Archives AI News

Caught in the Act: a mechanistic approach to detecting deception

arXiv:2508.19505v2 Announce Type: replace Abstract: Sophisticated instrumentation for AI systems might have indicators that signal misalignment from human values, not unlike a “check engine” light in cars. One such indicator of misalignment is deceptiveness in generated responses. Future AI instrumentation…

Programmable Cognitive Bias in Social Agents

arXiv:2509.13588v1 Announce Type: new Abstract: This paper introduces CoBRA, a novel toolkit for systematically specifying agent behavior in LLM-based social simulation. We found that conventional approaches that specify agent behaviors through implicit natural language descriptions cannot yield consistent behaviors across…

Apertus: Democratizing Open and Compliant LLMs for Global Language Environments

arXiv:2509.14233v1 Announce Type: cross Abstract: We present Apertus, a fully open suite of large language models (LLMs) designed to address two systemic shortcomings in today’s open model ecosystem: data compliance and multilingual representation. Unlike many prior models that release weights…

Gen AI in Proof-based Math Courses: A Pilot Study

arXiv:2509.13570v1 Announce Type: new Abstract: With the rapid rise of generative AI in higher education and the unreliability of current AI detection tools, developing policies that encourage student learning and critical thinking has become increasingly important. This study examines student…

SteeringControl: Holistic Evaluation of Alignment Steering in LLMs

arXiv:2509.13450v1 Announce Type: new Abstract: We introduce SteeringControl, a benchmark for evaluating representation steering methods across core alignment objectives–bias, harmful generation, and hallucination–and their effects on secondary behaviors such as sycophancy and commonsense morality. While prior alignment work often highlights…

Zero-Knowledge Proofs in Sublinear Space

arXiv:2509.05326v2 Announce Type: replace-cross Abstract: Zero-knowledge proofs allow verification of computations without revealing private information. However, existing systems require memory proportional to the computation size, which has historically limited use in large-scale applications and on mobile and edge devices. We…

From Next Token Prediction to (STRIPS) World Models — Preliminary Results

arXiv:2509.13389v1 Announce Type: new Abstract: We consider the problem of learning propositional STRIPS world models from action traces alone, using a deep learning architecture (transformers) and gradient descent. The task is cast as a supervised next token prediction problem where…