arXiv subjects

Emma Dixon

Publications and source records attributed to Emma Dixon.

5 recordsLinked to original sources

arXiv2025

A Careful Examination of Large Behavior Models for Multitask Dexterous Manipulation

Robot manipulation has seen tremendous progress in recent years, with imitation learning policies enabling successful performance of dexterous and hard-to-model tasks. Concurrently, scaling data and model size has led to the development of capable language and vision foundation models, motivating large-scale efforts to create general-purpose robot foundation models. While these models have garnered significant enthusiasm and investment, meaningful evaluation of real-world performance remains a challenge, limiting both the pace of development and inhibiting a nuanced understanding of current capabilities. In this paper, we rigorously evaluate multitask robot manipulation policies, referred to as Large Behavior Models (LBMs), by extending the Diffusion Policy paradigm across a corpus of simulated and real-world robot data. We propose and validate an evaluation pipeline to rigorously analyze the capabilities of these models with statistical confidence. We compare against single-task baselines through blind, randomized trials in a controlled setting, using both simulation and real-world experiments. We find that multi-task pretraining makes the policies more successful and robust, and enables teaching complex new tasks more quickly, using a fraction of the data when compared to single-task baselines. Moreover, performance predictably increases as pretraining scale and diversity grows. Project page: https://toyotaresearchinstitute.github.io/lbm1/

cs.RO↗

arXiv2025

Is Your Imitation Learning Policy Better than Mine? Policy Comparison with Near-Optimal Stopping

Imitation learning has enabled robots to perform complex, long-horizon tasks in challenging dexterous manipulation settings. As new methods are developed, they must be rigorously evaluated and compared against corresponding baselines through repeated evaluation trials. However, policy comparison is fundamentally constrained by a small feasible sample size (e.g., 10 or 50) due to significant human effort and limited inference throughput of policies. This paper proposes a novel statistical framework for rigorously comparing two policies in the small sample size regime. Prior work in statistical policy comparison relies on batch testing, which requires a fixed, pre-determined number of trials and lacks flexibility in adapting the sample size to the observed evaluation data. Furthermore, extending the test with additional trials risks inducing inadvertent p-hacking, undermining statistical assurances. In contrast, our proposed statistical test is sequential, allowing researchers to decide whether or not to run more trials based on intermediate results. This adaptively tailors the number of trials to the difficulty of the underlying comparison, saving significant time and effort without sacrificing probabilistic correctness. Extensive numerical simulation and real-world robot manipulation experiments show that our test achieves near-optimal stopping, letting researchers stop evaluation and make a decision in a near-minimal number of trials. Specifically, it reduces the number of evaluation trials by up to 32% as compared to state-of-the-art baselines, while preserving the probabilistic correctness and statistical power of the comparison. Moreover, our method is strongest in the most challenging comparison instances (requiring the most evaluation trials); in a multi-task comparison scenario, we save the evaluator more than 160 simulation rollouts.

cs.RO↗

arXiv2025

Can We Detect Failures Without Failure Data? Uncertainty-Aware Runtime Failure Detection for Imitation Learning Policies

Recent years have witnessed impressive robotic manipulation systems driven by advances in imitation learning and generative modeling, such as diffusion- and flow-based approaches. As robot policy performance increases, so does the complexity and time horizon of achievable tasks, inducing unexpected and diverse failure modes that are difficult to predict a priori. To enable trustworthy policy deployment in safety-critical human environments, reliable runtime failure detection becomes important during policy inference. However, most existing failure detection approaches rely on prior knowledge of failure modes and require failure data during training, which imposes a significant challenge in practicality and scalability. In response to these limitations, we present FAIL-Detect, a modular two-stage approach for failure detection in imitation learning-based robotic manipulation. To accurately identify failures from successful training data alone, we frame the problem as sequential out-of-distribution (OOD) detection. We first distill policy inputs and outputs into scalar signals that correlate with policy failures and capture epistemic uncertainty. FAIL-Detect then employs conformal prediction (CP) as a versatile framework for uncertainty quantification with statistical guarantees. Empirically, we thoroughly investigate both learned and post-hoc scalar signal candidates on diverse robotic manipulation tasks. Our experiments show learned signals to be mostly consistently effective, particularly when using our novel flow-based density estimator. Furthermore, our method detects failures more accurately and faster than state-of-the-art (SOTA) failure detection baselines. These results highlight the potential of FAIL-Detect to enhance the safety and reliability of imitation learning-based robotic systems as they progress toward real-world deployment.

cs.RO↗

arXiv2023

LLM-based Smart Reply (LSR): Enhancing Collaborative Performance with ChatGPT-mediated Smart Reply System

Interactive user interfaces have increasingly explored AI's role in enhancing communication efficiency and productivity in collaborative tasks. The emergence of Large Language Models (LLMs) such as ChatGPT has revolutionized conversational agents, employing advanced deep learning techniques to generate context-aware, coherent, and personalized responses. Consequently, LLM-based AI assistants provide a more natural and efficient user experience across various scenarios. In this paper, we study how LLM models can be used to improve work efficiency in collaborative workplaces. Specifically, we present an LLM-based Smart Reply (LSR) system utilizing the ChatGPT to generate personalized responses in professional collaborative scenarios while adapting to context and communication style based on prior responses. Our two-step process involves generating a preliminary response type (e.g., Agree, Disagree) to provide a generalized direction for message generation, thus reducing response drafting time. We conducted an experiment where participants completed simulated work tasks involving a Dual N-back test and subtask scheduling through Google Calendar while interacting with co-workers. Our findings indicate that the proposed LSR reduces overall workload, as measured by the NASA TLX, and improves work performance and productivity in the N-back task. We also provide qualitative analysis based on participants' experiences, as well as design considerations to provide future directions for improving such implementations.

cs.HC↗

arXiv2022

Investigating the Potential of Artificial Intelligence Powered Interfaces to Support Different Types of Memory for People with Dementia

There has been a growing interest in HCI to understand the specific technological needs of people with dementia and supporting them in self-managing daily activities. One of the most difficult challenges to address is supporting the fluctuating accessibility needs of people with dementia, which vary with the specific type of dementia and the progression of the condition. Researchers have identified auto-personalized interfaces, and more recently, Artificial Intelligence or AI-driven personalization as a potential solution to making commercial technology accessible in a scalable manner for users with fluctuating ability. However, there is a lack of understanding on the perceptions of people with dementia around AI as an aid to their everyday technology use and its role in their overall self-management systems, which include other non-AI technology, and human assistance. In this paper, we present future directions for the design of AI-based systems to personalize an interface for dementia-related changes in different types of memory, along with expectations for AI interactions with the user with dementia.

cs.HC↗