Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2610.16049v1 Announce Type: cross Abstract: Existing audio deepfake detection (ADD) methods frequently exhibit limited generalization, as fixed bottleneck architectures fail to adapt to the het...
arXiv:2610.16135v1 Announce Type: cross Abstract: Omni-modal large language models (Omni-LLMs) extend LLMs to process heterogeneous modalities (e.g., vision, audio, and IMU signals) alongside languag...
arXiv:2610.16411v1 Announce Type: cross Abstract: Deploying robots in human-shared environments requires deciding not only how to navigate, but also when dedicated social signals are worth emitting....
arXiv:2610.20472v1 Announce Type: new Abstract: Cultural attitudes and social norms evolve over time, yet most cultural benchmarks for large language models (LLMs) treat culture as a static attribute...
arXiv:2610.16911v1 Announce Type: new Abstract: Does providing additional context to automated AI auditors improve their effectiveness? Using the Petri tool to audit LLM coding agents, we test the im...
arXiv:2610.16545v1 Announce Type: cross Abstract: Feature caching has become a mainstream training-free strategy for accelerating video diffusion models by reusing intermediate outputs across denoisi...
arXiv:2610.17641v1 Announce Type: new Abstract: Bias scores in language-model evaluations often conflate what a model prefers with whether it answers at all. We identify this confound in \textsc{BBQ}...
arXiv:2610.18389v1 Announce Type: cross Abstract: Large language models are increasingly used as judges to reduce reliance on costly human evaluation, yet their scores can systematically disagree wit...
arXiv:2610.18141v1 Announce Type: new Abstract: Omnimodal large language models (OLLMs) are susceptible to cross-modal hallucinations under audio-visual interference. We find that, for a substantial...
arXiv:2610.17702v1 Announce Type: new Abstract: Evaluating the quality of process models discovered from event logs is a fundamental challenge in process mining. Traditional evaluation relies on conf...
arXiv:2610.18603v1 Announce Type: new Abstract: Benchmarks report what an Artificial Intelligence (AI) agent can do on a task suite, but not whether one released version has been engineered with the...
arXiv:2610.18304v1 Announce Type: new Abstract: Guardrail approaches that rely on reasoning-based large language models (LLM) improve safety decisions, but come with substantial inference overhead du...
arXiv:2610.18127v1 Announce Type: new Abstract: Looped language models (LMs) reuse one block of layers several times per token generation, promising more latent computation at no additional storage c...
arXiv:2610.19576v1 Announce Type: new Abstract: Temporal graph data are recorded in different representations, but whether these differences also determine the scope of learned knowledge reuse remain...
arXiv:2610.19603v1 Announce Type: new Abstract: Transformers have demonstrated strong performance in long-term time series forecasting (LTSF), yet modeling non-stationary temporal dependencies remain...
arXiv:2610.18606v1 Announce Type: new Abstract: Large language models increasingly tackle tasks that require multi-step reasoning, where errors in intermediate claims can affect the final answer. Sel...
arXiv:2610.18119v1 Announce Type: cross Abstract: {\em Specification-Guided Reinforcement Learning (RL)} has largely concerned itself with \emph{learning} an optimal policy against a formal specifica...
arXiv:2610.17269v1 Announce Type: cross Abstract: Learning symmetries through data augmentation offers an alternative to architectural equivariance, but does not guarantee that a network respects tho...
arXiv:2610.20009v1 Announce Type: new Abstract: Scaling self-supervised contrastive reinforcement learning (CRL) to 1000-layer Euclidean neural networks has enabled state-of-the-art goal-reaching per...
arXiv:2610.20014v1 Announce Type: new Abstract: Semi-gradient temporal-difference learning is a core algorithmic component of reinforcement learning, but bootstrapping with function approximation gen...