Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2610.34450v1 Announce Type: cross Abstract: Decentralized multi-agent systems (MAS) enable large language model agents to interact without centralized control. Agents need to adapt to other age...
arXiv:2610.34965v1 Announce Type: cross Abstract: Automatic red teaming increasingly relies on large language models (LLMs) to iteratively diagnose failed attacks and propose improved adversarial pro...
arXiv:2610.35188v1 Announce Type: cross Abstract: Large language models increasingly act through code, modifying repositories, executing programs, and interacting with services in agent form. Their a...
arXiv:2610.35725v1 Announce Type: cross Abstract: The gap between AI capability deployment and safety assurance is structural and cannot be closed through internal evaluations alone. Analyzing docume...
arXiv:2601.06640v2 Announce Type: replace Abstract: The transition towards sixth-generation (6G) wireless networks necessitates autonomous orchestration mechanisms capable of translating high-level o...
arXiv:2609.33357v2 Announce Type: replace Abstract: Reliable automated research requires agents to vet data, verify analyses, and generate hypotheses grounded in trustworthy evidence, potentially red...
arXiv:2610.26185v2 Announce Type: replace Abstract: Despite rapid advances in large language models (LLMs), their deployment in the sports and health domain remains high-risk due to the absence of do...
arXiv:2510.12044v2 Announce Type: replace-cross Abstract: Preference tuning typically distributes adapter updates across every Transformer layer. We study how the placement of a fixed LoRA budget aff...
arXiv:2512.04231v2 Announce Type: replace-cross Abstract: Assistive robots operating in unstructured environments must reason not only about what objects are, but how they can be used, including non-...
arXiv:2512.21711v2 Announce Type: replace-cross Abstract: Latent tokens are gaining attention for enhancing reasoning in large language models (LLMs), yet their internal mechanisms remain unclear. Th...
arXiv:2610.27293v2 Announce Type: replace Abstract: Safe Multi-Agent Reinforcement Learning (SMARL) aims to learn cooperative control policies under prescribed safety constraints. Al- though existing...
arXiv:2602.13444v2 Announce Type: replace-cross Abstract: Recent vision-language-action (VLA) models can generate plausible end-effector motions, yet they often fail in long-horizon, contact-rich tas...
arXiv:2602.23566v4 Announce Type: replace-cross Abstract: We study generative modeling of graphs with recurring subgraph motifs. We propose Flowette, a continuous flow matching framework that employs...
arXiv:2602.08054v2 Announce Type: replace-cross Abstract: Offline reinforcement learning (RL) offers a promising framework for deploying autonomous systems in safety-critical settings without the ris...
arXiv:2604.18161v2 Announce Type: replace-cross Abstract: In policy gradient reinforcement learning, access to a differentiable model enables 1st-order gradient estimation that accelerates learning c...
arXiv:2605.06908v2 Announce Type: replace-cross Abstract: Adaptive test-time compute for LLM agents aims to invoke extra computation only when it improves performance. Existing methods typically use...
arXiv:2610.26645v2 Announce Type: replace-cross Abstract: As people increasingly rely on large language models (LLMs) for information, model providers need to communicate meaningful estimates of whet...
arXiv:2610.30861v2 Announce Type: replace-cross Abstract: Few-step autoregressive video generation enables low-latency streaming, yet prevailing training pipelines separate causal adaptation from dis...
arXiv:2610.31704v1 Announce Type: new Abstract: Non-Destructive Inspection (NDI) is a cornerstone of modern quality assurance, critical for ensuring the safety and reliability of infrastructure and m...
arXiv:2610.31786v1 Announce Type: new Abstract: Federated unlearning (FU) aims to remove a designated target client's influence from a trained model without the prohibitive cost of retraining from sc...