arXiv AI
Jul 1
Wait, am I Being Fair? Characterizing Deductive Stereotyping and Mitigating It with Fair-GCG
arXiv:2606. 30989v1 Announce Type: cross Abstract: Warning: This paper contains several toxic and offensive statements.
AI safety and alignment
Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv Machine Learning
Jul 1
Multistage Defer Trees for Hybrid Interpretability: If at First You Can't Succeed, Tree Again
arXiv:2606. 30995v1 Announce Type: new Abstract: Recent work has shown that well-optimized individual decision trees can match complex black box models in some settings, primarily in noisy domains.
arXiv Machine Learning
Jul 1
On Optimizing Multimodal Jailbreaks for Spoken Language Models
arXiv:2603. 19127v2 Announce Type: replace Abstract: As Spoken Language Models (SLMs) integrate speech and text modalities, they inherit the safety vulnerabilities of their LLM backbone while introducing an expanded attack surface.
arXiv AI
Jul 1
Paper2Rebuttal: A Multi-Agent Framework for Transparent Author Response Assistance
arXiv:2601. 14171v2 Announce Type: replace Abstract: Writing effective rebuttals is a high-stakes task that demands more than linguistic fluency, as it requires precise alignment between reviewer intent and manuscript details.
arXiv AI
Jul 1
Towards Inclusive Mobility Modeling: Characterizing and Evaluating Elderly Trajectory Patterns in Urban Systems
arXiv:2606. 31207v1 Announce Type: new Abstract: The rapid advance of smart cities increasingly depends on trajectory data mining, yet underrepresented demographic groups, particularly the elderly, are often sparsely represented in public mobility datasets.
arXiv AI
Jul 1
Layout-Conditioned Autoregressive Text-to-Image Generation via Structured Masking
arXiv:2509. 12046v2 Announce Type: replace-cross Abstract: Although autoregressive (AR) models have demonstrated remarkable success in image generation, extending these models to layout-conditioned generation remains challenging due to the sparse nature of layout conditions and the risk of feature entanglement.
arXiv AI
Jul 1
TreeAgent: A Generalizable Multi-Agent Framework for Automated Bias Labeling in Forestry via Compiled Expert Rules and Vision-Language Models
arXiv:2606. 31976v1 Announce Type: new Abstract: Human-labeled data are widely used as reference annotations in ML, despite known variability across annotators in many expert-driven domains.
arXiv Machine Learning
Jul 1
FedLAB: Traceable Semantic Codebooks for Federated Multimodal Graph Foundation Learning
arXiv:2606. 32016v1 Announce Type: new Abstract: Multimodal graph foundation models aim to learn reusable knowledge from graphs enriched with text, images, attributes, and relational topology, thereby supporting diverse graph-centric and modality-centric tasks.
arXiv AI
Jul 1
Verify when Uncertain: Beyond Self-Consistency in Black Box Hallucination Detection
arXiv:2502. 15845v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) often hallucinate, limiting their reliability in sensitive applications.
arXiv Machine Learning
Jul 1
ShardNet: Training Neural Controllers with Hard, Non-Convex Constraints
arXiv:2606. 30935v1 Announce Type: cross Abstract: While neural network control policies are powerful, their deployment on safety critical systems depends on ensuring that they obey strict constraints.
arXiv Machine Learning
Jul 1
Surrogate Fidelity: When Can Open LLMs Explain Closed Ones?
arXiv:2606. 32008v1 Announce Type: new Abstract: Mechanistic interpretability (MI) requires full access to model internals, yet the APIs for most widely deployed language models at best expose log-probabilities over output tokens.
arXiv AI
Jul 1
Training Therapeutic Judges and Multi-Agent Systems for Human-Aligned Mental Health Support
arXiv:2606. 30887v1 Announce Type: cross Abstract: Large language models show promise for mental health support, yet therapeutic quality improves only when evaluation functions as an actionable control signal rather than a passive metric.
arXiv AI
Jul 1
When transformers learn "impossible" languages, what do they learn?
arXiv:2606. 30815v1 Announce Type: cross Abstract: Recent work suggests that transformer language models show a bias towards human languages over unnatural ("impossible") languages argued to be unacquirable by humans.
arXiv AI
Jul 1
Revealing Safety-Critical Scenarios for UTM via Transformer
arXiv:2606. 31114v1 Announce Type: new Abstract: Unmanned Traffic Management (UTM) systems are cloud-based platforms designed to manage and coordinate multiple aerial vehicles remotely.
arXiv AI
Jul 1
Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models
arXiv:2606. 31603v1 Announce Type: cross Abstract: Semantic segmentation models struggle with data sparsity and rare or visually diverse regions, e.
arXiv Machine Learning
Jul 1
Explainable Artificial Intelligence For The Detection and Characterisation of Stage B Heart Failure
arXiv:2606. 30665v1 Announce Type: cross Abstract: Stage B heart failure is characterized by asymptomatic structural or functional cardiac abnormalities.
arXiv AI
Jul 1
Harnessing Textual Refusal Directions for Multimodal Safety
arXiv:2606. 31876v1 Announce Type: new Abstract: To improve safety in Large Language Models (LLMs) we can either perform post-training alignment or exploit refusal directions in the activation space.
arXiv AI
Jul 1
FLARE-AI: Flaw Reporting for AI
arXiv:2606. 31567v1 Announce Type: cross Abstract: Flaw reporting for deployed AI systems is fundamental to identifying system failures and improving AI safety.
arXiv AI
Jul 1
GUIDE: Resolving Domain Bias in GUI Agents through Real-Time Web Video Retrieval and Plug-and-Play Annotation
arXiv:2603. 26266v3 Announce Type: replace Abstract: Large vision-language models have endowed GUI agents with strong general capabilities for interface understanding and interaction.
arXiv AI
Jul 1
Corruption Robust Offline Reinforcement Learning with Human Feedback
arXiv:2402. 06734v2 Announce Type: replace-cross Abstract: We study data corruption robustness for reinforcement learning with human feedback (RLHF) in an offline setting.