Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2607. 13162v1 Announce Type: cross Abstract: What a language model will and will not do is largely set during post-training, but which behaviors it expresses, hides, or resists is not revealed by prompting alone.
arXiv:2511. 13592v2 Announce Type: replace-cross Abstract: The existing method of GS-PowerOpt solves the non-convex optimization problem of the form $\max_{\boldsymbol{x} \in \mathbb{R}^d} f(\boldsymbol{x})$ through maximizing a Gaussian-smoothed surrogate $F_{N,\sigma}(\boldsymbol{\mu}) = \mathbb{E}_{\boldsymbol{x}\sim\mathcal{N}(\boldsymbol{\mu},\sigma^2 I_d)}[e^{N f(\boldsymbol{x})}]$.
arXiv:2607. 13436v1 Announce Type: new Abstract: Imitation learning (IL) has achieved remarkable success in complex decision-making tasks.
arXiv:2607. 13081v1 Announce Type: cross Abstract: We present nsfaguard, a guardrail framework for securing agentic AI systems against operational threats, such as prompt injection, sensitive information extraction, malicious code requests, dangerous tool misuse, and resource exhaustion.
arXiv:2607. 13899v1 Announce Type: new Abstract: We propose the AIMO Interpretability Challenge, a competition on distinguishing robust from spurious reasoning in frontier mathematical language models based on the models' internal mechanisms.
arXiv:2607. 13103v1 Announce Type: cross Abstract: Knowledge tracing (KT) aims to predict students' future performance by modeling their evolving knowledge states from historical interactions.
arXiv:2601. 02023v2 Announce Type: replace-cross Abstract: As Large Language Models (LLMs) increasingly utilize massive context windows as working memory for autonomous tasks, their reliability fluctuates significantly depending on how information is distributed in real-world corpora.
arXiv:2607. 13978v1 Announce Type: cross Abstract: Music-driven dance generation aims to produce human motion that is both rhythmically synchronized and semantically consistent with music.
arXiv:2607. 13039v1 Announce Type: cross Abstract: Safety evaluations for dual-use biology assistants often measure base-model capability, refusal behavior, or jailbreak success.
arXiv:2607. 13172v1 Announce Type: new Abstract: We address the problem of safely training an agent policy and deploying a good and safe policy, in settings where the environment dynamics are unknown and no suitable reward function is available.
arXiv:2511. 01680v4 Announce Type: replace-cross Abstract: Social scientists are increasingly turning to unstructured datasets to unlock new empirical insights, e.
arXiv:2508. 12466v2 Announce Type: replace-cross Abstract: Traditional multimodal learning approaches rely on alignment pre-training to bridge vision and language modalities, typically by projecting visual features into discrete text token spaces using large-scale image--text data.
arXiv:2510. 26707v2 Announce Type: replace-cross Abstract: As LLMs occupy an increasingly important role in society, they are more and more confronted with questions that require them not only to draw on their general knowledge but also to align with certain human value systems.
arXiv:2607. 13414v1 Announce Type: cross Abstract: Non-expansive two-time-scale stochastic approximation is governed by a slow stochastic Krasnoselskii--Mann fixed-point iteration rather than by contraction to a unique equilibrium.
arXiv:2607. 13203v1 Announce Type: cross Abstract: False alarms remain a major barrier to deploying network intrusion detection systems (NIDS).
arXiv:2607. 13046v1 Announce Type: new Abstract: We develop a framework for the information discarded by machine learning models whose inputs carry a Lie group action.
arXiv:2607. 13093v1 Announce Type: cross Abstract: On-device LLM inference faces a trilemma of response latency, limited hardware resources and user privacy.
arXiv:2603. 10330v2 Announce Type: replace-cross Abstract: Autonomous driving in complex traffic requires planners that generalize beyond hand-crafted rules, motivating data-driven approaches that learn behavior from expert demonstrations.
arXiv:2510. 10002v3 Announce Type: replace Abstract: As large language models (LLMs) are increasingly deployed in sensitive everyday contexts -- offering personal advice, mental health support, and moral guidance -- understanding their behavior in navigating complex moral reasoning is essential.
arXiv:2607. 13056v1 Announce Type: cross Abstract: Current vision-language-action (VLA) benchmarks primarily evaluate isolated manipulation skills while leaving human-robot interaction structure largely unmodeled.