Searcharxiv⌕ Search

arXiv subjects

Yichi Zhang

Publications and source records attributed to Yichi Zhang.

At least 253 records · Page 14Linked to original sources
arXiv2024

Segment Anything Model for Medical Image Segmentation: Current Applications and Future Directions

Due to the inherent flexibility of prompting, foundation models have emerged as the predominant force in the fields of natural language processing and computer vision. The recent introduction of the Segment Anything Model (SAM) signifies a noteworthy expansion of the prompt-driven paradigm into the domain of image segmentation, thereby introducing a plethora of previously unexplored capabilities. However, the viability of its application to medical image segmentation remains uncertain, given the substantial distinctions between natural and medical images. In this work, we provide a comprehensive overview of recent endeavors aimed at extending the efficacy of SAM to medical image segmentation tasks, encompassing both empirical benchmarking and methodological adaptations. Additionally, we explore potential avenues for future research directions in SAM's role within medical image segmentation. While direct application of SAM to medical image segmentation does not yield satisfactory performance on multi-modal and multi-target medical datasets so far, numerous insights gleaned from these efforts serve as valuable guidance for shaping the trajectory of foundational models in the realm of medical image analysis. To support ongoing research endeavors, we maintain an active repository that contains an up-to-date paper list and a succinct summary of open-source projects at https://github.com/YichiZhang98/SAM4MIS.

eess.IV↗
arXiv2023

Towards End-to-End Embodied Decision Making via Multi-modal Large Language Model: Explorations with GPT4-Vision and Beyond

In this study, we explore the potential of Multimodal Large Language Models (MLLMs) in improving embodied decision-making processes for agents. While Large Language Models (LLMs) have been widely used due to their advanced reasoning skills and vast world knowledge, MLLMs like GPT4-Vision offer enhanced visual understanding and reasoning capabilities. We investigate whether state-of-the-art MLLMs can handle embodied decision-making in an end-to-end manner and whether collaborations between LLMs and MLLMs can enhance decision-making. To address these questions, we introduce a new benchmark called PCA-EVAL, which evaluates embodied decision-making from the perspectives of Perception, Cognition, and Action. Additionally, we propose HOLMES, a multi-agent cooperation framework that allows LLMs to leverage MLLMs and APIs to gather multimodal information for informed decision-making. We compare end-to-end embodied decision-making and HOLMES on our benchmark and find that the GPT4-Vision model demonstrates strong end-to-end embodied decision-making abilities, outperforming GPT4-HOLMES in terms of average decision accuracy (+3%). However, this performance is exclusive to the latest GPT4-Vision model, surpassing the open-source state-of-the-art MLLM by 26%. Our results indicate that powerful MLLMs like GPT4-Vision hold promise for decision-making in embodied agents, offering new avenues for MLLM research. Code and data are open at https://github.com/pkunlp-icler/PCA-EVAL/.

cs.AI↗
arXiv2023

Eliciting Honest Information From Authors Using Sequential Review

In the setting of conference peer review, the conference aims to accept high-quality papers and reject low-quality papers based on noisy review scores. A recent work proposes the isotonic mechanism, which can elicit the ranking of paper qualities from an author with multiple submissions to help improve the conference's decisions. However, the isotonic mechanism relies on the assumption that the author's utility is both an increasing and a convex function with respect to the review score, which is often violated in peer review settings (e.g.~when authors aim to maximize the number of accepted papers). In this paper, we propose a sequential review mechanism that can truthfully elicit the ranking information from authors while only assuming the agent's utility is increasing with respect to the true quality of her accepted papers. The key idea is to review the papers of an author in a sequence based on the provided ranking and conditioning the review of the next paper on the review scores of the previous papers. Advantages of the sequential review mechanism include 1) eliciting truthful ranking information in a more realistic setting than prior work; 2) improving the quality of accepted papers, reducing the reviewing workload and increasing the average quality of papers being reviewed; 3) incentivizing authors to write fewer papers of higher quality.

cs.GT↗
arXiv2023

Learning with Limited Annotations: A Survey on Deep Semi-Supervised Learning for Medical Image Segmentation

Medical image segmentation is a fundamental and critical step in many image-guided clinical approaches. Recent success of deep learning-based segmentation methods usually relies on a large amount of labeled data, which is particularly difficult and costly to obtain especially in the medical imaging domain where only experts can provide reliable and accurate annotations. Semi-supervised learning has emerged as an appealing strategy and been widely applied to medical image segmentation tasks to train deep models with limited annotations. In this paper, we present a comprehensive review of recently proposed semi-supervised learning methods for medical image segmentation and summarized both the technical novelties and empirical results. Furthermore, we analyze and discuss the limitations and several unsolved problems of existing approaches. We hope this review could inspire the research community to explore solutions for this challenge and further promote the developments in medical image segmentation field.

cs.CV↗
arXiv2023

SPHINX: The Joint Mixing of Weights, Tasks, and Visual Embeddings for Multi-modal Large Language Models

We present SPHINX, a versatile multi-modal large language model (MLLM) with a joint mixing of model weights, tuning tasks, and visual embeddings. First, for stronger vision-language alignment, we unfreeze the large language model (LLM) during pre-training, and introduce a weight mix strategy between LLMs trained by real-world and synthetic data. By directly integrating the weights from two domains, the mixed LLM can efficiently incorporate diverse semantics with favorable robustness. Then, to enable multi-purpose capabilities, we mix a variety of tasks for joint visual instruction tuning, and design task-specific instructions to avoid inter-task conflict. In addition to the basic visual question answering, we include more challenging tasks such as region-level understanding, caption grounding, document layout detection, and human pose estimation, contributing to mutual enhancement over different scenarios. Additionally, we propose to extract comprehensive visual embeddings from various network architectures, pre-training paradigms, and information granularity, providing language models with more robust image representations. Based on our proposed joint mixing, SPHINX exhibits superior multi-modal understanding capabilities on a wide range of applications. On top of this, we further propose an efficient strategy aiming to better capture fine-grained appearances of high-resolution images. With a mixing of different scales and high-resolution sub-images, SPHINX attains exceptional visual parsing and reasoning performance on existing evaluation benchmarks. We hope our work may cast a light on the exploration of joint mixing in future MLLM research. Code is released at https://github.com/Alpha-VLLM/LLaMA2-Accessory.

cs.CV↗
arXiv2023

Can Foundation Models Watch, Talk and Guide You Step by Step to Make a Cake?

Despite tremendous advances in AI, it remains a significant challenge to develop interactive task guidance systems that can offer situated, personalized guidance and assist humans in various tasks. These systems need to have a sophisticated understanding of the user as well as the environment, and make timely accurate decisions on when and what to say. To address this issue, we created a new multimodal benchmark dataset, Watch, Talk and Guide (WTaG) based on natural interaction between a human user and a human instructor. We further proposed two tasks: User and Environment Understanding, and Instructor Decision Making. We leveraged several foundation models to study to what extent these models can be quickly adapted to perceptually enabled task guidance. Our quantitative, qualitative, and human evaluation results show that these models can demonstrate fair performances in some cases with no task-specific training, but a fast and reliable adaptation remains a significant challenge. Our benchmark and baselines will provide a stepping stone for future work on situated task guidance.

cs.AI↗
arXiv2023

Grounding Visual Illusions in Language: Do Vision-Language Models Perceive Illusions Like Humans?

Vision-Language Models (VLMs) are trained on vast amounts of data captured by humans emulating our understanding of the world. However, known as visual illusions, human's perception of reality isn't always faithful to the physical world. This raises a key question: do VLMs have the similar kind of illusions as humans do, or do they faithfully learn to represent reality? To investigate this question, we build a dataset containing five types of visual illusions and formulate four tasks to examine visual illusions in state-of-the-art VLMs. Our findings have shown that although the overall alignment is low, larger models are closer to human perception and more susceptible to visual illusions. Our dataset and initial findings will promote a better understanding of visual illusions in humans and machines and provide a stepping stone for future computational models that can better align humans and machines in perceiving and communicating about the shared visual world. The code and data are available at https://github.com/vl-illusion/dataset.

cs.AI↗
arXiv2023

Revisiting the Evaluation of Image Synthesis with GANs

A good metric, which promises a reliable comparison between solutions, is essential for any well-defined task. Unlike most vision tasks that have per-sample ground-truth, image synthesis tasks target generating unseen data and hence are usually evaluated through a distributional distance between one set of real samples and another set of generated samples. This study presents an empirical investigation into the evaluation of synthesis performance, with generative adversarial networks (GANs) as a representative of generative models. In particular, we make in-depth analyses of various factors, including how to represent a data point in the representation space, how to calculate a fair distance using selected samples, and how many instances to use from each set. Extensive experiments conducted on multiple datasets and settings reveal several important findings. Firstly, a group of models that include both CNN-based and ViT-based architectures serve as reliable and robust feature extractors for measurement evaluation. Secondly, Centered Kernel Alignment (CKA) provides a better comparison across various extractors and hierarchical layers in one model. Finally, CKA is more sample-efficient and enjoys better agreement with human judgment in characterizing the similarity between two internal data correlations. These findings contribute to the development of a new measurement system, which enables a consistent and reliable re-evaluation of current state-of-the-art generative models.

cs.CV↗
arXiv2023

Competitive Ensembling Teacher-Student Framework for Semi-Supervised Left Atrium MRI Segmentation

Semi-supervised learning has greatly advanced medical image segmentation since it effectively alleviates the need of acquiring abundant annotations from experts and utilizes unlabeled data which is much easier to acquire. Among existing perturbed consistency learning methods, mean-teacher model serves as a standard baseline for semi-supervised medical image segmentation. In this paper, we present a simple yet efficient competitive ensembling teacher student framework for semi-supervised for left atrium segmentation from 3D MR images, in which two student models with different task-level disturbances are introduced to learn mutually, while a competitive ensembling strategy is performed to ensemble more reliable information to teacher model. Different from the one-way transfer between teacher and student models, our framework facilitates the collaborative learning procedure of different student models with the guidance of teacher model and motivates different training networks for a competitive learning and ensembling procedure to achieve better performance. We evaluate our proposed method on the public Left Atrium (LA) dataset and it obtains impressive performance gains by exploiting the unlabeled data effectively and outperforms several existing semi-supervised methods.

cs.CV↗
arXiv2023

How Robust is Google's Bard to Adversarial Image Attacks?

Multimodal Large Language Models (MLLMs) that integrate text and other modalities (especially vision) have achieved unprecedented performance in various multimodal tasks. However, due to the unsolved adversarial robustness problem of vision models, MLLMs can have more severe safety and security risks by introducing the vision inputs. In this work, we study the adversarial robustness of Google's Bard, a competitive chatbot to ChatGPT that released its multimodal capability recently, to better understand the vulnerabilities of commercial MLLMs. By attacking white-box surrogate vision encoders or MLLMs, the generated adversarial examples can mislead Bard to output wrong image descriptions with a 22% success rate based solely on the transferability. We show that the adversarial examples can also attack other MLLMs, e.g., a 26% attack success rate against Bing Chat and a 86% attack success rate against ERNIE bot. Moreover, we identify two defense mechanisms of Bard, including face detection and toxicity detection of images. We design corresponding attacks to evade these defenses, demonstrating that the current defenses of Bard are also vulnerable. We hope this work can deepen our understanding on the robustness of MLLMs and facilitate future research on defenses. Our code is available at https://github.com/thu-ml/Attack-Bard. Update: GPT-4V is available at October 2023. We further evaluate its robustness under the same set of adversarial examples, achieving a 45% attack success rate.

cs.CV↗
arXiv2023

Emergence of ferromagnetism at the onset of moiré Kondo breakdown

The interaction of a lattice of localized magnetic moments with a sea of conduction electrons in Kondo lattice models induces rich quantum phases of matter, such as Fermi liquids with heavily renormalized electronic quasiparticles, quantum critical non-Fermi liquid metals and unconventional superconductors, among others. The recent demonstration of moiré Kondo lattices has opened the door to investigate the Kondo problem with continuously tunable parameters. Although a heavy Fermi liquid phase has been identified in moiré Kondo lattices, the magnetic phases and Kondo breakdown transitions remain unexplored. Here we report a density-tuned Kondo destruction in AB-stacked MoTe2/WSe2 moiré bilayers by combining magneto transport and optical studies. As the itinerant carrier density decreases, the Kondo temperature decreases. At a critical density, we observe a heavy Fermi liquid to insulator transition, and a nearly concomitant emergence of ferromagnetic order. The observation is consistent with the scenario of a ferromagnetic Anderson insulator and suppression of the Kondo screening effect. Our results pave the path for inducing other exotic quantum phase transitions in moiré Kondo lattices.

cond-mat.str-el↗
arXiv2023

PINNacle: A Comprehensive Benchmark of Physics-Informed Neural Networks for Solving PDEs

While significant progress has been made on Physics-Informed Neural Networks (PINNs), a comprehensive comparison of these methods across a wide range of Partial Differential Equations (PDEs) is still lacking. This study introduces PINNacle, a benchmarking tool designed to fill this gap. PINNacle provides a diverse dataset, comprising over 20 distinct PDEs from various domains, including heat conduction, fluid dynamics, biology, and electromagnetics. These PDEs encapsulate key challenges inherent to real-world problems, such as complex geometry, multi-scale phenomena, nonlinearity, and high dimensionality. PINNacle also offers a user-friendly toolbox, incorporating about 10 state-of-the-art PINN methods for systematic evaluation and comparison. We have conducted extensive experiments with these methods, offering insights into their strengths and weaknesses. In addition to providing a standardized means of assessing performance, PINNacle also offers an in-depth analysis to guide future research, particularly in areas such as domain decomposition methods and loss reweighting for handling multi-scale problems and complex geometry. To the best of our knowledge, it is the largest benchmark with a diverse and comprehensive evaluation that will undoubtedly foster further research in PINNs.

cs.LG↗
arXiv2023

Reconstructing microstructures from statistical descriptors using neural cellular automata

The problem of generating microstructures of complex materials in silico has been approached from various directions including simulation, Markov, deep learning and descriptor-based approaches. This work presents a hybrid method that is inspired by all four categories and has interesting scalability properties. A neural cellular automaton is trained to evolve microstructures based on local information. Unlike most machine learning-based approaches, it does not directly require a data set of reference micrographs, but is trained from statistical microstructure descriptors that can stem from a single reference. This means that the training cost scales only with the complexity of the structure and associated descriptors. Since the size of the reconstructed structures can be set during inference, even extremely large structures can be efficiently generated. Similarly, the method is very efficient if many structures are to be reconstructed from the same descriptor for statistical evaluations. The method is formulated and discussed in detail by means of various numerical experiments, demonstrating its utility and scalability.

cond-mat.mtrl-sci↗
arXiv2023

Robust Detection of Lead-Lag Relationships in Lagged Multi-Factor Models

In multivariate time series systems, key insights can be obtained by discovering lead-lag relationships inherent in the data, which refer to the dependence between two time series shifted in time relative to one another, and which can be leveraged for the purposes of control, forecasting or clustering. We develop a clustering-driven methodology for robust detection of lead-lag relationships in lagged multi-factor models. Within our framework, the envisioned pipeline takes as input a set of time series, and creates an enlarged universe of extracted subsequence time series from each input time series, via a sliding window approach. This is then followed by an application of various clustering techniques, (such as k-means++ and spectral clustering), employing a variety of pairwise similarity measures, including nonlinear ones. Once the clusters have been extracted, lead-lag estimates across clusters are robustly aggregated to enhance the identification of the consistent relationships in the original universe. We establish connections to the multireference alignment problem for both the homogeneous and heterogeneous settings. Since multivariate time series are ubiquitous in a wide range of domains, we demonstrate that our method is not only able to robustly detect lead-lag relationships in financial markets, but can also yield insightful results when applied to an environmental data set.

stat.ML↗
arXiv2023

Understanding the Robustness of 3D Object Detection with Bird's-Eye-View Representations in Autonomous Driving

3D object detection is an essential perception task in autonomous driving to understand the environments. The Bird's-Eye-View (BEV) representations have significantly improved the performance of 3D detectors with camera inputs on popular benchmarks. However, there still lacks a systematic understanding of the robustness of these vision-dependent BEV models, which is closely related to the safety of autonomous driving systems. In this paper, we evaluate the natural and adversarial robustness of various representative models under extensive settings, to fully understand their behaviors influenced by explicit BEV features compared with those without BEV. In addition to the classic settings, we propose a 3D consistent patch attack by applying adversarial patches in the 3D space to guarantee the spatiotemporal consistency, which is more realistic for the scenario of autonomous driving. With substantial experiments, we draw several findings: 1) BEV models tend to be more stable than previous methods under different natural conditions and common corruptions due to the expressive spatial representations; 2) BEV models are more vulnerable to adversarial noises, mainly caused by the redundant BEV features; 3) Camera-LiDAR fusion models have superior performance under different settings with multi-modal inputs, but BEV fusion model is still vulnerable to adversarial noises of both point cloud and image. These findings alert the safety issue in the applications of BEV detectors and could facilitate the development of more robust models.

cs.CV↗
arXiv2023

Federated Learning with Quantum Secure Aggregation

This article illustrates a novel Quantum Secure Aggregation (QSA) scheme that is designed to provide highly secure and efficient aggregation of local model parameters for federated learning. The scheme is secure in protecting private model parameters from being disclosed to semi-honest attackers by utilizing quantum bits i.e. qubits to represent model parameters. The proposed security mechanism ensures that any attempts to eavesdrop private model parameters can be immediately detected and stopped. The scheme is also efficient in terms of the low computational complexity of transmitting and aggregating model parameters through entangled qubits. Benefits of the proposed QSA scheme are showcased in a horizontal federated learning setting in which both a centralized and decentralized architectures are taken into account. It was empirically demonstrated that the proposed QSA can be readily applied to aggregate different types of local models including logistic regression (LR), convolutional neural networks (CNN) as well as quantum neural network (QNN), indicating the versatility of the QSA scheme. Performances of global models are improved to various extents with respect to local models obtained by individual participants, while no private model parameters are disclosed to semi-honest adversaries.

quant-ph↗
arXiv2023

Dynamic Time Warping for Lead-Lag Relationships in Lagged Multi-Factor Models

In multivariate time series systems, lead-lag relationships reveal dependencies between time series when they are shifted in time relative to each other. Uncovering such relationships is valuable in downstream tasks, such as control, forecasting, and clustering. By understanding the temporal dependencies between different time series, one can better comprehend the complex interactions and patterns within the system. We develop a cluster-driven methodology based on dynamic time warping for robust detection of lead-lag relationships in lagged multi-factor models. We establish connections to the multireference alignment problem for both the homogeneous and heterogeneous settings. Since multivariate time series are ubiquitous in a wide range of domains, we demonstrate that our algorithm is able to robustly detect lead-lag relationships in financial markets, which can be subsequently leveraged in trading strategies with significant economic benefits.

q-fin.ST↗
arXiv2023

MACO: A Modality Adversarial and Contrastive Framework for Modality-missing Multi-modal Knowledge Graph Completion

Recent years have seen significant advancements in multi-modal knowledge graph completion (MMKGC). MMKGC enhances knowledge graph completion (KGC) by integrating multi-modal entity information, thereby facilitating the discovery of unobserved triples in the large-scale knowledge graphs (KGs). Nevertheless, existing methods emphasize the design of elegant KGC models to facilitate modality interaction, neglecting the real-life problem of missing modalities in KGs. The missing modality information impedes modal interaction, consequently undermining the model's performance. In this paper, we propose a modality adversarial and contrastive framework (MACO) to solve the modality-missing problem in MMKGC. MACO trains a generator and discriminator adversarially to generate missing modality features that can be incorporated into the MMKGC model. Meanwhile, we design a cross-modal contrastive loss to improve the performance of the generator. Experiments on public benchmarks with further explorations demonstrate that MACO could achieve state-of-the-art results and serve as a versatile framework to bolster various MMKGC models. Our code and benchmark data are available at https://github.com/zjukg/MACO.

cs.CL↗