相关论文: Corrigibility as a Singular Target: A Vision for I…

Continual learning--the ability to acquire, retain, and refine knowledge over time--has always been fundamental to intelligence, both human and artificial. Historically, different AI paradigms have acknowledged this need, albeit with…

Model-based quantum optimal control promises to solve a wide range of critical quantum technology problems within a single, flexible framework. The catch is that highly-accurate models are needed if the optimized controls are to meet the…

The scaling of model and data sizes has reshaped the AI landscape, establishing finetuning pretrained models as the standard paradigm for solving downstream tasks. However, dominant finetuning methods typically rely on weight adaptation,…

It's widely expected that humanity will someday create AI systems vastly more intelligent than us, leading to the unsolved alignment problem of "how to control superintelligence." However, this commonly expressed problem is not only…

Static content-based AI value alignment is insufficient for robust alignment under capability scaling, distributional shift, and increasing autonomy. This holds for any approach that treats alignment as optimizing toward a fixed formal…

Foundation Models (FMs) are large-scale, pre-trained artificial intelligence (AI) systems that have revolutionized natural language processing and computer vision, and are now advancing geospatial analysis and Earth Observation (EO). They…

Recent tabular Foundational Models (FM) such as TabPFN and TabICL, leverage in-context learning to achieve strong performance without gradient updates or fine-tuning. However, their robustness to adversarial manipulation remains largely…

As large language models (LLMs) continue to evolve, ensuring their alignment with human goals and values remains a pressing challenge. A key concern is \textit{instrumental convergence}, where an AI system, in optimizing for a given…

Recently, major AI providers such as Google and OpenAI have introduced Finetuning-as-a-Service (FaaS), which allows users to customize Large Language Models (LLMs) using their own data. However, this service is vulnerable to safety…

Sim-to-real transfer for contact-rich manipulation remains challenging due to the inherent discrepancy in contact dynamics. While existing methods often rely on costly real-world data or utilize blind compliance through fixed controllers,…

Reinforcement learning (RL) algorithms are highly sensitive to reward function specification, which remains a central challenge limiting their broad applicability. We present ARM-FM: Automated Reward Machines via Foundation Models, a…

Researchers at artificial intelligence labs and universities are concerned that highly capable artificial intelligence (AI) systems may erode human control by pursuing instrumental goals. Existing mitigations remain largely technical and…

The integration of foundation models (FMs) into robotics has accelerated real-world deployment, while introducing new safety challenges arising from open-ended semantic reasoning and embodied physical action. These challenges require safety…

In equality-constrained optimization, a standard regularity assumption is often associated with feasible point methods, namely the gradients of constraints are linearly independent. In practice, the regularity assumption may be violated. To…

Reinforcement learning has been successfully applied to the problem of tuning PID controllers in several applications. The existing methods often utilize function approximation, such as neural networks, to update the controller parameters…

Reinforcement finetuning (RFT) has emerged as a powerful paradigm for unlocking reasoning capabilities in large language models. However, we identify a critical trade-off: while unconstrained RFT achieves strong reasoning performance, it…

Fine-tuning LLMs on benign data can still degrade alignment and adversarial robustness, yet direct analysis of the role of fine-tuning objectives in shaping these safety outcomes remain limited. We present a controlled comparison of six…

Machine fault diagnosis (FD) is a critical task for predictive maintenance, enabling early fault detection and preventing unexpected failures. Despite its importance, existing FD models are operation-specific with limited generalization…

Fine-tuning is a widely used strategy for adapting pre-trained models to new tasks, yet its methodology and theoretical properties in high-dimensional nonparametric settings with variable selection have not yet been developed. We propose a…

The AI alignment problem, which focusses on ensuring that artificial intelligence (AI), including AGI and ASI, systems act according to human values, presents profound challenges. With the progression from narrow AI to Artificial General…