Xiaodan Zhang

Chengxin Zheng, Junzhong Ji, Yanzhao Shi et al.

Brain CT report generation is significant to aid physicians in diagnosing cranial diseases. Recent studies concentrate on handling the consistency between visual and textual pathological features to improve the coherence of report. However, there exist some challenges: 1) Redundant visual representing: Massive irrelevant areas in 3D scans distract models from representing salient visual contexts. 2) Shifted semantic representing: Limited medical corpus causes difficulties for models to transfer the learned textual representations to generative layers. This study introduces a Pathological Clue-driven Representation Learning (PCRL) model to build cross-modal representations based on pathological clues and naturally adapt them for accurate report generation. Specifically, we construct pathological clues from perspectives of segmented regions, pathological entities, and report themes, to fully grasp visual pathological patterns and learn cross-modal feature representations. To adapt the representations for the text generation task, we bridge the gap between representation learning and report generation by using a unified large language model (LLM) with task-tailored instructions. These crafted instructions enable the LLM to be flexibly fine-tuned across tasks and smoothly transfer the semantic representation for report generation. Experiments demonstrate that our method outperforms previous methods and achieves SoTA performance. Our code is available at "https://github.com/Chauncey-Jheng/PCRL-MRG".

Yiming Huang, Aozhe Jia, Xiaodan Zhang et al.

Attention-based transformer models have achieved remarkable progress in multi-modal tasks, such as visual question answering. The explainability of attention-based methods has recently attracted wide interest as it can explain the inner changes of attention tokens by accumulating relevancy across attention layers. Current methods simply update relevancy by equally accumulating the token relevancy before and after the attention processes. However, the importance of token values is usually different during relevance accumulation. In this paper, we propose a weighted relevancy strategy, which takes the importance of token values into consideration, to reduce distortion when equally accumulating relevance. To evaluate our method, we propose a unified CLIP-based two-stage model, named CLIPmapper, to process Vision-and-Language tasks through CLIP encoder and a following mapper. CLIPmapper consists of self-attention, cross-attention, single-modality, and cross-modality attention, thus it is more suitable for evaluating our generic explainability method. Extensive perturbation tests on visual question answering and image captioning validate that our explainability method outperforms existing methods.

Pengxin Guo, Shuang Zeng, Wenhao Chen et al.

Federated Learning (FL) aims to protect data privacy by enabling clients to collectively train machine learning models without sharing their raw data. However, recent studies demonstrate that information exchanged during FL is subject to Gradient Inversion Attacks (GIA) and, consequently, a variety of privacy-preserving methods have been integrated into FL to thwart such attacks, such as Secure Multi-party Computing (SMC), Homomorphic Encryption (HE), and Differential Privacy (DP). Despite their ability to protect data privacy, these approaches inherently involve substantial privacy-utility trade-offs. By revisiting the key to privacy exposure in FL under GIA, which lies in the frequent sharing of model gradients that contain private data, we take a new perspective by designing a novel privacy preserve FL framework that effectively ``breaks the direct connection'' between the shared parameters and the local private data to defend against GIA. Specifically, we propose a Hypernetwork Federated Learning (HyperFL) framework that utilizes hypernetworks to generate the parameters of the local model and only the hypernetwork parameters are uploaded to the server for aggregation. Theoretical analyses demonstrate the convergence rate of the proposed HyperFL, while extensive experimental results show the privacy-preserving capability and comparable performance of HyperFL. Code is available at https://github.com/Pengxin-Guo/HyperFL.

Xiaodan Zhang, Sandeep Vemulapalli, Nabasmita Talukdar et al.

This study assesses the ability of state-of-the-art large language models (LLMs) including GPT-3.5, GPT-4, Falcon, and LLaMA 2 to identify patients with mild cognitive impairment (MCI) from discharge summaries and examines instances where the models' responses were misaligned with their reasoning. Utilizing the MIMIC-IV v2.2 database, we focused on a cohort aged 65 and older, verifying MCI diagnoses against ICD codes and expert evaluations. The data was partitioned into training, validation, and testing sets in a 7:2:1 ratio for model fine-tuning and evaluation, with an additional metastatic cancer dataset from MIMIC III used to further assess reasoning consistency. GPT-4 demonstrated superior interpretative capabilities, particularly in response to complex prompts, yet displayed notable response-reasoning inconsistencies. In contrast, open-source models like Falcon and LLaMA 2 achieved high accuracy but lacked explanatory reasoning, underscoring the necessity for further research to optimize both performance and interpretability. The study emphasizes the significance of prompt engineering and the need for further exploration into the unexpected reasoning-response misalignment observed in GPT-4. The results underscore the promise of incorporating LLMs into healthcare diagnostics, contingent upon methodological advancements to ensure accuracy and clinical coherence of AI-generated outputs, thereby improving the trustworthiness of LLMs for medical decision-making.

Bingkang Shi, Jen-tse Huang, Guoyi Li et al.

Leveraging their advanced capabilities, Large Language Models (LLMs) demonstrate vast application potential in video games--from dynamic scene generation and intelligent NPC interactions to adaptive opponents--replacing or enhancing traditional game mechanics. However, LLMs' trustworthiness in this application has not been sufficiently explored. In this paper, we reveal that the models' inherent social biases can directly damage game balance in real-world gaming environments. To this end, we present FairGamer, the first bias evaluation Benchmark for LLMs in video game scenarios, featuring six tasks and a novel metrics ${D_lstd}$. It covers three key scenarios in games where LLMs' social biases are particularly likely to manifest: Serving as Non-Player Characters, Interacting as Competitive Opponents, and Generating Game Scenes. FairGamer utilizes both reality-grounded and fully fictional game content, covering a variety of video game genres. Experiments reveal: (1) Decision biases directly cause game balance degradation, with Grok-3 (average ${D_lstd}$ score=0.431) exhibiting the most severe degradation; (2) LLMs demonstrate isomorphic social/cultural biases toward both real and virtual world content, suggesting their biases nature may stem from inherent model characteristics. These findings expose critical reliability gaps in LLMs' gaming applications. Our code and data are available at anonymous GitHub https://github.com/Anonymous999-xxx/FairGamer .

Bingkang Shi, Xiaodan Zhang, Dehan Kong et al.

The social biases and unwelcome stereotypes revealed by pretrained language models are becoming obstacles to their application. Compared to numerous debiasing methods targeting word level, there has been relatively less attention on biases present at phrase level, limiting the performance of debiasing in discipline domains. In this paper, we propose an automatic multi-token debiasing pipeline called \textbf{General Phrase Debiaser}, which is capable of mitigating phrase-level biases in masked language models. Specifically, our method consists of a \textit{phrase filter stage} that generates stereotypical phrases from Wikipedia pages as well as a \textit{model debias stage} that can debias models at the multi-token level to tackle bias challenges on phrases. The latter searches for prompts that trigger model's bias, and then uses them for debiasing. State-of-the-art results on standard datasets and metrics show that our approach can significantly reduce gender biases on both career and multiple disciplines, across models with varying parameter sizes.

Yanzhao Shi, Xiaodan Zhang, Junzhong Ji et al.

Automated 3D CT diagnosis empowers clinicians to make timely, evidence-based decisions by enhancing diagnostic accuracy and workflow efficiency. While multimodal large language models (MLLMs) exhibit promising performance in visual-language understanding, existing methods mainly focus on 2D medical images, which fundamentally limits their ability to capture complex 3D anatomical structures. This limitation often leads to misinterpretation of subtle pathologies and causes diagnostic hallucinations. In this paper, we present Hybrid Spatial Encoding Network (HSENet), a framework that exploits enriched 3D medical visual cues by effective visual perception and projection for accurate and robust vision-language understanding. Specifically, HSENet employs dual-3D vision encoders to perceive both global volumetric contexts and fine-grained anatomical details, which are pre-trained by dual-stage alignment with diagnostic reports. Furthermore, we propose Spatial Packer, an efficient multimodal projector that condenses high-resolution 3D spatial regions into a compact set of informative visual tokens via centroid-based compression. By assigning spatial packers with dual-3D vision encoders, HSENet can seamlessly perceive and transfer hybrid visual representations to LLM's semantic space, facilitating accurate diagnostic text generation. Experimental results demonstrate that our method achieves state-of-the-art performance in 3D language-visual retrieval (39.85% of R@100, +5.96% gain), 3D medical report generation (24.01% of BLEU-4, +8.01% gain), and 3D visual question answering (73.60% of Major Class Accuracy, +1.99% gain), confirming its effectiveness. Our code is available at https://github.com/YanzhaoShi/HSENet.

Wenchao Zhang, Jiahe Tian, Runze He et al.

Recent text-to-image (T2I) generation models have advanced significantly, enabling the creation of high-fidelity images from textual prompts. However, existing evaluation benchmarks primarily focus on the explicit alignment between generated images and prompts, neglecting the alignment with real-world knowledge beyond prompts. To address this gap, we introduce Align Beyond Prompts (ABP), a comprehensive benchmark designed to measure the alignment of generated images with real-world knowledge that extends beyond the explicit user prompts. ABP comprises over 2,000 meticulously crafted prompts, covering real-world knowledge across six distinct scenarios. We further introduce ABPScore, a metric that utilizes existing Multimodal Large Language Models (MLLMs) to assess the alignment between generated images and world knowledge beyond prompts, which demonstrates strong correlations with human judgments. Through a comprehensive evaluation of 8 popular T2I models using ABP, we find that even state-of-the-art models, such as GPT-4o, face limitations in integrating simple real-world knowledge into generated images. To mitigate this issue, we introduce a training-free strategy within ABP, named Inference-Time Knowledge Injection (ITKI). By applying this strategy to optimize 200 challenging samples, we achieved an improvement of approximately 43% in ABPScore. The dataset and code are available in https://github.com/smile365317/ABP.

Jiahua Xiao, Jiawei Zhang, Dongqing Zou et al.

Real-world image super-resolution (Real-ISR) has achieved a remarkable leap by leveraging large-scale text-to-image models, enabling realistic image restoration from given recognition textual prompts. However, these methods sometimes fail to recognize some salient objects, resulting in inaccurate semantic restoration in these regions. Additionally, the same region may have a strong response to more than one prompt and it will lead to semantic ambiguity for image super-resolution. To alleviate the above two issues, in this paper, we propose to consider semantic segmentation as an additional control condition into diffusion-based image super-resolution. Compared to textual prompt conditions, semantic segmentation enables a more comprehensive perception of salient objects within an image by assigning class labels to each pixel. It also mitigates the risks of semantic ambiguities by explicitly allocating objects to their respective spatial regions. In practice, inspired by the fact that image super-resolution and segmentation can benefit each other, we propose SegSR which introduces a dual-diffusion framework to facilitate interaction between the image super-resolution and segmentation diffusion models. Specifically, we develop a Dual-Modality Bridge module to enable updated information flow between these two diffusion models, achieving mutual benefit during the reverse diffusion process. Extensive experiments show that SegSR can generate realistic images while preserving semantic structures more effectively.

Xiaodan Zhang, Yanzhao Shi, Junzhong Ji et al.

The automatic generation of brain CT reports has gained widespread attention, given its potential to assist radiologists in diagnosing cranial diseases. However, brain CT scans involve extensive medical entities, such as diverse anatomy regions and lesions, exhibiting highly inconsistent spatial patterns in 3D volumetric space. This leads to biased learning of medical entities in existing methods, resulting in repetitiveness and inaccuracy in generated reports. To this end, we propose a Medical Entity-balanced Prompting Network (MEPNet), which harnesses the large language model (LLM) to fairly interpret various entities for accurate brain CT report generation. By introducing the visual embedding and the learning status of medical entities as enriched clues, our method prompts the LLM to balance the learning of diverse entities, thereby enhancing reports with comprehensive findings. First, to extract visual embedding of entities, we propose Knowledge-driven Joint Attention to explore and distill entity patterns using both explicit and implicit medical knowledge. Then, a Learning Status Scorer is designed to evaluate the learning of entity visual embeddings, resulting in unique learning status for individual entities. Finally, these entity visual embeddings and status are elaborately integrated into multi-modal prompts, to guide the text generation of LLM. This process allows LLM to self-adapt the learning process for biased-fitted entities, thereby covering detailed findings in generated reports. We conduct experiments on two brain CT report generation benchmarks, showing the effectiveness in clinical accuracy and text coherence.

Nabasmita Talukdar, Xiaodan Zhang, Shreya Paithankar et al.

Electronic Health Records (EHRs) have been increasingly used as real-world evidence (RWE) to support the discovery and validation of new drug indications. This paper surveys current approaches to EHR-based drug repurposing, covering data sources, processing methodologies, and representation techniques. It discusses study designs and statistical frameworks for evaluating drug efficacy. Key challenges in validation are discussed, with emphasis on the role of large language models (LLMs) and target trial emulation. By synthesizing recent developments and methodological advances, this work provides a foundational resource for researchers aiming to translate real-world data into actionable drug-repurposing evidence.

Shang Shang, Xinqiang Zhao, Zhongjiang Yao et al.

To demonstrate and address the underlying maliciousness, we propose a theoretical hypothesis and analytical approach, and introduce a new black-box jailbreak attack methodology named IntentObfuscator, exploiting this identified flaw by obfuscating the true intentions behind user prompts.This approach compels LLMs to inadvertently generate restricted content, bypassing their built-in content security measures. We detail two implementations under this framework: "Obscure Intention" and "Create Ambiguity", which manipulate query complexity and ambiguity to evade malicious intent detection effectively. We empirically validate the effectiveness of the IntentObfuscator method across several models, including ChatGPT-3.5, ChatGPT-4, Qwen and Baichuan, achieving an average jailbreak success rate of 69.21\%. Notably, our tests on ChatGPT-3.5, which claims 100 million weekly active users, achieved a remarkable success rate of 83.65\%. We also extend our validation to diverse types of sensitive content like graphic violence, racism, sexism, political sensitivity, cybersecurity threats, and criminal skills, further proving the substantial impact of our findings on enhancing 'Red Team' strategies against LLM content security frameworks.

Xiaowei Yuan, Jingyuan Hu, Xiaodan Zhang et al.

With the explosive growth of social media, opinionated postings with emojis have increased explosively. Many emojis are used to express emotions, attitudes, and opinions. Emoji representation learning can be helpful to improve the performance of emoji-related natural language processing tasks, especially in text sentiment analysis. However, most studies have only utilized the fixed descriptions provided by the Unicode Consortium without consideration of actual usage scenarios. As for the sentiment analysis task, many researchers ignore the emotional impact of the interaction between text and emojis. It results that the emotional semantics of emojis cannot be fully explored. In this work, we propose a method called EmoGraph2vec to learn emoji representations by constructing a co-occurrence graph network from social data and enriching the semantic information based on an external knowledge base EmojiNet to embed emoji nodes. Based on EmoGraph2vec model, we design a novel neural network to incorporate text and emoji information into sentiment analysis, which uses a hybrid-attention module combined with TextCNN-based classifier to improve performance. Experimental results show that the proposed model can outperform several baselines for sentiment analysis on benchmark datasets. Additionally, we conduct a series of ablation and comparison experiments to investigate the effectiveness and interpretability of our model.

Xiaowei Yuan, Jingyuan Hu, Xiaodan Zhang et al.

Emojis are widely used in online social networks to express emotions, attitudes, and opinions. As emotional-oriented characters, emojis can be modeled as important features of emotions towards the recipient or subject for sentiment analysis. However, existing methods mainly take emojis as heuristic information that fails to resolve the problem of ambiguity noise. Recent researches have utilized emojis as an independent input to classify text sentiment but they ignore the emotional impact of the interaction between text and emojis. It results that the emotional semantics of emojis cannot be fully explored. In this paper, we propose an emoji-based co-attention network that learns the mutual emotional semantics between text and emojis on microblogs. Our model adopts the co-attention mechanism based on bidirectional long short-term memory incorporating the text and emojis, and integrates a squeeze-and-excitation block in a convolutional neural network classifier to increase its sensitivity to emotional semantic features. Experimental results show that the proposed method can significantly outperform several baselines for sentiment analysis on short texts of social media.

Wei Zhou, Xin Cao, Xiaodan Zhang et al.

The existing 3D deep learning methods adopt either individual point-based features or local-neighboring voxel-based features, and demonstrate great potential for processing 3D data. However, the point based models are inefficient due to the unordered nature of point clouds and the voxel-based models suffer from large information loss. Motivated by the success of recent point-voxel representation, such as PVCNN, we propose a new convolutional neural network, called Multi Point-Voxel Convolution (MPVConv), for deep learning on point clouds. Integrating both the advantages of voxel and point-based methods, MPVConv can effectively increase the neighboring collection between point-based features and also promote independence among voxel-based features. Moreover, most of the existing approaches aim at solving one specific task, and only a few of them can handle a variety of tasks. Simply replacing the corresponding convolution module with MPVConv, we show that MPVConv can fit in different backbones to solve a wide range of 3D tasks. Extensive experiments on benchmark datasets such as ShapeNet Part, S3DIS and KITTI for various tasks show that MPVConv improves the accuracy of the backbone (PointNet) by up to \textbf{36\%}, and achieves higher accuracy than the voxel-based model with up to \textbf{34}$\times$ speedups. In addition, MPVConv outperforms the state-of-the-art point-based models with up to \textbf{8}$\times$ speedups. Notably, our MPVConv achieves better accuracy than the newest point-voxel-based model PVCNN (a model more efficient than PointNet) with lower latency.

Wei Zhou, Xin Cao, Xiaodan Zhang et al.

We present a new convolutional neural network, called Multi Voxel-Point Neurons Convolution (MVPConv), for fast and accurate 3D deep learning. The previous works adopt either individual point-based features or local-neighboring voxel-based features to process 3D model, which limits the performance of models due to the inefficient computation. Moreover, most of the existing 3D deep learning frameworks aim at solving one specific task, and only a few of them can handle a variety of tasks. Integrating both the advantages of the voxel and point-based methods, the proposed MVPConv can effectively increase the neighboring collection between point-based features and also promote the independence among voxel-based features. Simply replacing the corresponding convolution module with MVPConv, we show that MVPConv can fit in different backbones to solve a wide range of 3D tasks. Extensive experiments on benchmark datasets such as ShapeNet Part, S3DIS and KITTI for various tasks show that MVPConv improves the accuracy of the backbone (PointNet) by up to 36%, and achieves higher accuracy than the voxel-based model with up to 34 times speedup. In addition, MVPConv also outperforms the state-of-the-art point-based models with up to 8 times speedup. Notably, our MVPConv achieves better accuracy than the newest point-voxel-based model PVCNN (a model more efficient than PointNet) with lower latency.

Lingwei Wei, Dou Hu, Wei Zhou et al.

Document-level Sentiment Analysis (DSA) is more challenging due to vague semantic links and complicate sentiment information. Recent works have been devoted to leveraging text summarization and have achieved promising results. However, these summarization-based methods did not take full advantage of the summary including ignoring the inherent interactions between the summary and document. As a result, they limited the representation to express major points in the document, which is highly indicative of the key sentiment. In this paper, we study how to effectively generate a discriminative representation with explicit subject patterns and sentiment contexts for DSA. A Hierarchical Interaction Networks (HIN) is proposed to explore bidirectional interactions between the summary and document at multiple granularities and learn subject-oriented document representations for sentiment classification. Furthermore, we design a Sentiment-based Rethinking mechanism (SR) by refining the HIN with sentiment label information to learn a more sentiment-aware document representation. We extensively evaluate our proposed models on three public datasets. The experimental results consistently demonstrate the effectiveness of our proposed models and show that HIN-SR outperforms various state-of-the-art methods.

Chunyuan Yuan, Jiacheng Li, Wei Zhou et al.

Information diffusion prediction is a fundamental task for understanding the information propagation process. It has wide applications in such as misinformation spreading prediction and malicious account detection. Previous works either concentrate on utilizing the context of a single diffusion sequence or using the social network among users for information diffusion prediction. However, the diffusion paths of different messages naturally constitute a dynamic diffusion graph. For one thing, previous works cannot jointly utilize both the social network and diffusion graph for prediction, which is insufficient to model the complexity of the diffusion process and results in unsatisfactory prediction performance. For another, they cannot learn users' dynamic preferences. Intuitively, users' preferences are changing as time goes on and users' personal preference determines whether the user will repost the information. Thus, it is beneficial to consider users' dynamic preferences in information diffusion prediction. In this paper, we propose a novel dynamic heterogeneous graph convolutional network (DyHGCN) to jointly learn the structural characteristics of the social graph and dynamic diffusion graph. Then, we encode the temporal information into the heterogeneous graph to learn the users' dynamic preferences. Finally, we apply multi-head attention to capture the context-dependency of the current diffusion path to facilitate the information diffusion prediction task. Experimental results show that DyHGCN significantly outperforms the state-of-the-art models on three public datasets, which shows the effectiveness of the proposed model.

Xiaodan Zhang, Xinbo Gao, Wen Lu et al.

Learning fine-grained details is a key issue in image aesthetic assessment. Most of the previous methods extract the fine-grained details via random cropping strategy, which may undermine the integrity of semantic information. Extensive studies show that humans perceive fine-grained details with a mixture of foveal vision and peripheral vision. Fovea has the highest possible visual acuity and is responsible for seeing the details. The peripheral vision is used for perceiving the broad spatial scene and selecting the attended regions for the fovea. Inspired by these observations, we propose a Gated Peripheral-Foveal Convolutional Neural Network (GPF-CNN). It is a dedicated double-subnet neural network, i.e. a peripheral subnet and a foveal subnet. The former aims to mimic the functions of peripheral vision to encode the holistic information and provide the attended regions. The latter aims to extract fine-grained features on these key regions. Considering that the peripheral vision and foveal vision play different roles in processing different visual stimuli, we further employ a gated information fusion (GIF) network to weight their contributions. The weights are determined through the fully connected layers followed by a sigmoid function. We conduct comprehensive experiments on the standard AVA and Photo.net datasets for unified aesthetic prediction tasks: (i) aesthetic quality classification; (ii) aesthetic score regression; and (iii) aesthetic score distribution prediction. The experimental results demonstrate the effectiveness of the proposed method.