Search

LLM-powered coding and development assistants have become prevalent to programmers' workflows. However, concerns about the trustworthiness of LLMs for code persist despite their widespread use. Much of the existing research focused on either training or evaluation, raising questions about whether stakeholders in training and evaluation align in their understanding of model trustworthiness and whether they can move toward a unified direction. In this paper, we propose a vision for a unified trustworthiness auditing framework, DataTrust, which adopts a data-centric approach that synergistically emphasizes both training and evaluation data and their correlations. DataTrust aims to connect model trustworthiness indicators in evaluation with data quality indicators in training. It autonomously inspects training data and evaluates model trustworthiness using synthesized data, attributing potential causes from specific evaluation data to corresponding training data and refining indicator connections. Additionally, a trustworthiness arena powered by DataTrust will engage crowdsourced input and deliver quantitative outcomes. We outline the benefits that various stakeholders can gain from DataTrust and discuss the challenges and opportunities it presents., Comment: Short Vision Paper

Code Large Language Models (LLMs) have demonstrated remarkable capabilities in generating, understanding, and manipulating programming code. However, their training process inadvertently leads to the memorization of sensitive information, posing severe privacy risks. Existing studies on memorization in LLMs primarily rely on prompt engineering techniques, which suffer from limitations such as widespread hallucination and inefficient extraction of the target sensitive information. In this paper, we present a novel approach to characterize real and fake secrets generated by Code LLMs based on token probabilities. We identify four key characteristics that differentiate genuine secrets from hallucinated ones, providing insights into distinguishing real and fake secrets. To overcome the limitations of existing works, we propose DESEC, a two-stage method that leverages token-level features derived from the identified characteristics to guide the token decoding process. DESEC consists of constructing an offline token scoring model using a proxy Code LLM and employing the scoring model to guide the decoding process by reassigning token likelihoods. Through extensive experiments on four state-of-the-art Code LLMs using a diverse dataset, we demonstrate the superior performance of DESEC in achieving a higher plausible rate and extracting more real secrets compared to existing baselines. Our findings highlight the effectiveness of our token-level approach in enabling an extensive assessment of the privacy leakage risks associated with Code LLMs., Comment: 12 pages, 8 figures

Augmenting the multi-step reasoning abilities of Large Language Models (LLMs) has been a persistent challenge. Recently, verification has shown promise in improving solution consistency by evaluating generated outputs. However, current verification approaches suffer from sampling inefficiencies, requiring a large number of samples to achieve satisfactory performance. Additionally, training an effective verifier often depends on extensive process supervision, which is costly to acquire. In this paper, we address these limitations by introducing a novel verification method based on Twisted Sequential Monte Carlo (TSMC). TSMC sequentially refines its sampling effort to focus exploration on promising candidates, resulting in more efficient generation of high-quality solutions. We apply TSMC to LLMs by estimating the expected future rewards at partial solutions. This approach results in a more straightforward training target that eliminates the need for step-wise human annotations. We empirically demonstrate the advantages of our method across multiple math benchmarks, and also validate our theoretical analysis of both our approach and existing verification methods.

Code generation aims to automatically generate code from input requirements, significantly enhancing development efficiency. Recent large language models (LLMs) based approaches have shown promising results and revolutionized code generation task. Despite the promising performance, LLMs often generate contents with hallucinations, especially for the code generation scenario requiring the handling of complex contextual dependencies in practical development process. Although previous study has analyzed hallucinations in LLM-powered code generation, the study is limited to standalone function generation. In this paper, we conduct an empirical study to study the phenomena, mechanism, and mitigation of LLM hallucinations within more practical and complex development contexts in repository-level generation scenario. First, we manually examine the code generation results from six mainstream LLMs to establish a hallucination taxonomy of LLM-generated code. Next, we elaborate on the phenomenon of hallucinations, analyze their distribution across different models. We then analyze causes of hallucinations and identify four potential factors contributing to hallucinations. Finally, we propose an RAG-based mitigation method, which demonstrates consistent effectiveness in all studied LLMs. The replication package including code, data, and experimental results is available at https://github.com/DeepSoftwareAnalytics/LLMCodingHallucination, Comment: 11 pages, 13 figures

Recent advancements in deep learning have shown impressive results in image and video denoising, leveraging extensive pairs of noisy and noise-free data for supervision. However, the challenge of acquiring paired videos for dynamic scenes hampers the practical deployment of deep video denoising techniques. In contrast, this obstacle is less pronounced in image denoising, where paired data is more readily available. Thus, a well-trained image denoiser could serve as a reliable spatial prior for video denoising. In this paper, we propose a novel unsupervised video denoising framework, named ``Temporal As a Plugin'' (TAP), which integrates tunable temporal modules into a pre-trained image denoiser. By incorporating temporal modules, our method can harness temporal information across noisy frames, complementing its power of spatial denoising. Furthermore, we introduce a progressive fine-tuning strategy that refines each temporal module using the generated pseudo clean video frames, progressively enhancing the network's denoising performance. Compared to other unsupervised video denoising methods, our framework demonstrates superior performance on both sRGB and raw video denoising datasets.

Nonlinear optical (NLO) effects in materials with band crossings have attracted significant research interests due to the divergent band geometric quantities around these crossings. Most current research has focused on band crossings between the valence and conduction bands. However, such crossings are absent in insulators, which are more relevant for NLO applications. In this work, we demonstrate that NLO effects can be significantly enhanced by band crossings within the valence or conduction bands, which we designate as "deep band crossings" (DBCs). As an example, in two dimensions, we show that shift conductivity can be substantially enhanced or even divergent due to a mirror-protected "deep Dirac nodal point". In three dimensions, we propose GeTe as an ideal material where shift conductivity is enhanced by "deep Dirac nodal lines". The ubiquity of this enhancement is further confirmed by high-throughput calculations. Other types of DBCs and NLO effects are also discussed. By manipulating band crossings between arbitrary bands, our work offers a simple, practical, and universal way to greatly enhance NLO effects.

Recent breakthroughs in large language models (LLMs) offer unprecedented natural language understanding and generation capabilities. However, existing surveys on LLMs in biomedicine often focus on specific applications or model architectures, lacking a comprehensive analysis that integrates the latest advancements across various biomedical domains. This review, based on an analysis of 484 publications sourced from databases including PubMed, Web of Science, and arXiv, provides an in-depth examination of the current landscape, applications, challenges, and prospects of LLMs in biomedicine, distinguishing itself by focusing on the practical implications of these models in real-world biomedical contexts. Firstly, we explore the capabilities of LLMs in zero-shot learning across a broad spectrum of biomedical tasks, including diagnostic assistance, drug discovery, and personalized medicine, among others, with insights drawn from 137 key studies. Then, we discuss adaptation strategies of LLMs, including fine-tuning methods for both uni-modal and multi-modal LLMs to enhance their performance in specialized biomedical contexts where zero-shot fails to achieve, such as medical question answering and efficient processing of biomedical literature. Finally, we discuss the challenges that LLMs face in the biomedicine domain including data privacy concerns, limited model interpretability, issues with dataset quality, and ethics due to the sensitive nature of biomedical data, the need for highly reliable model outputs, and the ethical implications of deploying AI in healthcare. To address these challenges, we also identify future research directions of LLM in biomedicine including federated learning methods to preserve data privacy and integrating explainable AI methodologies to enhance the transparency of LLMs.

We study convex optimization problems under differential privacy (DP). With heavy-tailed gradients, existing works achieve suboptimal rates. The main obstacle is that existing gradient estimators have suboptimal tail properties, resulting in a superfluous factor of $d$ in the union bound. In this paper, we explore algorithms achieving optimal rates of DP optimization with heavy-tailed gradients. Our first method is a simple clipping approach. Under bounded $p$-th order moments of gradients, with $n$ samples, it achieves $\tilde{O}(\sqrt{d/n}+\sqrt{d}(\sqrt{d}/n\epsilon)^{1-1/p})$ population risk with $\epsilon\leq 1/\sqrt{d}$. We then propose an iterative updating method, which is more complex but achieves this rate for all $\epsilon\leq 1$. The results significantly improve over existing methods. Such improvement relies on a careful treatment of the tail behavior of gradient estimators. Our results match the minimax lower bound in \cite{kamath2022improved}, indicating that the theoretical limit of stochastic convex optimization under DP is achievable.

Noncollinear ferroic materials are sought after as testbeds to explore the intimate connections between topology and symmetry, which result in electronic, optical and magnetic functionalities not observed in collinear ferroic materials. For example, ferroaxial materials have ordered rotational structural distortions that break mirror symmetry and induce chirality. When ferroaxial order is coupled with ferroelectricity arising from a broken inversion symmetry, it offers the prospect of electric-field-control of the ferroaxial distortions and opens up new tunable functionalities. However, chiral ferroelectrics, especially ones stable at room temperature, are rare. We report the discovery of a strain-stabilized, room-temperature chiral ferroelectric phase in single crystals of BaTiS$_3$, a quasi-one-dimensional (1D) hexagonal chalcogenide. Using first-principles calculations, we predict the stabilization of this multiferroic phase having $P6_3$ space group for biaxial tensile strains exceeding 1.5% applied on the basal ab-plane of the room temperature $P6_3cm$ phase of BaTiS$_3$. The chiral ferroelectric phase is characterized by rotational distortions of select TiS$_6$ octahedra around the long $c$-axis and polar displacement of Ti atoms along the $c$-axis. We used an innovative approach using focused ion beam milling to make appropriately strained samples of BaTiS$_3$. The ferroaxial and ferroelectric distortions, and their domains in $P6_3$-BaTiS$_3$ were directly resolved using atomic resolution scanning transmission electron microscopy. Landau-based phenomenological modeling predicts a strong coupling between the ferroelectric and the ferroaxial order making $P6_3$-BaTiS$_3$ an attractive test bed for achieving electric-field control of chirality-related phenomena such as circular photo-galvanic current and the Rashba effect., Comment: 19 pages, 4 figures

Recent discoveries of zero-field fractional Chern insulators in moir\'e materials have attracted intensive research interests. However, most current theoretical and experimental attempts focus on systems with low Chern number bands, in analogy to the Landau levels. Here we propose candidate material systems for realizing fractional Chern insulators with higher Chern numbers. The material setup involves $\Gamma$-valley twisted homobilayer transition metal dichalcogenides in proximity to a skyrmion lattice. The skyrmion exchange potential induces a flat band with a high Chern number $C = -2$. Using the momentum-space projected exact diagonalization method, we perform a comprehensive study at various filling factors, confirming the generalized Jain series. Our research provides theoretical guidance on realizing unconventional fractional Chern insulators beyond the Landau level picture., Comment: 6 pages, 4 figures

We present foundation language models developed to power Apple Intelligence features, including a ~3 billion parameter model designed to run efficiently on devices and a large server-based language model designed for Private Cloud Compute. These models are designed to perform a wide range of tasks efficiently, accurately, and responsibly. This report describes the model architecture, the data used to train the model, the training process, how the models are optimized for inference, and the evaluation results. We highlight our focus on Responsible AI and how the principles are applied throughout the model development.

Deep learning electronic structures from ab initio calculations holds great potential to revolutionize computational materials studies. While existing methods proved success in deep-learning density functional theory (DFT) Hamiltonian matrices, they are limited to DFT programs using localized atomic-like bases and heavily depend on the form of the bases. Here, we propose the DeepH-r method for deep-learning DFT Hamiltonians in real space, facilitating the prediction of DFT Hamiltonian in a basis-independent manner. An equivariant neural network architecture for modeling the real-space DFT potential is developed, targeting a more fundamental quantity in DFT. The real-space potential exhibits simplified principles of equivariance and enhanced nearsightedness, further boosting the performance of deep learning. When applied to evaluate the Hamiltonian matrix, this method significantly improved in accuracy, as exemplified in multiple case studies. Given the abundance of data in the real-space potential, this work may pave a novel pathway for establishing a ``large materials model" with increased accuracy.

Shadow removal aims at restoring the image content within shadow regions, pursuing a uniform distribution of illumination that is consistent between shadow and non-shadow regions. {Comparing to other image restoration tasks, there are two unique challenges in shadow removal:} 1) The patterns of shadows are arbitrary, varied, and often have highly complex trace structures, making ``trace-less'' image recovery difficult. 2) The degradation caused by shadows is spatially non-uniform, resulting in inconsistencies in illumination and color between shadow and non-shadow areas. Recent developments in this field are primarily driven by deep learning-based solutions, employing a variety of learning strategies, network architectures, loss functions, and training data. Nevertheless, a thorough and insightful review of deep learning-based shadow removal techniques is still lacking. In this paper, we are the first to provide a comprehensive survey to cover various aspects ranging from technical details to applications. We highlight the major advancements in deep learning-based single-image shadow removal methods, thoroughly review previous research across various categories, and provide insights into the historical progression of these developments. Additionally, we summarize performance comparisons both quantitatively and qualitatively. Beyond the technical aspects of shadow removal methods, we also explore potential future directions for this field., Comment: url: https://github.com/GuoLanqing/Awesome-Shadow-Removal

Audio-visual segmentation (AVS) is an emerging task that aims to accurately segment sounding objects based on audio-visual cues. The success of AVS learning systems depends on the effectiveness of cross-modal interaction. Such a requirement can be naturally fulfilled by leveraging transformer-based segmentation architecture due to its inherent ability to capture long-range dependencies and flexibility in handling different modalities. However, the inherent training issues of transformer-based methods, such as the low efficacy of cross-attention and unstable bipartite matching, can be amplified in AVS, particularly when the learned audio query does not provide a clear semantic clue. In this paper, we address these two issues with the new Class-conditional Prompting Machine (CPM). CPM improves the bipartite matching with a learning strategy combining class-agnostic queries with class-conditional queries. The efficacy of cross-modal attention is upgraded with new learning objectives for the audio, visual and joint modalities. We conduct experiments on AVS benchmarks, demonstrating that our method achieves state-of-the-art (SOTA) segmentation accuracy.

Python's dynamic typing system offers flexibility and expressiveness but can lead to type-related errors, prompting the need for automated type inference to enhance type hinting. While existing learning-based approaches show promising inference accuracy, they struggle with practical challenges in comprehensively handling various types, including complex generic types and (unseen) user-defined types. In this paper, we introduce TIGER, a two-stage generating-then-ranking (GTR) framework, designed to effectively handle Python's diverse type categories. TIGER leverages fine-tuned pre-trained code models to train a generative model with a span masking objective and a similarity model with a contrastive training objective. This approach allows TIGER to generate a wide range of type candidates, including complex generics in the generating stage, and accurately rank them with user-defined types in the ranking stage. Our evaluation on the ManyTypes4Py dataset shows TIGER's advantage over existing methods in various type categories, notably improving accuracy in inferring user-defined and unseen types by 11.2% and 20.1% respectively in Top-5 Exact Match. Moreover, the experimental results not only demonstrate TIGER's superior performance and efficiency, but also underscore the significance of its generating and ranking stages in enhancing automated type inference., Comment: Accepted by ICSE'25

Automated cephalometric landmark detection is crucial in real-world orthodontic diagnosis. Current studies mainly focus on only adult subjects, neglecting the clinically crucial scenario presented by adolescents whose landmarks often exhibit significantly different appearances compared to adults. Hence, an open question arises about how to develop a unified and effective detection algorithm across various age groups, including adolescents and adults. In this paper, we propose CeLDA, the first work for Cephalometric Landmark Detection across Ages. Our method leverages a prototypical network for landmark detection by comparing image features with landmark prototypes. To tackle the appearance discrepancy of landmarks between age groups, we design new strategies for CeLDA to improve prototype alignment and obtain a holistic estimation of landmark prototypes from a large set of training images. Moreover, a novel prototype relation mining paradigm is introduced to exploit the anatomical relations between the landmark prototypes. Extensive experiments validate the superiority of CeLDA in detecting cephalometric landmarks on both adult and adolescent subjects. To our knowledge, this is the first effort toward developing a unified solution and dataset for cephalometric landmark detection across age groups. Our code and dataset will be made public on https://github.com/ShanghaiTech-IMPACT/Cephalometric-Landmark-Detection-across-Ages-with-Prototypical-Network, Comment: MICCAI 2024

Realizing large materials models has emerged as a critical endeavor for materials research in the new era of artificial intelligence, but how to achieve this fantastic and challenging objective remains elusive. Here, we propose a feasible pathway to address this paramount pursuit by developing universal materials models of deep-learning density functional theory Hamiltonian (DeepH), enabling computational modeling of the complicated structure-property relationship of materials in general. By constructing a large materials database and substantially improving the DeepH method, we obtain a universal materials model of DeepH capable of handling diverse elemental compositions and material structures, achieving remarkable accuracy in predicting material properties. We further showcase a promising application of fine-tuning universal materials models for enhancing specific materials models. This work not only demonstrates the concept of DeepH's universal materials model but also lays the groundwork for developing large materials models, opening up significant opportunities for advancing artificial intelligence-driven materials discovery.

Large language models (LLMs), pre-trained or fine-tuned on large code corpora, have shown effectiveness in generating code completions. However, in LLM-based code completion, LLMs may struggle to use correct and up-to-date Application Programming Interfaces (APIs) due to the rapid and continuous evolution of libraries. While existing studies have highlighted issues with predicting incorrect APIs, the specific problem of deprecated API usage in LLM-based code completion has not been thoroughly investigated. To address this gap, we conducted the first evaluation study on deprecated API usage in LLM-based code completion. This study involved seven advanced LLMs, 145 API mappings from eight popular Python libraries, and 28,125 completion prompts. The study results reveal the \textit{status quo} and \textit{root causes} of deprecated API usage in LLM-based code completion from the perspectives of \textit{model}, \textit{prompt}, and \textit{library}. Based on these findings, we propose two lightweight fixing approaches, \textsc{ReplaceAPI} and \textsc{InsertPrompt}, which can serve as baseline approaches for future research on mitigating deprecated API usage in LLM-based completion. Additionally, we provide implications for future research on integrating library evolution with LLM-driven software development.

The recent observation of the fractional quantum anomalous Hall effect in moir\'e fractional Chern insulators (FCI) provides opportunities for investigating zero magnetic field anyons. So far, both experimental and theoretical results suggest that filling > 1/3 FCI states in the first Chern band share features with those of the lowest Landau level (LL). To create the possibility of realizing non-Abelian anyons, one route is to engineer higher flat Chern bands that mimic higher LLs. Here, we investigate the interaction, topology, and ferromagnetism of the second moir\'e miniband in twisted MoTe2 bilayer (tMoTe2). Around filling factor v = -3, i.e., half-filling of the second miniband, we uncover spontaneous ferromagnetism and an incipient Chern insulator state. By measuring the anomalous Hall effect as a function of twist angle, we find that the Chern numbers (C) of the top two moir\'e flat bands have opposite sign (C = -+1) at twist angles above 3.1{\deg} but the same sign (C = -1) around 2.6{\deg}. This observation is consistent with the recently predicted twist-angle dependent band topology, resulting from the competition between moir\'e ferroelectricity and piezoelectricity. As we increase the magnetic field, only the small twist-angle device (2.6{\deg}) experiences a topological phase transition with an emergent C = -2 state. This is attributed to a Zeeman field-induced band crossing between opposite valleys, with the determined C = -1 for the top two bands. Our results lay a firm foundation for understanding the higher flat Chern bands, which is essential for the prediction or discovery of non-Abelian FCIs., Comment: 24 pages, 4 figures

Entanglement in quantum many-body systems is typically fragile to interactions with the environment. Generic unital quantum channels, for example, have the maximally mixed state with no entanglement as their unique steady state. However, we find that for a unital quantum channel that is `strongly symmetric', i.e. it preserves a global on-site symmetry, the maximally mixed steady state in certain symmetry sectors can be highly entangled. For a given symmetry, we analyze the entanglement and correlations of the maximally mixed state in the invariant sector (MMIS), and show that the entanglement of formation and distillation are exactly computable and equal for any bipartition. For all Abelian symmetries, the MMIS is separable, and for all non-Abelian symmetries, the MMIS is entangled. Remarkably, for non-Abelian continuous symmetries described by compact semisimple Lie groups (e.g. $SU(2)$), the bipartite entanglement of formation for the MMIS scales logarithmically $\sim \log N$ with the number of qudits $N$., Comment: 17+8 pages, 7 figures

Commit message is one of the most important textual information in software development and maintenance. However, it is time-consuming to write commit messages manually. Commit Message Generation (CMG) has become a research hotspot. Recently, several pre-trained language models (PLMs) and large language models (LLMs) with code capabilities have been introduced, demonstrating impressive performance on code-related tasks. Meanwhile, prior studies have explored the utilization of retrieval techniques for CMG, but it is still unclear what effects would emerge from combining advanced retrieval techniques with various generation models. This paper proposed REACT, a REtrieval-Augmented framework for CommiT message generation. It integrates advanced retrieval techniques with different PLMs and LLMs, to enhance the performance of these models on the CMG task. Specifically, a hybrid retriever is designed and used to retrieve the most relevant code diff and commit message pair as an exemplar. Then, the retrieved pair is utilized to guide and enhance the CMG task by PLMs and LLMs through fine-tuning and in-context learning. The experimental results show that REACT significantly enhances these models' performance on the CMG task, improving the BLEU score of CodeT5 by up to 55%, boosting Llama 3's BLEU score by 102%, and substantially surpassing all baselines., Comment: 22 pages, 5 images, 6 tables, Manuscript submitted to a journal (2024)

By forming measurement matrices with the Kronecker product of two random matrices, image encryption in computational ghost imaging is investigated. The two-dimensional images are conveniently reconstructed with the pseudo-inverse matrices of the two random matrices. To suppress the noise, the method of truncated singular value decomposition can be applied to either or both of the two pseudo-inverse matrices. Further, our proposal facilitates for image encryption since more matrices can be involved in forming the measurement matrix. Two permutation matrices are inserted into the matrix sequence. The image information can only be reconstructed with the correct permutation matrices and the matrix sequence in image decryption. The experimental results show the facilitations our proposal. The technique paves the way for the practicality and flexibility of computational ghost imaging., Comment: 5 pages, 4 figures

The development of AI-Generated Content (AIGC) has empowered the creation of remarkably realistic AI-generated videos, such as those involving Sora. However, the widespread adoption of these models raises concerns regarding potential misuse, including face video scams and copyright disputes. Addressing these concerns requires the development of robust tools capable of accurately determining video authenticity. The main challenges lie in the dataset and neural classifier for training. Current datasets lack a varied and comprehensive repository of real and generated content for effective discrimination. In this paper, we first introduce an extensive video dataset designed specifically for AI-Generated Video Detection (GenVidDet). It includes over 2.66 M instances of both real and generated videos, varying in categories, frames per second, resolutions, and lengths. The comprehensiveness of GenVidDet enables the training of a generalizable video detector. We also present the Dual-Branch 3D Transformer (DuB3D), an innovative and effective method for distinguishing between real and generated videos, enhanced by incorporating motion information alongside visual appearance. DuB3D utilizes a dual-branch architecture that adaptively leverages and fuses raw spatio-temporal data and optical flow. We systematically explore the critical factors affecting detection performance, achieving the optimal configuration for DuB3D. Trained on GenVidDet, DuB3D can distinguish between real and generated video content with 96.77% accuracy, and strong generalization capability even for unseen types.

With the rising popularity of Large Language Models (LLMs), assessing their trustworthiness through security tasks has gained critical importance. Regarding the new task of universal goal hijacking, previous efforts have concentrated solely on optimization algorithms, overlooking the crucial role of the prompt. To fill this gap, we propose a universal goal hijacking method called POUGH that incorporates semantic-guided prompt processing strategies. Specifically, the method starts with a sampling strategy to select representative prompts from a candidate pool, followed by a ranking strategy that prioritizes the prompts. Once the prompts are organized sequentially, the method employs an iterative optimization algorithm to generate the universal fixed suffix for the prompts. Experiments conducted on four popular LLMs and ten types of target responses verified the effectiveness of our method., Comment: 15 pages

A scenario of ghost imaging with hybrid transform approach is proposed by integrating Hadamard, discrete cosine, and Haar matrices. The measurement matrix is formed by the Kronecker product of the two different transform matrices. The image information can be conveniently reconstructed by the corresponding inverse matrices. In experiment, six hybridization sets are performed in computational ghost imaging. For an object of staggered stripes, only one bucket signal survives in the Hadamard-cosine, Haar-Hadamard, and Haar-cosine hybridization sets, demonstrating flexible image compression. For a handmade windmill object, the quality factors of the reconstructed images vary with the hybridization sets. Sub-Nyquist sampling can be applied to either or both of the different transform matrices in each hybridization set in experiment. The hybridization method can be extended to apply more transforms at once. Ghost imaging with hybrid transforms may find flexible applications in image processing, such as image compression and image encryption., Comment: 5 pages, 4 figures

Symmetry in mixed quantum states can manifest in two distinct forms: \textit{strong symmetry}, where each individual pure state in the quantum ensemble is symmetric with the same charge, and \textit{weak symmetry}, which applies only to the entire ensemble. This paper explores a novel type of spontaneous symmetry breaking (SSB) where a strong symmetry is broken to a weak one. While the SSB of a weak symmetry is measured by the long-ranged two-point correlation function $\mathrm{Tr}(O_xO^{\dagger}_y\rho)$, the strong-to-weak SSB (SW-SSB) is measured by the fidelity $F(\rho, O_xO^{\dagger}_y\rho O_yO^{\dagger}_x)$, dubbed the \textit{fidelity correlator}. We prove that SW-SSB is a universal property of mixed-state quantum phases, in the sense that the phenomenon of SW-SSB is robust against symmetric low-depth local quantum channels. { We also show that the symmetry breaking is "spontaneous " in the sense that the effect of a local symmetry-breaking measurement cannot be recovered locally.} We argue that a thermal state at a nonzero temperature in the canonical ensemble (with fixed symmetry charge) should have spontaneously broken strong symmetry. Additionally, we study non-thermal scenarios where decoherence induces SW-SSB, leading to phase transitions described by classical statistical models with bond randomness. In particular, the SW-SSB transition of a decohered Ising model can be viewed as the "ungauged" version of the celebrated toric code decodability transition. We confirm that, in the decohered Ising model, the SW-SSB transition defined by the fidelity correlator is the only physical transition in terms of channel recoverability. We also comment on other (inequivalent) definitions of SW-SSB, through correlation functions with higher R\'enyi indices., Comment: 17+6 pages, 4 figures

Reconstructing human dynamic vision from brain activity is a challenging task with great scientific significance. The difficulty stems from two primary issues: (1) vision-processing mechanisms in the brain are highly intricate and not fully revealed, making it challenging to directly learn a mapping between fMRI and video; (2) the temporal resolution of fMRI is significantly lower than that of natural videos. To overcome these issues, this paper propose a two-stage model named Mind-Animator, which achieves state-of-the-art performance on three public datasets. Specifically, during the fMRI-to-feature stage, we decouple semantic, structural, and motion features from fMRI through fMRI-vision-language tri-modal contrastive learning and sparse causal attention. In the feature-to-video stage, these features are merged to videos by an inflated Stable Diffusion. We substantiate that the reconstructed video dynamics are indeed derived from fMRI, rather than hallucinations of the generative model, through permutation tests. Additionally, the visualization of voxel-wise and ROI-wise importance maps confirms the neurobiological interpretability of our model.

Since the emergence of GPT-3, Large Language Models (LLMs) have caught the eyes of researchers, practitioners, and educators in the field of software engineering. However, there has been relatively little investigation regarding the performance of LLMs in assisting with requirements analysis and UML modeling. This paper explores how LLMs can assist novice analysts in creating three types of typical UML models: use case models, class diagrams, and sequence diagrams. For this purpose, we designed the modeling tasks of these three UML models for 45 undergraduate students who participated in a requirements modeling course, with the help of LLMs. By analyzing their project reports, we found that LLMs can assist undergraduate students as novice analysts in UML modeling tasks, but LLMs also have shortcomings and limitations that should be considered when using them., Comment: The 21st IEEE International Conference on Software Services Engineering (SSE)

Recent experimental discovery of fractional Chern insulators at zero magnetic field in moir\'e superlattices has sparked intense interests in bringing Landau level physics to flat Chern bands. In twisted MoTe$_2$ bilayers (tMoTe$_2$), recent theoretical and experimental studies have found three consecutive flat Chern bands at twist angle $\sim 2^\circ$. In this work, we investigate whether higher Landau level physics can be found in these consecutive Chern bands. At twist angles $2.00^\circ$ and $1.89^\circ$, we identify four consecutive $C = 1$ bands for the $K$ valley in tMoTe$_2$. By constructing Wannier functions directly from density functional theory (DFT) calculations, a six-orbital model is developed to describe the consecutive Chern bands, with the orbitals forming a honeycomb lattice. Exact diagonalization on top of Hartree-Fock calculations are carried out with the Wannier functions. Especially, when the second moir\'e miniband is half-filled, signatures of non-Abelian states are found. Our Wannier-based approach in modelling moir\'e superlattices is faithful to DFT wave functions and can serve as benchmarks for continuum models. The possibility of realizing non-Abelian anyons at zero magnetic field also opens up a new pathway for fault-tolerant quantum information processing.

Recently, Automated Vulnerability Localization (AVL) has attracted much attention, aiming to facilitate diagnosis by pinpointing the lines of code responsible for discovered vulnerabilities. Large Language Models (LLMs) have shown potential in various domains, yet their effectiveness in vulnerability localization remains underexplored. In this work, we perform the first comprehensive study of LLMs for AVL. Our investigation encompasses 10+ leading LLMs suitable for code analysis, including ChatGPT and various open-source models, across three architectural types: encoder-only, encoder-decoder, and decoder-only, with model sizes ranging from 60M to 16B parameters. We explore the efficacy of these LLMs using 4 distinct paradigms: zero-shot learning, one-shot learning, discriminative fine-tuning, and generative fine-tuning. Our evaluation framework is applied to the BigVul-based dataset for C/C++, and an additional dataset comprising smart contract vulnerabilities. The results demonstrate that discriminative fine-tuning of LLMs can significantly outperform existing learning-based methods for AVL, while other paradigms prove less effective or unexpectedly ineffective for the task. We also identify challenges related to input length and unidirectional context in fine-tuning processes for encoders and decoders. We then introduce two remedial strategies: the sliding window and the right-forward embedding, both of which substantially enhance performance. Furthermore, our findings highlight certain generalization capabilities of LLMs across Common Weakness Enumerations (CWEs) and different projects, indicating a promising pathway toward their practical application in vulnerability localization.

Shadow removal is a task aimed at erasing regional shadows present in images and reinstating visually pleasing natural scenes with consistent illumination. While recent deep learning techniques have demonstrated impressive performance in image shadow removal, their robustness against adversarial attacks remains largely unexplored. Furthermore, many existing attack frameworks typically allocate a uniform budget for perturbations across the entire input image, which may not be suitable for attacking shadow images. This is primarily due to the unique characteristic of spatially varying illumination within shadow images. In this paper, we propose a novel approach, called shadow-adaptive adversarial attack. Different from standard adversarial attacks, our attack budget is adjusted based on the pixel intensity in different regions of shadow images. Consequently, the optimized adversarial noise in the shadowed regions becomes visually less perceptible while permitting a greater tolerance for perturbations in non-shadow regions. The proposed shadow-adaptive attacks naturally align with the varying illumination distribution in shadow images, resulting in perturbations that are less conspicuous. Building on this, we conduct a comprehensive empirical evaluation of existing shadow removal methods, subjecting them to various levels of attack on publicly available datasets., Comment: Accepted to ICASSP 2024

Deep unfolding networks (DUN) have emerged as a popular iterative framework for accelerated magnetic resonance imaging (MRI) reconstruction. However, conventional DUN aims to reconstruct all the missing information within the entire null space in each iteration. Thus it could be challenging when dealing with highly ill-posed degradation, usually leading to unsatisfactory reconstruction. In this work, we propose a Progressive Divide-And-Conquer (PDAC) strategy, aiming to break down the subsampling process in the actual severe degradation and thus perform reconstruction sequentially. Starting from decomposing the original maximum-a-posteriori problem of accelerated MRI, we present a rigorous derivation of the proposed PDAC framework, which could be further unfolded into an end-to-end trainable network. Specifically, each iterative stage in PDAC focuses on recovering a distinct moderate degradation according to the decomposition. Furthermore, as part of the PDAC iteration, such decomposition is adaptively learned as an auxiliary task through a degradation predictor which provides an estimation of the decomposed sampling mask. Following this prediction, the sampling mask is further integrated via a severity conditioning module to ensure awareness of the degradation severity at each stage. Extensive experiments demonstrate that our proposed method achieves superior performance on the publicly available fastMRI and Stanford2D FSE datasets in both multi-coil and single-coil settings., Comment: Accepted to CVPR 2024

We present Recurrent Drafter (ReDrafter), an advanced speculative decoding approach that achieves state-of-the-art speedup for large language models (LLMs) inference. The performance gains are driven by three key aspects: (1) leveraging a recurrent neural network (RNN) as the draft model conditioning on LLM's hidden states, (2) applying a dynamic tree attention algorithm over beam search results to eliminate duplicated prefixes in candidate sequences, and (3) training through knowledge distillation from the LLM. ReDrafter accelerates Vicuna inference in MT-Bench by up to 3.5x with a PyTorch implementation on Nvidia H100 GPUs. To demonstrate its practicality in production environments, we integrate ReDrafter into TensorRT-LLM, reaching up to 2.5x speedup on H100 GPUs. We also validated its effectiveness for on-device applications by implementing the approach in MLX and benchmarking performance on Metal GPUs in Apple Silicon chips, achieving up to 2.3x speedup., Comment: 15 pages, 7 figures

In this work, we discuss building performant Multimodal Large Language Models (MLLMs). In particular, we study the importance of various architecture components and data choices. Through careful and comprehensive ablations of the image encoder, the vision language connector, and various pre-training data choices, we identified several crucial design lessons. For example, we demonstrate that for large-scale multimodal pre-training using a careful mix of image-caption, interleaved image-text, and text-only data is crucial for achieving state-of-the-art (SOTA) few-shot results across multiple benchmarks, compared to other published pre-training results. Further, we show that the image encoder together with image resolution and the image token count has substantial impact, while the vision-language connector design is of comparatively negligible importance. By scaling up the presented recipe, we build MM1, a family of multimodal models up to 30B parameters, including both dense models and mixture-of-experts (MoE) variants, that are SOTA in pre-training metrics and achieve competitive performance after supervised fine-tuning on a range of established multimodal benchmarks. Thanks to large-scale pre-training, MM1 enjoys appealing properties such as enhanced in-context learning, and multi-image reasoning, enabling few-shot chain-of-thought prompting.

Most models for weakly supervised video anomaly detection (WS-VAD) rely on multiple instance learning, aiming to distinguish normal and abnormal snippets without specifying the type of anomaly. However, the ambiguous nature of anomaly definitions across contexts may introduce inaccuracy in discriminating abnormal and normal events. To show the model what is anomalous, a novel framework is proposed to guide the learning of suspected anomalies from event prompts. Given a textual prompt dictionary of potential anomaly events and the captions generated from anomaly videos, the semantic anomaly similarity between them could be calculated to identify the suspected events for each video snippet. It enables a new multi-prompt learning process to constrain the visual-semantic features across all videos, as well as provides a new way to label pseudo anomalies for self-training. To demonstrate its effectiveness, comprehensive experiments and detailed ablation studies are conducted on four datasets, namely XD-Violence, UCF-Crime, TAD, and ShanghaiTech. Our proposed model outperforms most state-of-the-art methods in terms of AP or AUC (86.5\%, \hl{90.4}\%, 94.4\%, and 97.4\%). Furthermore, it shows promising performance in open-set and cross-dataset cases. The data, code, and models can be found at: \url{https://github.com/shiwoaz/lap}.