Related papers: Safe Transformer: An Explicit Safety Bit For Inter…

Transformers are powerful for sequence modeling. Nearly all state-of-the-art language models and pre-trained language models are based on the Transformer architecture. However, it distinguishes sequential tokens only with the token position…

Large language models (LLMs) are increasingly deployed in enterprise settings where they handle sensitive documents and user context, raising acute concerns over security and controllability. Conventional access control regulates whether…

People are regularly confronted with potentially deceptive statements (e.g., fake news, misleading product reviews, or lies about activities). Only few works on automated text-based deception detection have exploited the potential of deep…

Feedback control actively dissipates uncertainty from a dynamical system by means of actuation. We develop a notion of "control capacity" that gives a fundamental limit (in bits) on the rate at which a controller can dissipate the…

Safety is a critical requirement for the real-world deployment of robotic systems. Unfortunately, while current robot foundation models show promising generalization capabilities across a wide variety of tasks, they fail to address safety,…

As assistive and collaborative robots become more ubiquitous in the real-world, we need to develop interfaces and controllers that are safe for users to build trust and encourage adoption. In this Blue Sky paper, we discuss the need for…

World modelling, i.e. building a representation of the rules that govern the world so as to predict its evolution, is an essential ability for any agent interacting with the physical world. Recent applications of the Transformer…

Until recently, the concept of soft robot safety was an informal notion, often attributed solely to the fact that soft robots are less likely to damage their operating environment than rigid robots. As the field moves toward feedback…

Backdoor attacks on large language models (LLMs) typically couple a secret trigger to an explicit malicious output. We show that this explicit association is unnecessary for common LLMs. We introduce a compliance-only backdoor: supervised…

When a human supervisor collaborates with a team of robots, their attention is divided and cognitive resources are at a premium. We aim to optimize the distribution of these resources and the flow of attention. To this end, we propose the…

A new framework is developed for control of constrained nonlinear systems with structured parametric uncertainties. Forward invariance of a safe set is achieved through online parameter adaptation and data-driven model estimation. The new…

Transformer-based models generate hidden states that are difficult to interpret. In this work, we analyze hidden states and modify them at inference, with a focus on motion forecasting. We use linear probing to analyze whether interpretable…

Robots must balance compliance with safety and social expectations as blind obedience can cause harm, while over-refusal erodes trust. Existing safe reinforcement learning (RL) benchmarks emphasize physical hazards, while human-robot…

Safety-oriented instruction-following is supposed to keep LLM-controlled robots safe. We show it also creates an availability attack surface. By injecting short safety-plausible phrases (1-5 tokens) into a robots audio channel, an adversary…

This paper addresses learning safe output feedback control laws from partial observations of expert demonstrations. We assume that a model of the system dynamics and a state estimator are available along with corresponding error bounds,…

In this paper, we propose a novel gender bias detection method by utilizing attention map for transformer-based models. We 1) give an intuitive gender bias judgement method by comparing the different relation degree between the genders and…

Recent advancements in large language models (LLMs) have accelerated progress toward artificial general intelligence, yet their potential to generate harmful content poses critical safety challenges. Existing alignment methods often…

Current language model safety paradigms often fall short in emotionally charged or high-stakes settings, where refusal-only approaches may alienate users and naive compliance can amplify risk. We propose ProSocialAlign, a test-time,…

Transformers have recently shown superior performances on various vision tasks. The large, sometimes even global, receptive field endows Transformer models with higher representation power over their CNN counterparts. Nevertheless, simply…

With the attention mechanism, transformers achieve significant empirical successes. Despite the intuitive understanding that transformers perform relational inference over long sequences to produce desirable representations, we lack a…