Tootfinder

BadViM: Backdoor Attack against Vision Mamba
Yinghao Wu, Liyan Zhang
https://arxiv.org/abs/2507.00577 https://arxiv.org/pdf/2507.0057…

BadViM: Backdoor Attack against Vision Mamba
Vision State Space Models (SSMs), particularly architectures like Vision Mamba (ViM), have emerged as promising alternatives to Vision Transformers (ViTs). However, the security implications of this novel architecture, especially their vulnerability to backdoor attacks, remain critically underexplored. Backdoor attacks aim to embed hidden triggers into victim models, causing the model to misclassify inputs containing these triggers while maintaining normal behavior on clean inputs. This paper i…

Push or Light: Nudging Standing to Break Prolonged Sitting
Sohshi Yoshida, Ko Watanabe, Andreas Dengel, Shoya Ishimaru, Shingo Ata, Manato Fujimoto
https://arxiv.org/abs/2507.08659

Push or Light: Nudging Standing to Break Prolonged Sitting
Prolonged sitting is a health risk leading to metabolic and cardiovascular diseases. To combat this, various "nudging" strategies encourage stand-ups. Behavior change triggers use explicit prompts such as smartphone push notifications or light controls. However, comparisons of the effects of such interactions, discomfort, and user context have not yet been performed. The present study evaluated these methods in a mixed design experiment with 15 college students. Three intervention methods (none…

Execution-Aware Program Reduction for WebAssembly via Record and Replay
Doehyun Baek, Daniel Lehmann, Ben L. Titzer, Sukyoung Ryu, Michael Pradel
https://arxiv.org/abs/2506.07834 …

Execution-Aware Program Reduction for WebAssembly via Record and Replay
WebAssembly (Wasm) programs may trigger bugs in their engine implementations. To aid debugging, program reduction techniques try to produce a smaller variant of the input program that still triggers the bug. However, existing execution-unaware program reduction techniques struggle with large and complex Wasm programs, because they rely on static information and apply syntactic transformations, while ignoring the valuable information offered by the input program's execution behavior. We presen…

Hierarchical Temporal Point Process Modeling of Aggressive Behavior Onset in Psychiatric Inpatient Youth with Autism for Branching Factor Estimation
Michael Potter, Michael Everett, Deniz Erdogmus, Yuna Watanabe, Tales Imbiriba, Matthew S. Goodwin
https://arxiv.org/abs/2507.12424

Hierarchical Temporal Point Process Modeling of Aggressive Behavior Onset in Psychiatric Inpatient Youth with Autism for Branching Factor Estimation
Aggressive behavior in autistic inpatient youth often arises in temporally clustered bursts complicating efforts to distinguish external triggers from internal escalation. The sample population branching factor-the expected number of new onsets triggered by a given event-is a key summary of self-excitation in behavior dynamics. Prior pooled models overestimate this quantity by ignoring patient-specific variability. We addressed this using a hierarchical Hawkes process with an exponential kernel…

InverTune: Removing Backdoors from Multimodal Contrastive Learning Models via Trigger Inversion and Activation Tuning
Mengyuan Sun, Yu Li, Yuchen Liu, Bo Du, Yunjie Ge
https://arxiv.org/abs/2506.12411

InverTune: Removing Backdoors from Multimodal Contrastive Learning Models via Trigger Inversion and Activation Tuning
Multimodal contrastive learning models like CLIP have demonstrated remarkable vision-language alignment capabilities, yet their vulnerability to backdoor attacks poses critical security risks. Attackers can implant latent triggers that persist through downstream tasks, enabling malicious control of model behavior upon trigger presentation. Despite great success in recent defense mechanisms, they remain impractical due to strong assumptions about attacker knowledge or excessive clean data requir…

Endo-exo classification of episodic rock creep in deep mines: Implications for forecasting catastrophic failure
Qinghua Lei, Daniel Francois Malan, Didier Sornette
https://arxiv.org/abs/2506.12547

Endo-exo classification of episodic rock creep in deep mines: Implications for forecasting catastrophic failure
Rock masses in deep underground environments under high in-situ stress often exhibit episodic creep behavior, driven by complex interactions between external perturbation and internal reorganization. The causes of these creep episodes and their link to potential catastrophic failure remain poorly understood. Here, we present a novel 'endo-exo' framework for analyzing episodic rock creep in deep underground mines, capturing the interplay between exogenous triggers (e.g., blasting and excavation)…

3S-Attack: Spatial, Spectral and Semantic Invisible Backdoor Attack Against DNN Models
Jianyao Yin, Luca Arnaboldi, Honglong Chen, Pascal Berrang
https://arxiv.org/abs/2507.10733 …

3S-Attack: Spatial, Spectral and Semantic Invisible Backdoor Attack Against DNN Models
Backdoor attacks involve either poisoning the training data or directly modifying the model in order to implant a hidden behavior, that causes the model to misclassify inputs when a specific trigger is present. During inference, the model maintains high accuracy on benign samples but misclassifies poisoned samples into an attacker-specified target class. Existing research on backdoor attacks has explored developing triggers in the spatial, spectral (frequency), and semantic (feature) domains, a…

TED-LaST: Towards Robust Backdoor Defense Against Adaptive Attacks
Xiaoxing Mo, Yuxuan Cheng, Nan Sun, Leo Yu Zhang, Wei Luo, Shang Gao
https://arxiv.org/abs/2506.10722

TED-LaST: Towards Robust Backdoor Defense Against Adaptive Attacks
Deep Neural Networks (DNNs) are vulnerable to backdoor attacks, where attackers implant hidden triggers during training to maliciously control model behavior. Topological Evolution Dynamics (TED) has recently emerged as a powerful tool for detecting backdoor attacks in DNNs. However, TED can be vulnerable to backdoor attacks that adaptively distort topological representation distributions across network layers. To address this limitation, we propose TED-LaST (Topological Evolution Dynamics agai…