Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2606. 26891v1 Announce Type: cross Abstract: Concept Bottleneck Models (CBMs) promise transparent reasoning by predicting through human-interpretable concepts, yet their effectiveness fundamentally depends on how well visual and textual representations are aligned or matched.
arXiv:2606. 26793v1 Announce Type: cross Abstract: Multimodal agentic retrieval-augmented generation (RAG) systems expand the attack surface beyond prompt injection to include text poisoning, image injection, direct-query attacks, and orchestrator-level tool manipulation.
arXiv:2606. 26857v1 Announce Type: new Abstract: The interpretation phase of life cycle assessment often lacks structured mechanisms for translating quantified improvement opportunities addressing environmental hotspots into actionable strategic pathways under technological, social, and policy uncertainty.
arXiv:2606. 26204v1 Announce Type: new Abstract: Floods frequently impact regions around the world.
arXiv:2606. 26922v1 Announce Type: cross Abstract: Continuous driver monitoring in automated vehicles requires low-latency inference while avoiding unsafe decisions under uncertain driver states.
arXiv:2606. 26523v1 Announce Type: new Abstract: We develop a framework for interpreting AI systems as agents, drawing on the philosophical tradition of radical interpretation and the tools of mechanistic interpretability.
arXiv:2606. 27215v1 Announce Type: new Abstract: Deep learning models have achieved impressive performance across various fields but remain vulnerable to adversarial inputs, particularly in NLP, where such attacks can have significant real-world consequences.
arXiv:2606. 27095v1 Announce Type: cross Abstract: Cold-start exemplar-free class-incremental learning requires learning a growing set of classes without replay, external pretraining, or a large initial task.
arXiv:2606. 26994v1 Announce Type: cross Abstract: Existing referring video segmentation methods often treat a video as a single event consisting of multiple images, overlooking the fact that a video typically contains multiple distinct events.
arXiv:2606. 26987v1 Announce Type: cross Abstract: Recent work identified emotion vectors in Claude Sonnet 4.
arXiv:2606. 26981v1 Announce Type: cross Abstract: Synthesizing human motion from textual descriptions is essential for immersive digital applications, yet existing methods face a persistent trade-off between semantic fidelity and physical realism.
arXiv:2410. 01574v4 Announce Type: replace-cross Abstract: The rapid advancement of Generative Artificial Intelligence (GenAI) capabilities is accompanied by a concerning rise in its misuse.
arXiv:2604. 04535v2 Announce Type: replace Abstract: Modern machine learning systems, such as generative models and recommendation systems, often evolve through a cycle of deployment, user interaction, and periodic model updates.
arXiv:2509. 07295v4 Announce Type: replace-cross Abstract: Unified multimodal models (UMMs) unify visual understanding and generation within a single architecture.
arXiv:2606. 26369v1 Announce Type: cross Abstract: Scoring functions are used to represent the relevance of individual documents.
arXiv:2606. 26387v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) extend large language models (LLMs) with visual perception, enabling joint reasoning over images and text.
arXiv:2606. 26396v1 Announce Type: new Abstract: Pre-trained transformers have demonstrated remarkable generalization abilities, at times extending beyond the scope of their training data.
arXiv:2606. 26519v1 Announce Type: new Abstract: Large language models (LLMs) can make scientific software easier to use.
arXiv:2606. 26205v1 Announce Type: new Abstract: Patients increasingly seek medication information online, yet safety knowledge for psychiatric drugs is split between regulatory adverse-event records, which are authoritative but abstract, and patient narratives, which are experience-near but unvalidated.
arXiv:2606. 27023v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) applied to Medical Visual Question Answering (VQA) tend to produce overconfident outputs regardless of actual correctness, and existing verbalized confidence calibration methods, developed primarily for text only LLMs, do not account for the multimodal nature of medical image understanding.