Artificial Intelligence and Robotics Commons^™

Aligning Large Language Models For Faithful Integrity Against Opposing Argument, Yong Zhao, Yang Deng, See-Kiong Ng, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Large Language Models (LLMs) have demonstrated impressive capabilities in complex reasoning tasks. However, they can be easily misled by unfaithful arguments during conversations, even when their original statements are correct. To this end, we investigate the problem of maintaining faithful integrity in LLMs. This involves ensuring that LLMs adhere to their faithful statements in the face of opposing arguments and are able to correct their incorrect statements when presented with faithful arguments. In this work, we propose a novel framework, named Alignment for Faithful Integrity with Confidence Estimation (AFICE), which aims to align the LLM responses with faithful integrity. Specifically, …

Dualopt: A Dual Divide-And-Optimize Algorithm For The Large-Scale Traveling Salesman Problem, Shipei Zhou, Yuandong Ding, Chi Zhang, Zhiguang Cao, Yan Jin

Research Collection School Of Computing and Information Systems

This paper proposes a dual divide-and-optimize algorithm (DualOpt) for solving the large-scale traveling salesman problem (TSP). DualOpt combines two complementary strategies to improve both solution quality and computational efficiency. The first strategy is a grid-based divide-and-conquer procedure that partitions the TSP into smaller subproblems, solving them in parallel and iteratively refining the solution by merging nodes and partial routes. The process continues until only one grid remains, yielding a high-quality initial solution. The second strategy involves a path-based divide-and-optimize procedure that further optimizes the solution by dividing it into sub-paths, optimizing each using a neural solver, and merging them back …

Leveraging Constraint Violation Signals For Action Constrained Reinforcement Learning, Janaka Chathuranga Brahmanage, Jiajing Ling, Akshat Kumar

Research Collection School Of Computing and Information Systems

In many RL applications, ensuring an agent’s actions adhere to constraints is crucial for safety. Most previous methods in Action-Constrained Reinforcement Learning (ACRL) employ a projection layer after the policy network to correct the action. However projection-based methods suffer from issues like the zero gradient problem and higher runtime due to the usage of optimization solvers. Recently methods were proposed to train generative models to learn a differentiable mapping between latent variables and feasible actions to address this issue. However, generative models require training using samples from the constrained action space, which itself is challenging. To address such limitations, first, …

Offline Safe Reinforcement Learning Using Trajectory Classification, Ze Gong, Akshat Kumar, Pradeep Varakantham

Research Collection School Of Computing and Information Systems

Offline safe reinforcement learning (RL) has emerged as a promising approach for learning safe behaviors without engaging in risky online interactions with the environment. Most existing methods in offline safe RL rely on cost constraints at each time step (derived from global cost constraints) and this can result in either overly conservative policies or violation of safety constraints. In this paper, we propose to learn a policy that generates desirable trajectories and avoids undesirable trajectories. To be specific, we first partition the pre-collected dataset of state-action trajectories into desirable and undesirable subsets. Intuitively, the desirable set contains high reward and …

Occlusion-Insensitive Talking Head Video Generation Via Facelet Compensation, Yuhui Deng, Yuqin Lu, Yangyang Xu, Yongwei Nie, Shengfeng He

Research Collection School Of Computing and Information Systems

Talking head video generation involves animating a still face image using facial motion cues derived from a driving video to replicate target poses and expressions. Traditional methods often rely on the assumption that the relative positions of facial keypoints remain unchanged. However, this assumption fails when keypoints are occluded or when the head is in a profile pose, leading to inconsistencies in identity and blurring in certain facial regions. In this paper, we introduce Occlusion-Insensitive Talking Head Video Generation, a novel approach that eliminates the reliance on spatial correlation of keypoints and instead leverages semantic correlation. Our method transforms facial …

Adversarial Attacks On Event-Based Pedestrian Detectors: A Physical Approach, Guixu Lin, Muyao Niu, Qingtian Zhu, Zhengwei Yin, Zhuoxiao Li, Shengfeng He, Yinqiang Zheng

Research Collection School Of Computing and Information Systems

Event cameras, known for their low latency and high dynamic range, show great potential in pedestrian detection applications. However, while recent research has primarily focused on improving detection accuracy, the robustness of event-based visual models against physical adversarial attacks has received limited attention. For example, adversarial physical objects, such as specific clothing patterns or accessories, can exploit inherent vulnerabilities in these systems, leading to misdetections or misclassifications. This study is the first to explore physical adversarial attacks on event-driven pedestrian detectors, specifically investigating whether certain clothing patterns worn by pedestrians can cause these detectors to fail, effectively rendering them unable …

Personamagic: Stage-Regulated High-Fidelity Face Customization With Tandem Equilibrium, Xinzhe Li, Jiahui Zhan, Shengfeng He, Yangyang Xu, Junyu Dong, Huaidong Zhang, Yong Du

Research Collection School Of Computing and Information Systems

Personalized image generation has made significant strides in adapting content to novel concepts. However, a persistent challenge remains: balancing the accurate reconstruction of unseen concepts with the need for editability according to the prompt, especially when dealing with the complex nuances of facial features. In this study, we delve into the temporal dynamics of the text-to-image conditioning process, emphasizing the crucial role of stage partitioning in introducing new concepts. We present PersonaMagic, a stage-regulated generative technique designed for high-fidelity face customization. Using a simple MLP network, our method learns a series of embeddings within a specific timestep interval to capture …

An Aspect Performance-Aware Hypergraph Neural Network For Review-Based Recommendation, Junrui Liu, Tong Li, Di Wu, Zifang Tang, Yuan Fang, Zhen Yang

Research Collection School Of Computing and Information Systems

Online reviews allow consumers to provide detailed feedback on various aspects of items. Existing methods utilize these aspects to model users' fine-grained preferences for specific item features through graph neural networks. We argue that the performance of items on different aspects is important for making precise recommendations, which has not been taken into account by existing approaches, due to lack of data. In this paper, we propose an aspect performance-aware hypergraph neural network (APH) for the review-based recommendation, which learns the performance of items from the conflicting sentiment polarity of user reviews. Specifically, APH comprehensively models the relationships among users, …

Marginal Benefit Driven Rl Teacher For Unsupervised Environment Design, Dexun Li, Wenjun Li, Pradeep Varakantham

Research Collection School Of Computing and Information Systems

Training generally capable agents in complex environments is a challenging task that involves identifying the “right” environments at the training stage. Recent research has highlighted the potential of the Unsupervised Environment Design framework, which generates environment instances/levels adaptively at the frontier of the agent’s capabilities using regret measures. While regret approaches have shown promise in generating feasible environments, they can produce difficult environments that are challenging for an RL agent to learn from. This is because regret represents the best-case (upper bound) learning potential and not the actual learning potential of an environment. To address this, we propose an alternative …

Adapting Knowledge Prompt Tuning For Enhanced Automated Program Repair, Xuemeng Cai, Lingxiao Jiang

Research Collection School Of Computing and Information Systems

Automated Program Repair (APR) aims to enhance software reliability by automatically generating bug-fixing patches. Recent work has improved the state-of-the-art of APR by fine-tuning pre-trained large language models (LLMs), such as CodeT5, for APR. However, the effectiveness of fine-tuning be-comes weakened in data scarcity scenarios, and data scarcity can be a common issue in practice, limiting fine-tuning performance. To alleviate this limitation, this paper adapts prompt tuning for enhanced APR and conducts a comprehensive study to evaluate its effectiveness in data scarcity scenarios, using three LLMs of different sizes and six diverse datasets across four programming languages. Prompt tuning rewrites …

Evaluating Software Development Agents: Patch Patterns, Code Quality, And Issue Complexity In Real-World Github Scenarios, Zhi Chen, Lingxiao Jiang

Research Collection School Of Computing and Information Systems

In recent years, AI-based software engineering has progressed from pre-trained models to advanced agentic workflows, with Software Development Agents representing the next major leap. These agents, capable of reasoning, planning, and interacting with external environments, offer promising solutions to complex software engineering tasks. However, while much research has evaluated code generated by large language models (LLMs), comprehensive studies on agent-generated patches, particularly in real-world settings, are lacking. This study addresses that gap by evaluating 4,892 patches from 10 top-ranked agents on 500 real-world GitHub issues from SWE-Bench Verified, focusing on their impact on code quality. Our analysis shows no single …

Drone Delivery Network Design With Uncertainties, Wenjia Zeng, Jiang Ruiwei, Hai Yang, Hai Wang

Research Collection School Of Computing and Information Systems

Unmanned aerial vehicles (UAVs), also called drones, are gaining popularity as an alternative delivery mode due to their faster delivery speed and reduced labor costs. Several companies, especially e-commerce giants, are conducting pilot projects that use drones to deliver fast food and groceries. In 2021, for example, Walmart partnered with Zipline in the United States to provide delivery services for areas near Walmart stores in Arkansas. In China, Meituan drone delivery services have been launched in Shenzhen and have conducted trial food delivery that cover more than 8,000 households.

Unlocking The Potential Of Black-Box Pre-Trained Gnns For Graph Few-Shot Learning, Qiannan Zhang, Shichao Pei, Yuan Fang, Xiangliang Zhang

Research Collection School Of Computing and Information Systems

Few-shot learning has emerged as an important problem on graphs to combat label scarcity, which can be approached by current trends in pre-trained graph neural networks (GNNs) and meta-learning. Recent efforts integrate both paradigms in a white-box setting, leaving the more realistic black-box setting largely underexplored, where the parameters and gradients in the pre-trained GNNs are inaccessible. In this paper, we study the critical problem: Leveraging black-box pre-trained GNNs for graph few-shot learning. Despite its appeal, two key issues hinder the unlocking of its potential: the inherent task gap between pre-training and downstream stages, which can introduce irrelevant knowledge and …

Learning To Identify Seen, Unseen And Unknown In The Open World: A Practical Setting For Zero-Shot Learning, Sethupathy Parameswaran, Yuan Fang, Chandan Gautam, Savitha Ramasamy, Xiaoli Li

Research Collection School Of Computing and Information Systems

As vision-language models advance, addressing the Zero-Shot Learning (ZSL) problem in the open world becomes increasingly crucial. Specifically, a robust model must handle three types of samples during inference: seen classes with visual and semantic information provided in training, unseen classes with only the semantic information in training, and unknown samples with no prior information from training. Existing methods either handle seen and unseen classes together (ZSL) or seen and unknown classes (known as Open-Set Recognition, OSR). However, none addresses the simultaneous handling of all three, which we term Open-Set Zero-Shot Learning (OZSL). To address this problem, we propose a …

Enhancing Item‑Level Bundle Representation For Bundle Recommendation, Xiaoyu Du, Kun Qian, Yunshan Ma, Xinguang Xiang

Research Collection School Of Computing and Information Systems

Bundle recommendation approaches offer users a set of related items on a particular topic. The current state-of-the-art (SOTA) method utilizes contrastive learning to learn representations at both the bundle and item levels. However, due to the inherent difference between the bundle-level and item-level preferences, the item-level representations may not receive sufficient information from the bundle affiliations to make accurate predictions. In this article, we propose a novel approach, Enhanced Bundle Recommendation (EBRec), which incorporates two enhanced modules to explore inherent item-level bundle representations. First, we propose to incorporate the bundle-user-item (B-U-I) high-order correlations to explore more collaborative information, thus to …

Attackg+: Boosting Attack Graph Construction With Large Language Models, Yongheng Zhang, Tingwen Du, Yunshan Ma, Xiang Wang, Yi Xie, Guozheng Yang, Yuliang Lu, Ee‑Chien Chang

Research Collection School Of Computing and Information Systems

Attack graph construction seeks to convert textual cyber threat intelligence (CTI) reports into structuredrepresentations, portraying the evolutionary traces of cyber attacks. Even though previous research hasproposed various methods to construct attack graphs, they generally suffer from limited generalizationcapability to diverse knowledge types as well as requirement of expertise in model design and tuning.Addressing these limitations, we seek to utilize Large Language Models (LLMs), which have achieved enormoussuccess in a broad range of tasks given exceptional capabilities in both language understanding and zeroshot task fulfillment. Thus, we propose a fully automatic LLM-based framework to construct attack graphsnamed: AttacKG+. Our framework consists …

Enhanced Sample Selection With Confidence Tracking: Identifying Correctly Labeled Yet Hard-To-Learn Samples In Noisy Data, Weiran Pan, Wei Wei, Feida Zhu, Yong Deng

Research Collection School Of Computing and Information Systems

We propose a novel sample selection method for image classification in the presence of noisy labels. Existing methods typically consider small-loss samples as correctly labeled. However, some correctly labeled samples are inherently difficult for the model to learn and can exhibit high loss similar to mislabeled samples in the early stages of training. Consequently, setting a threshold on per-sample loss to select correct labels results in a trade-off between precision and recall in sample selection: a lower threshold may miss many correctly labeled hard-to-learn samples (low recall), while a higher threshold may include many mislabeled samples (low precision). To address …

Explainable Neural Networks With Guarantee: A Sparse Estimation Approach, Antoine Ledent, Peng Liu

Research Collection School Of Computing and Information Systems

Balancing predictive power and interpretability has long been a challenging research area, particularly in powerful yet complex models like neural networks, where nonlinearity obstructs direct interpretation. This paper introduces a novel approach to constructing an explainable neural network that harmonizes predictiveness and explainability. Our model is designed as a linear combination of a sparse set of jointly learned features, each derived from a different trainable function applied to a single 1-dimensional input feature. Leveraging the ability to learn arbitrarily complex relationships, our neural network architecture enables automatic selection of a sparse set of important features, with the final prediction being …

Enhancing Virtual Reality Usability With A Ml Based Dynamic Adaptive System, Ananth Ramaseri-Chandra, Hassan Reza

Computer Science Posters and Presentations

Virtual reality (VR) holds tremendous potential, but cybersickness degrades the user experience. Since individuals vary, a one-size-fits-all design is insufficient. Our work introduces a dynamic adaptive system that personalizes VR experiences by learning individual cybersickness profiles from head-tracking data and sickness questionnaires while adjusting settings such as field of view and foveated rendering strength. Early results show that our system reduces post-exposure sickness scores, enhancing the user experience and highlighting the importance of personalizing VR.

What's The Art In Artificial Intelligence?, Emily Verla Bovino

Open Educational Resources

This workbook learns from Black and Indigenous artists working with Artificial Intelligence to confront issues of ethics and aesthetics in its technologies. It features guided learning activities with links to publicly available video lectures and online articles, and includes options for experiential learning through both a tutorial in Midjourney and a visit to the public art collection at York College in Jamaica, Queens. Featured artists include: American Artist, Rizvana Bradley, Beth Coleman, Denise Ferreira da Silva, Suzanne Kite, Sondra Perry, Mimi Onuoha and Alisha B. Wormsley. Works by Martin Puryear and Maren Hassinger are explored in the Midjourney exercise.

About …