Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2609.07107v1 Announce Type: new Abstract: Micro-dramas, characterized by ultra-short durations and hyper-dense storylines, pose unique challenges for video understanding that conventional bench...
arXiv:2609.07128v1 Announce Type: new Abstract: Reliable risk assessment remains a central challenge for Autonomous Vehicles (AVs). Despite advances in automation, passenger cognition provides a non-...
arXiv:2609.05946v1 Announce Type: new Abstract: Interpretability and fairness are two of the most emphasized dimensions in trustworthy artificial intelligence (AI). Various explainable AI methods hav...
arXiv:2609.06011v1 Announce Type: cross Abstract: Omni-modal large language models (OLLMs) jointly process vision, audio, and text, yet their modality bias under cross-modal conflict remains underexp...
arXiv:2609.07984v1 Announce Type: new Abstract: Process mining has long turned event logs into process knowledge: discovered models, conformance evidence, bottleneck diagnoses, and runtime prediction...
arXiv:2609.07998v1 Announce Type: new Abstract: We study the control of Markov decision processes in which the quality of a policy is evaluated by a dynamic, time-consistent Markov risk measure rathe...
The paper presents a distributed primal‑dual algorithm for quadratic Generalized Nash Equilibrium Problems (GNEPs) that eliminates the requirement for multiplier consensus. By removing shared multipliers, the method reduces communication overhead and enhances privacy, allowing different initializations to converge to distinct GNEs, including non‑variational equilibria. Convergence is proven under sufficient conditions using an input‑to‑state stability (ISS) framework.
arXiv:2609.08082v1 Announce Type: new Abstract: Preference-based fine-tuning methods such as RLHF and DPO require substantial compute and large preference datasets. They also need direct access to th...
arXiv:2109.02355v2 Announce Type: replace Abstract: The last decade of progress in machine learning (ML), especially the deep learning era, has raised a number of scientific questions that challenge...
arXiv:2609.06761v1 Announce Type: new Abstract: While diffusion models excel at capturing complex data distributions, scientific discovery often requires steering generation toward specific, uncharac...
arXiv:2609.08188v1 Announce Type: new Abstract: Vision-language models (VLMs) augmented with retrieval-augmented generation (RAG) benefit from access to external evidence. However, standard retriever...
arXiv:2602.19600v2 Announce Type: replace Abstract: Many high-dimensional datasets concentrate near a low-dimensional structure embedded in the ambient space. Generative models for such data must con...
arXiv:2609.07147v1 Announce Type: new Abstract: Federated learning, as a privacy-preserving distributed machine learning paradigm, faces significant threats from backdoor attacks. Compared to central...
arXiv:2609.08445v1 Announce Type: new Abstract: Working entirely on topologically anonymized embeddings, we perform fraud detection using iterative rounds of unsupervised filtering followed by superv...
arXiv:2609.08253v1 Announce Type: new Abstract: Diffusion models have shown remarkable performance on diverse generation tasks. Recent work finds that imposing representation alignment on the hidden...
arXiv:2609.07493v1 Announce Type: new Abstract: In this paper, we propose a class-wise dimension (channel) selection framework for Multivariate Time Series Classification (MTSC). Rather than applying...
arXiv:2609.06484v1 Announce Type: cross Abstract: Planning with a generative model aims to estimate the value of a state using as few simulator calls as possible. SmoothCruiser achieves problem-indep...
arXiv:2609.06490v1 Announce Type: cross Abstract: Recursive Super-Resolution (SR) extends fixed-scale SR to extreme magnification by repeatedly feeding predictions back into the same model, analogous...
arXiv:2609.09188v1 Announce Type: new Abstract: Lensless near-eye sensing is often described as privacy-friendly because its coded measurements are visually unintelligible. Yet visual unintelligibili...
arXiv:2609.08861v1 Announce Type: new Abstract: Benchmark scores are a central currency in model releases: they inform purchasing decisions, shape public trust, and influence policy. Yet, a key assum...