Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2609.07738v1 Announce Type: cross Abstract: LiDAR-based 3D Single Object Tracking (3D SOT) is critical for robotic perception and navigation and aims to localize dynamic objects across frames i...
arXiv:2609.08080v1 Announce Type: new Abstract: Ensuring safety in reinforcement learning under nonstationarity requires anticipating changes in risk before they lead to unsafe behavior. Existing app...
arXiv:2609.06100v1 Announce Type: cross Abstract: Verifiable outcome rewards guide language-model post-training, but sequence-level advantages do not identify which token-level decisions should be pr...
arXiv:2609.05517v1 Announce Type: cross Abstract: Human observers prioritize visual information according to task goals. Most computational models of naturalistic viewing are gaze-trained for free vi...
arXiv:2609.05522v1 Announce Type: cross Abstract: Eye-tracking data are expensive to collect, requiring specialized hardware and controlled laboratory conditions, and difficult to share because of pr...
arXiv:2609.05531v1 Announce Type: cross Abstract: No specification says how a governed autotelic AI agent organization, where agents pursue self-generated goals inside guardrails, should be designed...
arXiv:2609.05794v1 Announce Type: cross Abstract: Open-weight large language models face a low-cost white-box threat from representation engineering attacks. Attackers can estimate refusal directions...
arXiv:2609.06934v1 Announce Type: cross Abstract: Post-hoc safety training (RLHF, DPO) is the dominant way to align large language models, yet jailbreaks (Zou et al., 2023b), fine-tuning attacks (Qi...
arXiv:2609.06951v1 Announce Type: cross Abstract: Activation steering promises modular control of language model behavior: a behavior such as politeness corresponds to a direction in a model's activa...
arXiv:2609.06433v1 Announce Type: cross Abstract: The generation of safety-critical traffic scenarios is essential for training and evaluating autonomous vehicles. Prior approaches typically perturb...
arXiv:2609.05459v1 Announce Type: new Abstract: Reinforcement learning and large language models often struggle to accurately capture the causal mechanics of game environments. Standard reinforcement...
arXiv:2609.06543v1 Announce Type: new Abstract: Enterprise AI is evolving into an Enterprise Operating System where autonomous AI agents can plan, reason, use memory, invoke tools, execute workflows,...
arXiv:2609.07174v1 Announce Type: new Abstract: Efficient, fully automatic, and physically plausible 4D Gaussian synthesis is an important goal for dynamic scene generation. Recent physics-based meth...
arXiv:2609.07559v1 Announce Type: new Abstract: How do you validate a cheap, deterministic proxy for an oracle that is expensive, rate-limited, and non-stationary? We present a protocol built on adve...
arXiv:2609.07713v1 Announce Type: new Abstract: Generative and agentic AI are reshaping both the production and evaluation of scientific research. These developments are often studied separately, as...
arXiv:2609.07731v1 Announce Type: new Abstract: We show that ordinary business language --- "maximize profitability" --- induces profit-oriented ambiguity resolution: LLMs systematically dismiss ambi...
arXiv:2609.07901v1 Announce Type: new Abstract: Weight quantization largely determines the economics of serving open-weight LLMs. Its costs are usually assessed with capability benchmarks, on which 4...
arXiv:2609.05988v1 Announce Type: cross Abstract: Recurrent reasoning models (RRMs) can solve structured problems, achieving easy-to-hard generalization through iterative computation in hidden space....
arXiv:2609.06052v1 Announce Type: cross Abstract: Autonomous agent systems increasingly depend on reusable skill abstractions for consolidating experiential knowledge and domain expertise. These arti...
arXiv:2609.06085v1 Announce Type: cross Abstract: Understanding the joint dynamics of prices and trades is central to market microstructure, where returns and order flow interact through nonlinear an...