purpose at: \url{https://figshare.com/articles/dataset/AATTCT-IDS/23807256}.

eess.IV↗
arXiv2023

Asymptotic behavior of solutions for the thermoviscous acoustic systems

We study some asymptotic properties of solutions for the acoustic coupled systems in thermoviscous fluids which was proposed by [Karlsen-Bruus, \emph{Phys. Rev. E} (2015)]. Basing on the WKB analysis and the Fourier analysis, we derive optimal estimates and large time asymptotic profiles of the energy term via diagonalization procedure, and of the velocity potential via reduction methodology. We found that the wave effect has a dominant influence for lower dimensions comparing with thermal-viscous effects. Moreover, by employing suitable energy methods, we rigorously demonstrate global (in time) inviscid limits as the momentum diffusion coefficient vanishes, whose limit model can be regarded as the thermoelastic acoustic systems in isotropic solids. These results explain some influence of the momentum diffusion on asymptotic behavior of solutions.

math.AP↗
arXiv2023

The influence of viscous dissipations on the nonlinear acoustic wave equation with second sound

We study the effect of a viscous dissipation on the Cauchy problem for a Cattaneo-type model in nonlinear acoustics, established by applying the Lighthill approximation for the viscous or inviscid fluid model. The contribution of this paper is twofold. For the nonlinear viscous Cattaneo-type model involving a fractional Laplacian $(-Δ)^α$ in the viscous damping with $α\in[0,1]$, we derive optimal decay rates for global (in time) solutions with small data in certain Sobolev spaces. Furthermore, by introducing a threshold $α=1/2$ for the power of the fractional viscous dissipation, we derive an anomalous diffusion profile when $α\in[0,1/2)$ and a diffusion wave profile when $α\in[1/2,1]$ for large-time. Whereas, for the nonlinear inviscid Cattaneo-type model (or the Jordan-Moore-Gibson-Thompson equation in the critical case), we obtain the blow-up of the energy solutions in finite time under suitable assumptions for the initial data. Thus, the presence of a viscous dissipation in the nonlinear Cattaneo-type model is a criterion for the global (in time) existence and blow-up of solutions.

math.AP↗
arXiv2023

Attention-based 3D CNN with Multi-layer Features for Alzheimer's Disease Diagnosis using Brain Images

Structural MRI and PET imaging play an important role in the diagnosis of Alzheimer's disease (AD), showing the morphological changes and glucose metabolism changes in the brain respectively. The manifestations in the brain image of some cognitive impairment patients are relatively inconspicuous, for example, it still has difficulties in achieving accurate diagnosis through sMRI in clinical practice. With the emergence of deep learning, convolutional neural network (CNN) has become a valuable method in AD-aided diagnosis, but some CNN methods cannot effectively learn the features of brain image, making the diagnosis of AD still presents some challenges. In this work, we propose an end-to-end 3D CNN framework for AD diagnosis based on ResNet, which integrates multi-layer features obtained under the effect of the attention mechanism to better capture subtle differences in brain images. The attention maps showed our model can focus on key brain regions related to the disease diagnosis. Our method was verified in ablation experiments with two modality images on 792 subjects from the ADNI database, where AD diagnostic accuracies of 89.71% and 91.18% were achieved based on sMRI and PET respectively, and also outperformed some state-of-the-art methods.

eess.IV↗
arXiv2023

GLA-GCN: Global-local Adaptive Graph Convolutional Network for 3D Human Pose Estimation from Monocular Video

3D human pose estimation has been researched for decades with promising fruits. 3D human pose lifting is one of the promising research directions toward the task where both estimated pose and ground truth pose data are used for training. Existing pose lifting works mainly focus on improving the performance of estimated pose, but they usually underperform when testing on the ground truth pose data. We observe that the performance of the estimated pose can be easily improved by preparing good quality 2D pose, such as fine-tuning the 2D pose or using advanced 2D pose detectors. As such, we concentrate on improving the 3D human pose lifting via ground truth data for the future improvement of more quality estimated pose data. Towards this goal, a simple yet effective model called Global-local Adaptive Graph Convolutional Network (GLA-GCN) is proposed in this work. Our GLA-GCN globally models the spatiotemporal structure via a graph representation and backtraces local joint features for 3D human pose estimation via individually connected layers. To validate our model design, we conduct extensive experiments on three benchmark datasets: Human3.6M, HumanEva-I, and MPI-INF-3DHP. Experimental results show that our GLA-GCN implemented with ground truth 2D poses significantly outperforms state-of-the-art methods (e.g., up to around 3%, 17%, and 14% error reductions on Human3.6M, HumanEva-I, and MPI-INF-3DHP, respectively). GitHub: https://github.com/bruceyo/GLA-GCN.

cs.CV↗
arXiv2023

Mechanical Artifacts in Optical Projection Tomography: Classification and Automatic Calibration

Optical projection tomography (OPT) is a powerful tool for biomedical studies. It achieves 3D visualization of mesoscopic biological samples with high spatial resolution using conventional tomographic-reconstruction algorithms. However, various artifacts degrade the quality of the reconstructed images due to experimental imperfections in the OPT instruments. While many efforts have been made to characterize and correct for these artifacts, they focus on one specific type of artifacts. This work has two contributions. First, we systematically document a catalog of mechanical artifacts based on a 3D description of the imaging system that uses a set of angular and translational parameters. Then, we introduce a calibration algorithm that recovers the unknown system parameters fed into the final 3D iterative reconstruction algorithm for a distortion-free volumetric image. Simulations with beads data and experimental results on a fluorescent textile fiber confirm that our algorithm successfully removes miscalibration artifacts in the reconstruction.

cs.CV↗
arXiv2023

Secrets of RLHF in Large Language Models Part I: PPO

Large language models (LLMs) have formulated a blueprint for the advancement of artificial general intelligence. Its primary objective is to function as a human-centric (helpful, honest, and harmless) assistant. Alignment with humans assumes paramount significance, and reinforcement learning with human feedback (RLHF) emerges as the pivotal technological paradigm underpinning this pursuit. Current technical routes usually include \textbf{reward models} to measure human preferences, \textbf{Proximal Policy Optimization} (PPO) to optimize policy model outputs, and \textbf{process supervision} to improve step-by-step reasoning capabilities. However, due to the challenges of reward design, environment interaction, and agent training, coupled with huge trial and error cost of large language models, there is a significant barrier for AI researchers to motivate the development of technical alignment and safe landing of LLMs. The stable training of RLHF has still been a puzzle. In the first report, we dissect the framework of RLHF, re-evaluate the inner workings of PPO, and explore how the parts comprising PPO algorithms impact policy agent training. We identify policy constraints being the key factor for the effective implementation of the PPO algorithm. Therefore, we explore the PPO-max, an advanced version of PPO algorithm, to efficiently improve the training stability of the policy model. Based on our main results, we perform a comprehensive analysis of RLHF abilities compared with SFT models and ChatGPT. The absence of open-source implementations has posed significant challenges to the investigation of LLMs alignment. Therefore, we are eager to release technical reports, reward models and PPO codes, aiming to make modest contributions to the advancement of LLMs.

cs.CL↗
arXiv2023

SAITS: Self-Attention-based Imputation for Time Series

Missing data in time series is a pervasive problem that puts obstacles in the way of advanced analysis. A popular solution is imputation, where the fundamental challenge is to determine what values should be filled in. This paper proposes SAITS, a novel method based on the self-attention mechanism for missing value imputation in multivariate time series. Trained by a joint-optimization approach, SAITS learns missing values from a weighted combination of two diagonally-masked self-attention (DMSA) blocks. DMSA explicitly captures both the temporal dependencies and feature correlations between time steps, which improves imputation accuracy and training speed. Meanwhile, the weighted-combination design enables SAITS to dynamically assign weights to the learned representations from two DMSA blocks according to the attention map and the missingness information. Extensive experiments quantitatively and qualitatively demonstrate that SAITS outperforms the state-of-the-art methods on the time-series imputation task efficiently and reveal SAITS' potential to improve the learning performance of pattern recognition models on incomplete time-series data from the real world. The code is open source on GitHub at https://github.com/WenjieDu/SAITS.

cs.LG↗
arXiv2023

DSRM: Boost Textual Adversarial Training with Distribution Shift Risk Minimization

Adversarial training is one of the best-performing methods in improving the robustness of deep language models. However, robust models come at the cost of high time consumption, as they require multi-step gradient ascents or word substitutions to obtain adversarial samples. In addition, these generated samples are deficient in grammatical quality and semantic consistency, which impairs the effectiveness of adversarial training. To address these problems, we introduce a novel, effective procedure for instead adversarial training with only clean data. Our procedure, distribution shift risk minimization (DSRM), estimates the adversarial loss by perturbing the input data's probability distribution rather than their embeddings. This formulation results in a robust model that minimizes the expected global loss under adversarial attacks. Our approach requires zero adversarial samples for training and reduces time consumption by up to 70\% compared to current best-performing adversarial training methods. Experiments demonstrate that DSRM considerably improves BERT's resistance to textual adversarial attacks and achieves state-of-the-art robust accuracy on various benchmarks.

cs.CL↗
arXiv2023

Engineering Perovskite Emissions via Optical Quasi-Bound-States-in-the-Continuum

Metal halide perovskite quantum dots (PQDs) have emerged as promising materials due to their exceptional photoluminescence (PL) properties. A wide range of applications could benefit from adjustable luminescence properties, while preserving the physical and chemical properties of the PQDs. Therefore, post-synthesis engineering has gained attention recently, involving the use of ion-exchange or external stimuli, such as extreme pressure, magnetic and electric fields. Nevertheless, these methods typically suffer from spectrum broadening, intensity quenching or yield multiple bands. Alternatively, photonic antennas can modify the radiative decay channel of perovskites via the Purcell effect, with the largest wavelength shift being 8 nm to date, at an expense of 5-fold intensity loss. Here, we present an optical nanoantenna array with polarization-controlled quasi-bound-states-in-the-continuum (q-BIC) resonances, which can engineer and shift the photoluminescence wavelength over a ~39 nm range and confers a 21-fold emission enhancement of FAPbI3 perovskite QDs. The spectrum is engineered in a non-invasive manner via lithographically defined antennas and the pump laser polarization at ambient conditions. Our research provides a path towards advanced optoelectronic devices, such as spectrally tailored quantum emitters and lasers.

physics.optics↗
arXiv2023

On the Generalization and Advancement of Half-Sine-Based Pulse Shaping Filters for Constant Envelope OQPSK Modulation

The offset quadrature phase-shift keying (OQPSK) modulation is a key factor for the technique of ZigBee, which has been adopted in IEEE 802.15.4 for wireless communications of Internet of Things (IoT) and Internet of Vehicles (IoV), etc. In this paper, we propose the general conditions of pulse shaping filters (PSFs) with constant envelope (CE) property for OQPSK modulation, which can be easily leveraged to design the PSFs with CE property. Based on these conditions, we further design an advanced PSF called $α$-half-sine PSF. It is verified that the newly designed $α$-half-sine PSF can not only keep the CE property for OQPSK but also achieve better performance than the traditional PSFs in certain scenarios. Moreover, the $α$-half-sine PSF can be simply adjusted to achieve a flexible performance tradeoff between the transition roll-off speed and out-of-band leakage.

eess.SP↗