skip to main content

Search for: All records

Creators/Authors contains: "Liu, Han"

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. Visual Question Answering (VQA) is a fundamental task in computer vision and natural language process fields. Although the “pre-training & finetuning” learning paradigm significantly improves the VQA performance, the adversarial robustness of such a learning paradigm has not been explored. In this paper, we delve into a new problem: using a pre-trained multimodal source model to create adversarial image-text pairs and then transferring them to attack the target VQA models. Correspondingly, we propose a novel VQATTACK model, which can iteratively generate both im- age and text perturbations with the designed modules: the large language model (LLM)-enhanced image attack and the cross-modal joint attack module. At each iteration, the LLM-enhanced image attack module first optimizes the latent representation-based loss to generate feature-level image perturbations. Then it incorporates an LLM to further enhance the image perturbations by optimizing the designed masked answer anti-recovery loss. The cross-modal joint attack module will be triggered at a specific iteration, which updates the image and text perturbations sequentially. Notably, the text perturbation updates are based on both the learned gradients in the word embedding space and word synonym-based substitution. Experimental results on two VQA datasets with five validated models demonstrate the effectiveness of the proposed VQATTACK in the transferable attack setting, compared with state-of-the-art baselines. This work revealsa significant blind spot in the “pre-training & fine-tuning” paradigm on VQA tasks. The source code can be found in the link

    more » « less
    Free, publicly-accessible full text available March 25, 2025
  2. Abstract

    Architected materials design across orders of magnitude length scale intrigues exceptional mechanical responses nonexistent in their natural bulk state. However, the so‐termed mechanical metamaterials, when scaling bottom down to the atomistic or microparticle level, remain largely unexplored and conventionally fall out of their coarse‐resolution, ordered‐pattern design space. Here, combining high‐throughput molecular dynamics (MD) simulations and machine learning (ML) strategies, some intriguing atomistic families of disordered mechanical metamaterials are discovered, as fabricated by melt quenching and exemplified herein by lightweight‐yet‐stiff cellular materials featuring a theoretical limit of linear stiffness–density scaling, whose structural disorder—rather than order—is key to reduce the scaling exponent and is simply controlled by the bonding interactions and their directionality that enable flexible tunability experimentally. Importantly, a systematic navigation in the forcefield landscape reveals that, in‐between directional and non‐directional bonding such as covalent and ionic bonds, modest bond directionality is most likely to promotes disordered packing of polyhedral, stretching‐dominated structures responsible for the formation of metamaterials. This work pioneers a bottom‐down atomistic scheme to design mechanical metamaterials formatted disorderly, unlocking a largely untapped field in leveraging structural disorder in devising metamaterials atomistically and, potentially, generic to conventional upscaled designs.

    more » « less
    Free, publicly-accessible full text available January 25, 2025
  3. Numerical simulations have revolutionized material design. However, although simulations excel at mapping an input material to its output property, their direct application to inverse design has traditionally been limited by their high computing cost and lack of differentiability. Here, taking the example of the inverse design of a porous matrix featuring targeted sorption isotherm, we introduce a computational inverse design framework that addresses these challenges, by programming differentiable simulation on TensorFlow platform that leverages automated end-to-end differentiation. Thanks to its differentiability, the simulation is used to directly train a deep generative model, which outputs an optimal porous matrix based on an arbitrary input sorption isotherm curve. Importantly, this inverse design pipeline leverages the power of tensor processing units (TPU)—an emerging family of dedicated chips, which, although they are specialized in deep learning, are flexible enough for intensive scientific simulations. This approach holds promise to accelerate inverse materials design. 
    more » « less
    Free, publicly-accessible full text available December 1, 2024
  4. Free, publicly-accessible full text available October 1, 2024
  5. Abstract Background

    Computational drug repurposing is a cost- and time-efficient approach that aims to identify new therapeutic targets or diseases (indications) of existing drugs/compounds. It is especially critical for emerging and/or orphan diseases due to its cheaper investment and shorter research cycle compared with traditional wet-lab drug discovery approaches. However, the underlying mechanisms of action (MOAs) between repurposed drugs and their target diseases remain largely unknown, which is still a main obstacle for computational drug repurposing methods to be widely adopted in clinical settings.


    In this work, we propose KGML-xDTD: a Knowledge Graph–based Machine Learning framework for explainably predicting Drugs Treating Diseases. It is a 2-module framework that not only predicts the treatment probabilities between drugs/compounds and diseases but also biologically explains them via knowledge graph (KG) path-based, testable MOAs. We leverage knowledge-and-publication–based information to extract biologically meaningful “demonstration paths” as the intermediate guidance in the Graph-based Reinforcement Learning (GRL) path-finding process. Comprehensive experiments and case study analyses show that the proposed framework can achieve state-of-the-art performance in both predictions of drug repurposing and recapitulation of human-curated drug MOA paths.


    KGML-xDTD is the first model framework that can offer KG path explanations for drug repurposing predictions by leveraging the combination of prediction outcomes and existing biological knowledge and publications. We believe it can effectively reduce “black-box” concerns and increase prediction confidence for drug repurposing based on predicted path-based explanations and further accelerate the process of drug discovery for emerging diseases.

    more » « less
  6. A graph-based machine learning model is built to predict atom dynamics from their static structure, which, in turn, unveils the predictive power of static structure in dynamical evolution of disordered phases. 
    more » « less
    Free, publicly-accessible full text available August 29, 2024
  7. Tiny machine learning (TinyML) is an essential component of emerging smart microcontrollers (MCUs). However, the protection of the intellectual property (IP) of the model is an increasing concern due to the lack of desktop/server-grade resources on these power-constrained devices. In this paper, we propose STML, a system and algorithm co-design to Secure IP of TinyML on MCUs with ARM TrustZone. Our design jointly optimizes memory utilization and latency while ensuring the security and accuracy of emerging models. We implemented a prototype and benchmarked with 7 models, demonstrating STML reduces 40% of model protection runtime overhead on average. 
    more » « less
    Free, publicly-accessible full text available July 9, 2024
  8. Free, publicly-accessible full text available June 1, 2024
  9. The field of text-to-image generation has made remarkable strides in creating high-fidelity and photorealistic images. As this technology gains popularity, there is a growing concern about its potential security risks. However, there has been limited exploration into the robustness of these models from an adversarial perspective. Existing research has primarily focused on untargeted settings, and lacks holistic consideration for reliability (attack success rate) and stealthiness (imperceptibility). In this paper, we propose RIATIG, a reliable and imperceptible adversarial attack against text-to-image models via inconspicuous examples. By formulating the example crafting as an optimization process and solving it using a genetic-based method, our proposed attack can generate imperceptible prompts for text-to-image generation models in a reliable way. Evaluation of six popular text-to-image generation models demonstrates the efficiency and stealthiness of our attack in both white-box and black-box settings. To allow the community to build on top of our findings, we’ve made the artifacts available. 
    more » « less
    Free, publicly-accessible full text available June 1, 2024
  10. Algorithmic case-based decision support provides examples to help human make sense of predicted labels and aid human in decision-making tasks. Despite the promising performance of supervised learning, representations learned by supervised models may not align well with human intuitions: what models consider as similar examples can be perceived as distinct by humans. As a result, they have limited effectiveness in case-based decision support. In this work, we incorporate ideas from metric learning with supervised learning to examine the importance of alignment for effective decision support. In addition to instance-level labels, we use human-provided triplet judgments to learn human-compatible decision-focused representations. Using both synthetic data and human subject experiments in multiple classification tasks, we demonstrate that such representation is better aligned with human perception than representation solely optimized for classification. Human-compatible representations identify nearest neighbors that are perceived as more similar by humans and allow humans to make more accurate predictions, leading to substantial improvements in human decision accuracies (17.8% in butterfly vs. moth classification and 13.2% in pneumonia classification). 
    more » « less