papers

Publications (89)

eess.AS2026

Dimensionality-Aware Anomaly Detection in Learned Representations of Self-Supervised Speech Models

Sandra Arcos-Holzinger, Sarah M. Erfani, James Bailey +1

Self-supervised speech models (S3Ms) achieve strong downstream performance, yet their learned representations remain poorly understood under natural and adversarial perturbations.…

cs.LG2025

Coarse-to-Fine Open-Set Graph Node Classification with Large Language Models

Xueqi Ma, Xingjun Ma, Sarah Monazam Erfani +2

Developing open-set classification methods capable of classifying in-distribution (ID) data while detecting out-of-distribution (OOD) samples is essential for deploying graph neura…

cs.CV2020

Clean-Label Backdoor Attacks on Video Recognition Models

Shihao Zhao, Xingjun Ma, Xiang Zheng +3

Deep neural networks (DNNs) are vulnerable to backdoor attacks which can hide backdoor triggers in DNNs by poisoning training data. A backdoored model behaves normally on clean tes…

cs.LG2017

Adversarial Generation of Real-time Feedback with Neural Networks for Simulation-based Training

Xingjun Ma, Sudanthi Wijewickrema, Shuo Zhou +4

Simulation-based training (SBT) is gaining popularity as a low-cost and convenient training technique in a vast range of applications. However, for a SBT platform to be fully utili…

cs.AI2026

Attention in Space: Functional Roles of VLM Heads for Spatial Reasoning

Xueqi Ma, Shuo Yang, Yanbei Jiang +6

Despite remarkable advances in large Vision-Language Models (VLMs), spatial reasoning remains a persistent challenge. In this work, we investigate how attention heads within VLMs c…

cs.LG2022

On the Convergence and Robustness of Adversarial Training

Yisen Wang, Xingjun Ma, James Bailey +3

Improving the robustness of deep neural networks (DNNs) to adversarial examples is an important yet challenging problem for secure deep learning. Across existing defense techniques…

cs.LG2020

Normalized Loss Functions for Deep Learning with Noisy Labels

Xingjun Ma, Hanxun Huang, Yisen Wang +3

Robust loss functions are essential for training accurate deep neural networks (DNNs) in the presence of noisy (incorrect) labels. It has been shown that the commonly used Cross En…

cs.LG2026

E-BATS: Efficient Backpropagation-Free Test-Time Adaptation for Speech Foundation Models

Jiaheng Dong, Hong Jia, Soumyajit Chatterjee +3

Speech Foundation Models encounter significant performance degradation when deployed in real-world scenarios involving acoustic domain shifts, such as background noise and speaker…

cs.CV2022

Alpha-IoU: A Family of Power Intersection over Union Losses for Bounding Box Regression

Jiabo He, Sarah Erfani, Xingjun Ma +3

Bounding box (bbox) regression is a fundamental task in computer vision. So far, the most commonly used loss functions for bbox regression are the Intersection over Union (IoU) los…

stat.ML2016

Ground Truth Bias in External Cluster Validity Indices

Yang Lei, James C. Bezdek, Simone Romano +3

It has been noticed that some external CVIs exhibit a preferential bias towards a larger or smaller number of clusters which is monotonic (directly or inversely) in the number of c…

cs.LG2021

Dual Head Adversarial Training

Yujing Jiang, Xingjun Ma, Sarah Monazam Erfani +1

Deep neural networks (DNNs) are known to be vulnerable to adversarial examples/attacks, raising concerns about their reliability in safety-critical applications. A number of defens…

stat.ML2016

Invariant backpropagation: how to train a transformation-invariant neural network

Sergey Demyanov, James Bailey, Ramamohanarao Kotagiri +1

In many classification problems a classifier should be robust to small variations in the input vector. This is a desired property not only for particular transformations, such as t…

cs.LG2019

FCC-GAN: A Fully Connected and Convolutional Net Architecture for GANs

Sukarna Barua, Sarah Monazam Erfani, James Bailey

Generative Adversarial Networks (GANs) are a powerful class of generative models. Despite their successes, the most appropriate choice of a GAN network architecture is still not we…

cs.LG2024

Time Series Representation Learning with Supervised Contrastive Temporal Transformer

Yuansan Liu, Sudanthi Wijewickrema, Christofer Bester +2

Finding effective representations for time series data is a useful but challenging task. Several works utilize self-supervised or unsupervised learning methods to address this. How…

cs.CL2023

Semantic-Preserving Adversarial Text Attacks

Xinghao Yang, Weifeng Liu, James Bailey +2

Deep neural networks (DNNs) are known to be vulnerable to adversarial images, while their robustness in text classification is rarely studied. Several lines of text attack methods…

cond-mat.mtrl-sci2021

On local intrinsic dimensionality of deformation in complex materials

Shuo Zhou, Antoinette Tordesillas, Mehdi Pouragha +2

We propose a new metric called s-LID based on the concept of Local Intrinsic Dimensionality to identify and quantify hierarchies of kinematic patterns in heterogeneous media. s-LID…

cs.LG2026

Humanity's Last Exam

Long Phan, Alice Gatti, Ziwen Han +1144

Benchmarks are important tools for tracking the rapid advancements in large language model (LLM) capabilities. However, benchmarks are not keeping pace in difficulty: LLMs now achi…

cs.CV2025

X-Transfer Attacks: Towards Super Transferable Adversarial Attacks on CLIP

Hanxun Huang, Sarah Erfani, Yige Li +2

As Contrastive Language-Image Pre-training (CLIP) models are increasingly adopted for diverse downstream tasks and integrated into large vision-language models (VLMs), their suscep…

cs.LG2022

Exploring Architectural Ingredients of Adversarially Robust Deep Neural Networks

Hanxun Huang, Yisen Wang, Sarah Monazam Erfani +3

Deep neural networks (DNNs) are known to be vulnerable to adversarial attacks. A range of defense methods have been proposed to train adversarially robust DNNs, among which adversa…

cs.LG2021

Neural Architecture Search via Combinatorial Multi-Armed Bandit

Hanxun Huang, Xingjun Ma, Sarah M. Erfani +1

Neural Architecture Search (NAS) has gained significant popularity as an effective tool for designing high performance deep neural networks (DNNs). NAS can be performed via policy…

cs.CV2021

What Do Deep Nets Learn? Class-wise Patterns Revealed in the Input Space

Shihao Zhao, Xingjun Ma, Yisen Wang +3

Deep neural networks (DNNs) are increasingly deployed in different applications to achieve state-of-the-art performance. However, they are often applied as a black box with limited…

cs.LG2024

LDReg: Local Dimensionality Regularized Self-Supervised Learning

Hanxun Huang, Ricardo J. G. B. Campello, Sarah Monazam Erfani +3

Representations learned via self-supervised learning (SSL) can be susceptible to dimensional collapse, where the learned representation subspace is of extremely low dimensionality…

cs.LG2024

Dimensionality-Aware Outlier Detection: Theoretical and Experimental Analysis

Alastair Anderberg, James Bailey, Ricardo J. G. B. Campello +4

We present a nonparametric method for outlier detection that takes full account of local variations in intrinsic dimensionality within the dataset. Using the theory of Local Intrin…

cs.CV2020

Understanding Adversarial Attacks on Deep Learning Based Medical Image Analysis Systems

Xingjun Ma, Yuhao Niu, Lin Gu +4

Deep neural networks (DNNs) have become popular for medical image analysis tasks like cancer diagnosis and lesion detection. However, a recent study demonstrates that medical deep…

cs.LG2025

Stochastic Diffusion: A Diffusion Probabilistic Model for Stochastic Time Series Forecasting

Yuansan Liu, Sudanthi Wijewickrema, Dongting Hu +3

Recent innovations in diffusion probabilistic models have paved the way for significant progress in image, text and audio generation, leading to their applications in generative ti…

cs.LG2017

Efficient Orthogonal Parametrisation of Recurrent Neural Networks Using Householder Reflections

Zakaria Mhammedi, Andrew Hellicar, Ashfaqur Rahman +1

The problem of learning long-term dependencies in sequences using Recurrent Neural Networks (RNNs) is still a major challenge. Recent methods have been suggested to solve this prob…

cs.LG2024

End-to-End Anti-Backdoor Learning on Images and Time Series

Yujing Jiang, Xingjun Ma, Sarah Monazam Erfani +2

Backdoor attacks present a substantial security concern for deep learning models, especially those utilized in applications critical to safety and security. These attacks manipulat…

cs.LG2023

Backdoor Attacks on Time Series: A Generative Approach

Yujing Jiang, Xingjun Ma, Sarah Monazam Erfani +1

Backdoor attacks have emerged as one of the major security threats to deep learning models as they can easily control the model's test-time predictions by pre-injecting a backdoor…

cs.CV2024

Part-aware Unified Representation of Language and Skeleton for Zero-shot Action Recognition

Anqi Zhu, Qiuhong Ke, Mingming Gong +1

While remarkable progress has been made on supervised skeleton-based action recognition, the challenge of zero-shot recognition remains relatively unexplored. In this paper, we arg…

physics.ao-ph2025

Sea ice floe segmentation in close-range optical imagery using active contour and foundation models

Giulio Passerotti, Alberto Alberello, Marcello Vichi +3

The size of sea ice floes in the marginal ice zone (MIZ) is a key factor influencing ice coverage, albedo, wave propagation, and ocean--atmosphere energy exchanges. Floe size can b…

cs.CV2023

Imbalanced Gradients: A Subtle Cause of Overestimated Adversarial Robustness

Xingjun Ma, Linxi Jiang, Hanxun Huang +3

Evaluating the robustness of a defense model is a challenging task in adversarial robustness research. Obfuscated gradients have previously been found to exist in many defense meth…

cs.LG2026

On the Use of Bagging for Local Intrinsic Dimensionality Estimation

Kristóf Péter, Ricardo J. G. B. Campello, James Bailey +1

The theory of Local Intrinsic Dimensionality (LID) has become a valuable tool for characterizing local complexity within and across data manifolds, supporting a range of data minin…

cs.LG2017

TopicResponse: A Marriage of Topic Modelling and Rasch Modelling for Automatic Measurement in MOOCs

Jiazhen He, Benjamin I. P. Rubinstein, James Bailey +2

This paper explores the suitability of using automatically discovered topics from MOOC discussion forums for modelling students' academic abilities. The Rasch model from psychometr…

cs.HC2021

GAVIN: Gaze-Assisted Voice-Based Implicit Note-taking

Anam Ahmad Khan, Joshua Newn, Ryan Kelly +3

Annotation is an effective reading strategy people often undertake while interacting with digital text. It involves highlighting pieces of text and making notes about them. Annotat…

cs.LG2023

Distilling Cognitive Backdoor Patterns within an Image

Hanxun Huang, Xingjun Ma, Sarah Erfani +1

This paper proposes a simple method to distill and detect backdoor patterns within an image: \emph{Cognitive Distillation} (CD). The idea is to extract the "minimal essence" from a…

cs.CV2025

Exploring Weak-to-Strong Generalization for CLIP-based Classification

Jinhao Li, Sarah M. Erfani, Lei Feng +2

Aligning large-scale commercial models with user intent is crucial to preventing harmful outputs. Current methods rely on human supervision but become impractical as model complexi…

cs.CV2022

Adaptive Local-Component-aware Graph Convolutional Network for One-shot Skeleton-based Action Recognition

Anqi Zhu, Qiuhong Ke, Mingming Gong +1

Skeleton-based action recognition receives increasing attention because the skeleton representations reduce the amount of training data by eliminating visual information irrelevant…

cs.LG2018

Learning Deep Hidden Nonlinear Dynamics from Aggregate Data

Yisen Wang, Bo Dai, Lingkai Kong +3

Learning nonlinear dynamics from diffusion data is a challenging problem since the individuals observed may be different at different time points, generally following an aggregate…

cs.LG2019

Black-box Adversarial Attacks on Video Recognition Models

Linxi Jiang, Xingjun Ma, Shaoxiang Chen +2

Deep neural networks (DNNs) are known for their vulnerability to adversarial examples. These are examples that have undergone small, carefully crafted perturbations, and which can…

cs.DB2024

Learned Indexes with Distribution Smoothing via Virtual Points

Kasun Amarasinghe, Farhana Choudhury, Jianzhong Qi +1

Recent research on learned indexes has created a new perspective for indexes as models that map keys to their respective storage locations. These learned indexes are created to app…

cs.CL2024

Whose Side Are You On? Investigating the Political Stance of Large Language Models

Pagnarasmey Pit, Xingjun Ma, Mike Conway +6

Large Language Models (LLMs) have gained significant popularity for their application in various everyday tasks such as text generation, summarization, and information retrieval. A…

cs.AI2025

Investigating The Functional Roles of Attention Heads in Vision Language Models: Evidence for Reasoning Modules

Yanbei Jiang, Xueqi Ma, Shu Liu +5

Despite excelling on multimodal benchmarks, vision-language models (VLMs) largely remain a black box. In this paper, we propose a novel interpretability framework to systematically…

cs.SD2026

Emotion-Aware Quantization for Discrete Speech Representations: An Analysis of Emotion Preservation

Haoguang Zhou, Siyi Wang, Jingyao Wu +2

Modern speech systems increasingly use discretized self-supervised speech representations for compression and integration with token-based models, yet their impact on emotional inf…

cs.AI2021

Adversarial Interaction Attack: Fooling AI to Misinterpret Human Intentions

Nodens Koren, Qiuhong Ke, Yisen Wang +2

Understanding the actions of both humans and artificial intelligence (AI) agents is important before modern AI systems can be fully integrated into our daily life. In this paper, w…

cs.CR2026

Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses

Xiao Li, Xiang Zheng, Yifeng Gao +35

Embodied Artificial Intelligence (Embodied AI) integrates perception, cognition, planning, and interaction into agents that operate in open-world, safety-critical environments. As…

cs.CV2022

IGFormer: Interaction Graph Transformer for Skeleton-based Human Interaction Recognition

Yunsheng Pang, Qiuhong Ke, Hossein Rahmani +2

Human interaction recognition is very important in many applications. One crucial cue in recognizing an interaction is the interactive body parts. In this work, we propose a novel…

cs.CV2024

DDA: Dimensionality Driven Augmentation Search for Contrastive Learning in Laparoscopic Surgery

Yuning Zhou, Henry Badgery, Matthew Read +2

Self-supervised learning (SSL) has potential for effective representation learning in medical imaging, but the choice of data augmentation is critical and domain-specific. It remai…

cs.SI2012

A Time Decoupling Approach for Studying Forum Dynamics

Andrey Kan, Jeffrey Chan, Conor Hayes +3

Online forums are rich sources of information about user communication activity over time. Finding temporal patterns in online forum communication threads can advance our understan…

cs.SI2022

Finding Influentials in Twitter: A Temporal Influence Ranking Model

Xingjun Ma, Chunping Li, James Bailey +1

With the growing popularity of online social media, identifying influential users in these social networks has become very popular. Existing works have studied user attributes, net…

cs.LG2015

MOOCs Meet Measurement Theory: A Topic-Modelling Approach

Jiazhen He, Benjamin I. P. Rubinstein, James Bailey +3

This paper adapts topic models to the psychometric testing of MOOC students based on their online forum postings. Measurement theory from education and psychology provides statisti…

cs.AI2017

Providing Effective Real-time Feedback in Simulation-based Surgical Training

Xingjun Ma, Sudanthi Wijewickrema, Yun Zhou +3

Virtual reality simulation is becoming popular as a training platform in surgical education. However, one important aspect of simulation-based surgical training that has not receiv…

cs.CV2018

Dimensionality-Driven Learning with Noisy Labels

Xingjun Ma, Yisen Wang, Michael E. Houle +5

Datasets with significant proportions of noisy (incorrect) class labels present challenges for training accurate Deep Neural Networks (DNNs). We propose a new perspective for under…

cs.LG2021

Unlearnable Examples: Making Personal Data Unexploitable

Hanxun Huang, Xingjun Ma, Sarah Monazam Erfani +2

The volume of "free" data on the internet has been key to the current success of deep learning. However, it also raises privacy concerns about the unauthorized exploitation of pers…

cs.LG2020

Skip Connections Matter: On the Transferability of Adversarial Examples Generated with ResNets

Dongxian Wu, Yisen Wang, Shu-Tao Xia +2

Skip connections are an essential component of current state-of-the-art deep neural networks (DNNs) such as ResNet, WideResNet, DenseNet, and ResNeXt. Despite their huge success in…

cs.SD2026

CoCoEmo: Composable and Controllable Human-Like Emotional TTS via Activation Steering

Siyi Wang, Shihong Tan, Siyi Liu +4

Emotional expression in human speech is nuanced and compositional, often involving multiple, sometimes conflicting, affective cues that may diverge from linguistic content. In cont…

cs.LG2026

Local Intrinsic Dimensionality of Ground Motion Data for Early Detection of Catastrophic Slope Failure

Yuansan Liu, James Bailey, Antoinette Tordesillas

Local Intrinsic Dimensionality (LID) has shown strong potential for anomaly detection in high-dimensional data, including landslide failure detection in granular media, where early…

cs.LG2026

Time-Transformer: Integrating Local and Global Features for Better Time Series Generation (Extended Version)

Yuansan Liu, Sudanthi Wijewickrema, Ang Li +3

Generating time series data is a promising approach to address data deficiency problems. However, it is also challenging due to the complex temporal properties of time series data,…

stat.ML2015

Adjusting for Chance Clustering Comparison Measures

Simone Romano, Nguyen Xuan Vinh, James Bailey +1

Adjusted for chance measures are widely used to compare partitions/clusterings of the same data set. In particular, the Adjusted Rand Index (ARI) based on pair-counting, and the Ad…

cs.CR2026

Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety

Xingjun Ma, Yifeng Gao, Yixu Wang +45

The rapid advancement of large models, driven by their exceptional abilities in learning and generalization through large-scale pre-training, has reshaped the landscape of Artifici…

cs.HC2020

Using voice note-taking to promote learners' conceptual understanding

Anam Ahmad Khan, Sadia Nawaz, Joshua Newn +3

Though recent technological advances have enabled note-taking through different modalities (e.g., keyboard, digital ink, voice), there is still a lack of understanding of the effec…

cs.LG2019

Symmetric Cross Entropy for Robust Learning with Noisy Labels

Yisen Wang, Xingjun Ma, Zaiyi Chen +3

Training accurate deep neural networks (DNNs) in the presence of noisy labels is an important and challenging task. Though a number of approaches have been proposed for learning wi…

q-bio.NC2025

Cognitive Mirrors: Exploring the Diverse Functional Roles of Attention Heads in LLM Reasoning

Xueqi Ma, Jun Wang, Yanbei Jiang +3

Large language models (LLMs) have achieved state-of-the-art performance in a variety of tasks, but remain largely opaque in terms of their internal mechanisms. Understanding these…

cs.CV2024

Mitigating Challenges of the Space Environment for Onboard Artificial Intelligence: Design Overview of the Imaging Payload on SpIRIT

Miguel Ortiz del Castillo, Jonathan Morgan, Jack McRobbie +12

Artificial intelligence (AI) and autonomous edge computing in space are emerging areas of interest to augment capabilities of nanosatellites, where modern sensors generate orders o…

physics.optics2024

Bidirectional Adversarial Autoencoders for the design of Plasmonic Metasurfaces

Yuansan Liu, Jeygopi Panisilvam, Peter Dower +2

Deep Learning has been a critical part of designing inverse design methods that are computationally efficient and accurate. An example of this is the design of photonic metasurface…

cs.HC2017

Feedback Techniques in Computer-Based Simulation Training: A Survey

Sudanthi Wijewickrema, Xingjun Ma, James Bailey +2

Computer-based simulation training (CBST) is gaining popularity in a vast range of applications such as surgery, rehabilitation therapy, military applications, and driver/pilot tra…

cs.CV2020

Adversarial Camouflage: Hiding Physical-World Attacks with Natural Styles

Ranjie Duan, Xingjun Ma, Yisen Wang +3

Deep neural networks (DNNs) are known to be vulnerable to adversarial examples. Existing works have mostly focused on either digital adversarial examples created via small and impe…

cs.CV2022

A Survey of Automated Data Augmentation Algorithms for Deep Learning-based Image Classification Tasks

Zihan Yang, Richard O. Sinnott, James Bailey +1

In recent years, one of the most popular techniques in the computer vision community has been the deep learning technique. As a data-driven technique, deep model requires enormous…

cs.SD2026

RAIL: Rethinking Auditory Intelligence in Large Audio-Language Models with a CHC-Grounded Benchmark

Hongyu Jin, Siyi Wang, Yang Xiao +10

Humans process rich auditory environments through tightly integrated cognitive capabilities such as audio perception, audio reasoning, and memory. Despite recent progress in large…

cs.LG2023

PELP: Pioneer Event Log Prediction Using Sequence-to-Sequence Neural Networks

Wenjun Zhou, Artem Polyvyanyy, James Bailey

Process mining, a data-driven approach for analyzing, visualizing, and improving business processes using event logs, has emerged as a powerful technique in the field of business p…

cs.CV2026

DynaPURLS: Dynamic Refinement of Part-Aware Representations for Skeleton-Based Zero-Shot Action Recognition

Jingmin Zhu, Anqi Zhu, James Bailey +5

Zero-shot skeleton-based action recognition (ZS-SAR) is fundamentally constrained by prevailing approaches that rely on aligning skeleton features with static, class-level semantic…

cs.CV2018

Iterative Learning with Open-set Noisy Labels

Yisen Wang, Weiyang Liu, Xingjun Ma +4

Large-scale datasets possessing clean label annotations are crucial for training Convolutional Neural Networks (CNNs). However, labeling large-scale data can be very costly and err…

cs.LG2024

Unlearnable Examples For Time Series

Yujing Jiang, Xingjun Ma, Sarah Monazam Erfani +1

Unlearnable examples (UEs) refer to training samples modified to be unlearnable to Deep Neural Networks (DNNs). These examples are usually generated by adding error-minimizing nois…

cs.CV2024

Visual-Text Cross Alignment: Refining the Similarity Score in Vision-Language Models

Jinhao Li, Haopeng Li, Sarah Erfani +3

It has recently been discovered that using a pre-trained vision-language model (VLM), e.g., CLIP, to align a whole query image with several finer text descriptions generated by a l…

cs.LG2022

MurTree: Optimal Classification Trees via Dynamic Programming and Search

Emir Demirović, Anna Lukina, Emmanuel Hebrard +5

Decision tree learning is a widely used approach in machine learning, favoured in applications that require concise and interpretable models. Heuristic methods are traditionally us…

cs.CV2020

Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networks

Yunfei Liu, Xingjun Ma, James Bailey +1

Recent studies have shown that DNNs can be compromised by backdoor attacks crafted at training time. A backdoor attack installs a backdoor into the victim model by injecting a back…

stat.ML2018

Online Cluster Validity Indices for Streaming Data

Masud Moshtaghi, James C. Bezdek, Sarah M. Erfani +2

Cluster analysis is used to explore structure in unlabeled data sets in a wide range of applications. An important part of cluster analysis is validating the quality of computation…

cs.LG2020

Divide and Learn: A Divide and Conquer Approach for Predict+Optimize

Ali Ugur Guler, Emir Demirovic, Jeffrey Chan +3

The predict+optimize problem combines machine learning ofproblem coefficients with a combinatorial optimization prob-lem that uses the predicted coefficients. While this problemcan…

cs.LG2018

Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionality

Xingjun Ma, Bo Li, Yisen Wang +6

Deep Neural Networks (DNNs) have recently been shown to be vulnerable against adversarial examples, which are carefully crafted instances that can mislead DNNs to make errors durin…

stat.ML2016

A Framework to Adjust Dependency Measure Estimates for Chance

Simone Romano, Nguyen Xuan Vinh, James Bailey +1

Estimating the strength of dependency between two variables is fundamental for exploratory analysis and many other applications in data mining. For example: non-linear dependencies…

cs.CV2025

Reasoning Like Experts: Leveraging Multimodal Large Language Models for Drawing-based Psychoanalysis

Xueqi Ma, Yanbei Jiang, Sarah Erfani +4

Multimodal Large Language Models (MLLMs) have demonstrated exceptional performance across various objective multimodal perception tasks, yet their application to subjective, emotio…

cs.LO2008

Logical Queries over Views: Decidability and Expressiveness

James Bailey, Guozhu Dong, Anthony Widjaja To

We study the problem of deciding satisfiability of first order logic queries over views, our aim being to delimit the boundary between the decidable and the undecidable fragments o…

cs.LG2019

Quality Evaluation of GANs Using Cross Local Intrinsic Dimensionality

Sukarna Barua, Xingjun Ma, Sarah Monazam Erfani +2

Generative Adversarial Networks (GANs) are an elegant mechanism for data generation. However, a key challenge when using GANs is how to best measure their ability to generate reali…

cs.CV2026

Latent Video Prediction Learns Better World Models

Ali J Alrasheed, Aryan Yazdan Parast, Basim Azam +2

Self-supervised video models are increasingly framed as world models, yet their evaluation remains largely confined to a single top-1 accuracy score on clean benchmarks. This leave…

cs.CV2026

GeoDetect: Geometric Adversarial Detection for VLPs

Afsaneh Hasanebrahimi, Hanxun Huang, Christopher Leckie +2

The paper introduces GeoDetect, a method that uses geometric properties of vision‑language model embeddings to detect adversarial examples by measuring how far they deviate from th…

#adversarial detection#vision-language models#geometric analysis#embedding space anisotropy
cs.SD2026

A Geometric Perspective on Composable Emotion Steering in Text-to-Speech Models

Siyi Wang, James Bailey, Ting Dang

While prior work has explored emotion control in hybrid text-to-speech systems, the geometric properties of these modules, and their implications for steerability, remain poorly un…

cs.LG2026

Spectral Flattening Is All Muon Needs: How Orthogonalization Controls Learning Rate and Convergence

Tien-Phat Nguyen, Truong Nguyen, Minh-Phuc Truong +3

Muon orthogonalizes the momentum buffer before each update, replacing its singular values with ones via Newton-Schulz iterations. This simple change lets Muon tolerate far larger l…

cs.CV2025

Parsimonious Dataset Construction for Laparoscopic Cholecystectomy Structure Segmentation

Yuning Zhou, Henry Badgery, Matthew Read +2

Labeling has always been expensive in the medical context, which has hindered related deep learning application. Our work introduces active learning in surgical video frame selecti…

cs.LG2025

Detecting Backdoor Samples in Contrastive Language Image Pretraining

Hanxun Huang, Sarah Erfani, Yige Li +2

Contrastive language-image pretraining (CLIP) has been found to be vulnerable to poisoning backdoor attacks where the adversary can achieve an almost perfect attack success rate on…

q-bio.QM2021

De Novo Molecular Generation with Stacked Adversarial Model

Yuansan Liu, James Bailey

Generating novel drug molecules with desired biological properties is a time consuming and complex task. Conditional generative adversarial models have recently been proposed as pr…