arXiv subjects

Yuxuan Sun

Publications and source records attributed to Yuxuan Sun.

At least 73 records · Page 4Linked to original sources

arXiv2024

PathGen-1.6M: 1.6 Million Pathology Image-text Pairs Generation through Multi-agent Collaboration

Vision Language Models (VLMs) like CLIP have attracted substantial attention in pathology, serving as backbones for applications such as zero-shot image classification and Whole Slide Image (WSI) analysis. Additionally, they can function as vision encoders when combined with large language models (LLMs) to support broader capabilities. Current efforts to train pathology VLMs rely on pathology image-text pairs from platforms like PubMed, YouTube, and Twitter, which provide limited, unscalable data with generally suboptimal image quality. In this work, we leverage large-scale WSI datasets like TCGA to extract numerous high-quality image patches. We then train a large multimodal model to generate captions for these images, creating PathGen-1.6M, a dataset containing 1.6 million high-quality image-caption pairs. Our approach involves multiple agent models collaborating to extract representative WSI patches, generating and refining captions to obtain high-quality image-text pairs. Extensive experiments show that integrating these generated pairs with existing datasets to train a pathology-specific CLIP model, PathGen-CLIP, significantly enhances its ability to analyze pathological images, with substantial improvements across nine pathology-related zero-shot image classification tasks and three whole-slide image tasks. Furthermore, we construct 200K instruction-tuning data based on PathGen-1.6M and integrate PathGen-CLIP with the Vicuna LLM to create more powerful multimodal models through instruction tuning. Overall, we provide a scalable pathway for high-quality data generation in pathology, paving the way for next-generation general pathology models.

cs.CV↗

arXiv2024

MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI

We introduce MMMU: a new benchmark designed to evaluate multimodal models on massive multi-discipline tasks demanding college-level subject knowledge and deliberate reasoning. MMMU includes 11.5K meticulously collected multimodal questions from college exams, quizzes, and textbooks, covering six core disciplines: Art & Design, Business, Science, Health & Medicine, Humanities & Social Science, and Tech & Engineering. These questions span 30 subjects and 183 subfields, comprising 30 highly heterogeneous image types, such as charts, diagrams, maps, tables, music sheets, and chemical structures. Unlike existing benchmarks, MMMU focuses on advanced perception and reasoning with domain-specific knowledge, challenging models to perform tasks akin to those faced by experts. The evaluation of 14 open-source LMMs as well as the proprietary GPT-4V(ision) and Gemini highlights the substantial challenges posed by MMMU. Even the advanced GPT-4V and Gemini Ultra only achieve accuracies of 56% and 59% respectively, indicating significant room for improvement. We believe MMMU will stimulate the community to build next-generation multimodal foundation models towards expert artificial general intelligence.

cs.CL↗

arXiv2024

Benchmarking PathCLIP for Pathology Image Analysis

Accurate image classification and retrieval are of importance for clinical diagnosis and treatment decision-making. The recent contrastive language-image pretraining (CLIP) model has shown remarkable proficiency in understanding natural images. Drawing inspiration from CLIP, PathCLIP is specifically designed for pathology image analysis, utilizing over 200,000 image and text pairs in training. While the performance the PathCLIP is impressive, its robustness under a wide range of image corruptions remains unknown. Therefore, we conduct an extensive evaluation to analyze the performance of PathCLIP on various corrupted images from the datasets of Osteosarcoma and WSSS4LUAD. In our experiments, we introduce seven corruption types including brightness, contrast, Gaussian blur, resolution, saturation, hue, and markup at four severity levels. Through experiments, we find that PathCLIP is relatively robustness to image corruptions and surpasses OpenAI-CLIP and PLIP in zero-shot classification. Among the seven corruptions, blur and resolution can cause server performance degradation of the PathCLIP. This indicates that ensuring the quality of images is crucial before conducting a clinical test. Additionally, we assess the robustness of PathCLIP in the task of image-image retrieval, revealing that PathCLIP performs less effectively than PLIP on Osteosarcoma but performs better on WSSS4LUAD under diverse corruptions. Overall, PathCLIP presents impressive zero-shot classification and retrieval performance for pathology images, but appropriate care needs to be taken when using it. We hope this study provides a qualitative impression of PathCLIP and helps understand its differences from other CLIP models.

cs.CV↗

arXiv2024

Task-Oriented Wireless Communications for Collaborative Perception in Intelligent Unmanned Systems

Collaborative Perception (CP) has shown great potential to achieve more holistic and reliable environmental perception in intelligent unmanned systems (IUSs). However, implementing CP still faces key challenges due to the characteristics of the CP task and the dynamics of wireless channels. In this article, a task-oriented wireless communication framework is proposed to jointly optimize the communication scheme and the CP procedure. We first propose channel-adaptive compression and robust fusion approaches to extract and exploit the most valuable semantic information under wireless communication constraints. We then propose a task-oriented distributed scheduling algorithm to identify the best collaborators for CP under dynamic environments. The main idea is learning while scheduling, where the collaboration utility is effectively learned with low computation and communication overhead. Case studies are carried out in connected autonomous driving scenarios to verify the proposed framework. Finally, we identify several future research directions.

cs.MA↗

arXiv2024

PathMMU: A Massive Multimodal Expert-Level Benchmark for Understanding and Reasoning in Pathology

The emergence of large multimodal models has unlocked remarkable potential in AI, particularly in pathology. However, the lack of specialized, high-quality benchmark impeded their development and precise evaluation. To address this, we introduce PathMMU, the largest and highest-quality expert-validated pathology benchmark for Large Multimodal Models (LMMs). It comprises 33,428 multimodal multi-choice questions and 24,067 images from various sources, each accompanied by an explanation for the correct answer. The construction of PathMMU harnesses GPT-4V's advanced capabilities, utilizing over 30,000 image-caption pairs to enrich captions and generate corresponding Q&As in a cascading process. Significantly, to maximize PathMMU's authority, we invite seven pathologists to scrutinize each question under strict standards in PathMMU's validation and test sets, while simultaneously setting an expert-level performance benchmark for PathMMU. We conduct extensive evaluations, including zero-shot assessments of 14 open-sourced and 4 closed-sourced LMMs and their robustness to image corruption. We also fine-tune representative LMMs to assess their adaptability to PathMMU. The empirical findings indicate that advanced LMMs struggle with the challenging PathMMU benchmark, with the top-performing LMM, GPT-4V, achieving only a 49.8% zero-shot performance, significantly lower than the 71.8% demonstrated by human pathologists. After fine-tuning, significantly smaller open-sourced LMMs can outperform GPT-4V but still fall short of the expertise shown by pathologists. We hope that the PathMMU will offer valuable insights and foster the development of more specialized, next-generation LMMs for pathology.

cs.CV↗

arXiv2024

MUFFIN: Curating Multi-Faceted Instructions for Improving Instruction-Following

In the realm of large language models (LLMs), enhancing instruction-following capability often involves curating expansive training data. This is achieved through two primary schemes: i) Scaling-Inputs: Amplifying (input, output) pairs per task instruction, aiming for better instruction adherence. ii) Scaling Input-Free Tasks: Enlarging tasks, each composed of an (instruction, output) pair (without requiring a separate input anymore). However, LLMs under Scaling-Inputs tend to be overly sensitive to inputs, leading to misinterpretation or non-compliance with instructions. Conversely, Scaling Input-Free Tasks demands a substantial number of tasks but is less effective in instruction following when dealing with instances in Scaling-Inputs. This work introduces MUFFIN, a new scheme of instruction-following dataset curation. Specifically, we automatically Scale Tasks per Input by diversifying these tasks with various input facets. Experimental results across four zero-shot benchmarks, spanning both Scaling-Inputs and Scaling Input-Free Tasks schemes, reveal that LLMs, at various scales, trained on MUFFIN generally demonstrate superior instruction-following capabilities compared to those trained on the two aforementioned schemes.

cs.CL↗

arXiv2024

PathAsst: A Generative Foundation AI Assistant Towards Artificial General Intelligence of Pathology

As advances in large language models (LLMs) and multimodal techniques continue to mature, the development of general-purpose multimodal large language models (MLLMs) has surged, offering significant applications in interpreting natural images. However, the field of pathology has largely remained untapped, particularly in gathering high-quality data and designing comprehensive model frameworks. To bridge the gap in pathology MLLMs, we present PathAsst, a multimodal generative foundation AI assistant to revolutionize diagnostic and predictive analytics in pathology. The development of PathAsst involves three pivotal steps: data acquisition, CLIP model adaptation, and the training of PathAsst's multimodal generative capabilities. Firstly, we collect over 207K high-quality pathology image-text pairs from authoritative sources. Leveraging the advanced power of ChatGPT, we generate over 180K instruction-following samples. Furthermore, we devise additional instruction-following data specifically tailored for invoking eight pathology-specific sub-models we prepared, allowing the PathAsst to effectively collaborate with these models, enhancing its diagnostic ability. Secondly, by leveraging the collected data, we construct PathCLIP, a pathology-dedicated CLIP, to enhance PathAsst's capabilities in interpreting pathology images. Finally, we integrate PathCLIP with the Vicuna-13b and utilize pathology-specific instruction-tuning data to enhance the multimodal generation capacity of PathAsst and bolster its synergistic interactions with sub-models. The experimental results of PathAsst show the potential of harnessing AI-powered generative foundation model to improve pathology diagnosis and treatment processes.

cs.CV↗

arXiv2024

Unleashing the Power of Prompt-driven Nucleus Instance Segmentation

Nucleus instance segmentation in histology images is crucial for a broad spectrum of clinical applications. Current dominant algorithms rely on regression of nuclear proxy maps. Distinguishing nucleus instances from the estimated maps requires carefully curated post-processing, which is error-prone and parameter-sensitive. Recently, the Segment Anything Model (SAM) has earned huge attention in medical image segmentation, owing to its impressive generalization ability and promptable property. Nevertheless, its potential on nucleus instance segmentation remains largely underexplored. In this paper, we present a novel prompt-driven framework that consists of a nucleus prompter and SAM for automatic nucleus instance segmentation. Specifically, the prompter learns to generate a unique point prompt for each nucleus while the SAM is fine-tuned to output the corresponding mask for the prompted nucleus. Furthermore, we propose the inclusion of adjacent nuclei as negative prompts to enhance the model's capability to identify overlapping nuclei. Without complicated post-processing, our proposed method sets a new state-of-the-art performance on three challenging benchmarks. Code is available at \url{github.com/windygoo/PromptNucSeg}

cs.CV↗

arXiv2024

Mobility Accelerates Learning: Convergence Analysis on Hierarchical Federated Learning in Vehicular Networks

Hierarchical federated learning (HFL) enables distributed training of models across multiple devices with the help of several edge servers and a cloud edge server in a privacy-preserving manner. In this paper, we consider HFL with highly mobile devices, mainly targeting at vehicular networks. Through convergence analysis, we show that mobility influences the convergence speed by both fusing the edge data and shuffling the edge models. While mobility is usually considered as a challenge from the perspective of communication, we prove that it increases the convergence speed of HFL with edge-level heterogeneous data, since more diverse data can be incorporated. Furthermore, we demonstrate that a higher speed leads to faster convergence, since it accelerates the fusion of data. Simulation results show that mobility increases the model accuracy of HFL by up to 15.1% when training a convolutional neural network on the CIFAR-10 dataset.

cs.LG↗

arXiv2024

Digital-SC: Digital Semantic Communication with Adaptive Network Split and Learned Non-Linear Quantization

Semantic communication, an intelligent communication paradigm that aims to transmit useful information in the semantic domain, is facilitated by deep learning techniques. Robust semantic features can be learned and transmitted in an analog fashion, but it poses new challenges to hardware, protocol, and encryption. In this paper, we propose a digital semantic communication system, which consists of an encoding network deployed on a resource-limited device and a decoding network deployed at the edge. To acquire better semantic representation for digital transmission, a novel non-linear quantization module is proposed to efficiently quantize semantic features with trainable quantization levels. Additionally, structured pruning is incorporated to reduce the dimension of the transmitted features. We also introduce a semantic learning loss (SLL) function to reduce semantic error. To adapt to various channel conditions and inputs under constraints of communication and computing resources, a policy network is designed to adaptively choose the split point and the dimension of the transmitted semantic features. Experiments using the CIFAR-10 and ImageNet dataset for image classification are employed to evaluate the proposed digital semantic communication network, and ablation studies are conducted to assess the proposed quantization module, structured pruning and SLL.

eess.SP↗

arXiv2023

Test-Time Training for Semantic Segmentation with Output Contrastive Loss

Although deep learning-based segmentation models have achieved impressive performance on public benchmarks, generalizing well to unseen environments remains a major challenge. To improve the model's generalization ability to the new domain during evaluation, the test-time training (TTT) is a challenging paradigm that adapts the source-pretrained model in an online fashion. Early efforts on TTT mainly focus on the image classification task. Directly extending these methods to semantic segmentation easily experiences unstable adaption due to segmentation's inherent characteristics, such as extreme class imbalance and complex decision spaces. To stabilize the adaptation process, we introduce contrastive loss (CL), known for its capability to learn robust and generalized representations. Nevertheless, the traditional CL operates in the representation space and cannot directly enhance predictions. In this paper, we resolve this limitation by adapting the CL to the output space, employing a high temperature, and simplifying the formulation, resulting in a straightforward yet effective loss function called Output Contrastive Loss (OCL). Our comprehensive experiments validate the efficacy of our approach across diverse evaluation scenarios. Notably, our method excels even when applied to models initially pre-trained using domain adaptation methods on test domain data, showcasing its resilience and adaptability.\footnote{Code and more information could be found at~ \url{https://github.com/dazhangyu123/OCL}}

cs.CV↗

arXiv2023

Optimizing Task-Specific Timeliness With Edge-Assisted Scheduling for Status Update

Intelligent real-time applications, such as video surveillance, demand intensive computation to extract status information from raw sensing data. This poses a substantial challenge in orchestrating computation and communication resources to provide fresh status information. In this paper, we consider a scenario where multiple energy-constrained devices served by an edge server. To extract status information, each device can either do the computation locally or offload it to the edge server. A scheduling policy is needed to determine when and where to compute for each device, taking into account communication and computation capabilities, as well as task-specific timeliness requirements. To that end, we first model the timeliness requirements as general penalty functions of Age of Information (AoI). A convex optimization problem is formulated to provide a lower bound of the minimum AoI penalty given system parameters. Using KKT conditions, we proposed a novel scheduling policy which evaluates status update priorities based on communication and computation delays and task-specific timeliness requirements. The proposed policy is applied to an object tracking application and carried out on a large video dataset. Simulation results show that our policy improves tracking accuracy compared with scheduling policies based on video content information.

cs.NI↗

arXiv2023

Age of Information Guaranteed Scheduling for Asynchronous Status Updates in Collaborative Perception

We consider collaborative perception (CP) systems where a fusion center monitors various regions by multiple sources. The center has different age of information (AoI) constraints for different regions. Multi-view sensing data for a region generated by sources can be fused by the center for a reliable representation of the region. To ensure accurate perception, differences between generation time of asynchronous status updates for CP fusion should not exceed a certain threshold. An algorithm named scheduling for CP with asynchronous status updates (SCPA) is proposed to minimize the number of required channels and subject to AoI constraints with asynchronous status updates. SCPA first identifies a set of sources that can satisfy the constraints with minimum updating rates. It then chooses scheduling intervals and offsets for the sources such that the number of required channels is optimized. According to numerical results, the number of channels required by SCPA can reach only 12% more than a derived lower bound.

cs.IT↗

arXiv2023

A Grouping-based Scheduler for Efficient Channel Utilization under Age of Information Constraints

We consider a status information updating system where a fusion center collects the status information from a large number of sources and each of them has its own age of information (AoI) constraints. A novel grouping-based scheduler is proposed to solve this complex large-scale problem by dividing the sources into different scheduling groups. The problem is then transformed into deriving the optimal grouping scheme. A two-step grouping algorithm (TGA) is proposed: 1) Given AoI constraints, we first identify the sources with harmonic AoI constraints, then design a fast grouping method and an optimal scheduler for these sources. Under harmonic AoI constraints, each constraint is divisible by the smallest one and the sum of reciprocals of the constraints with the same value is divisible by the reciprocal of the smallest one. 2) For the other sources without such a special property, we pack the sources which can be scheduled together with minimum update rates into the same group. Simulations show the channel usage of the proposed TGA is significantly reduced as compared to a recent work and is 0.42% larger than a derived lower bound when the number of sources is large.

cs.IT↗

arXiv2023

Multimodal Question Answering for Unified Information Extraction

Multimodal information extraction (MIE) aims to extract structured information from unstructured multimedia content. Due to the diversity of tasks and settings, most current MIE models are task-specific and data-intensive, which limits their generalization to real-world scenarios with diverse task requirements and limited labeled data. To address these issues, we propose a novel multimodal question answering (MQA) framework to unify three MIE tasks by reformulating them into a unified span extraction and multi-choice QA pipeline. Extensive experiments on six datasets show that: 1) Our MQA framework consistently and significantly improves the performances of various off-the-shelf large multimodal models (LMM) on MIE tasks, compared to vanilla prompting. 2) In the zero-shot setting, MQA outperforms previous state-of-the-art baselines by a large margin. In addition, the effectiveness of our framework can successfully transfer to the few-shot setting, enhancing LMMs on a scale of 10B parameters to be competitive or outperform much larger language models such as ChatGPT and GPT-4. Our MQA framework can serve as a general principle of utilizing LMMs to better solve MIE and potentially other downstream multimodal tasks.

cs.CL↗

arXiv2023

Programmable Photonic Simulator for Spin Glass Models

Spin glasses featured by frustrated interactions and metastable states have important applications in chemistry, material sciences and artificial neural networks. However, the solution of the spin glass models is hindered by the computational complexity that exponentially increases with the sample size. Photonic Ising machines based on spatial light modulation can speed up the calculation by obtaining the Hamiltonian from the modulated light intensity. However, the large-scale generalization to various spin couplings and higher dimensions is still elusive. Here, we develop a Fourier-mask method to program the spin couplings in photonic Ising machines. We observe the phase transition of the two-dimensional Mattis model and the J$\mathrm{_1}$-J$\mathrm{_2}$ model and study the critical phenomena. We also demonstrate that the three-dimensional Ising model, which has not been analytically solved, can be effectively constructed and simulated in two-dimensional lattices with Fourier masks. Our strategy provides a flexible route to tuning couplings and dimensions of statistical spin models, and improves the applicability of optical simulation in neural networks and combinatorial optimization problems.

cond-mat.dis-nn↗

arXiv2023

A Data Source for Reasoning Embodied Agents

Recent progress in using machine learning models for reasoning tasks has been driven by novel model architectures, large-scale pre-training protocols, and dedicated reasoning datasets for fine-tuning. In this work, to further pursue these advances, we introduce a new data generator for machine reasoning that integrates with an embodied agent. The generated data consists of templated text queries and answers, matched with world-states encoded into a database. The world-states are a result of both world dynamics and the actions of the agent. We show the results of several baseline models on instantiations of train sets. These include pre-trained language models fine-tuned on a text-formatted representation of the database, and graph-structured Transformers operating on a knowledge-graph representation of the database. We find that these models can answer some questions about the world-state, but struggle with others. These results hint at new research directions in designing neural reasoning models and database representations. Code to generate the data will be released at github.com/facebookresearch/neuralmemory

cs.LG↗

arXiv2023

Deep Joint Source-Channel Coding for Semantic Communications

Semantic communications is considered as a promising technology to increase the efficiency of next-generation communication systems, particularly targeting human-machine and machine-type communications. In contrast to the source-agnostic approach of conventional wireless communication systems, semantic communication seeks to ensure that only the relevant information for the underlying task is communicated to the receiver. Considering that most semantic communication applications have strict latency, bandwidth, and power constraints, a prominent approach is to model them as a joint source-channel coding (JSCC) problem. Although JSCC has been a long-standing open problem in communication and coding theory, remarkable performance gains have been shown recently over existing separate source and channel coding systems, particularly in low-latency and low-power scenarios. Recent progress is thanks to the adoption of deep learning techniques for joint source-channel code design that outperform the concatenation of state-of-the-art compression and channel coding schemes, which are results of decades-long research efforts. In this article, we present an adaptive deep learning based JSCC (DeepJSCC) architecture for semantic communications, introduce its design principles, highlight its benefits, and outline future research challenges that lie ahead.

cs.IT↗