A Conversational User Interface for Instructional Maintenance Reports

Freire, Samuel Kernan; Niforatos, Evangelos; Rusák, Zoltán; Aschenbrenner, Doris; Bozzon, Alessandro

doi:10.1145/3543829.3544516

Cited by 8 publications

(4 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They noted that using voice interaction instead of a busy chat modality may allow people to take a step back and refect on their work. Other work has shown that reporting by voice to an IA results in reports with more information, including valuable explanations, than writing on paper [19]. These studies demonstrate that conversational interaction is an exciting technique for supporting self-refection in the workplace.…”

Section: Introductionmentioning

confidence: 55%

Tacit Knowledge Elicitation for Shop-floor Workers with an Intelligent Assistant

Freire

Wang

Ruiz-Arenas

et al. 2023

Extended Abstracts of the 2023 CHI Conference on Human Factors in Computing Systems

Self Cite

View full text Add to dashboard Cite

Many industries face the challenge of capturing workers' knowledge to share it, particularly tacit knowledge. The operation of complex systems such as a manufacturing line is knowledge-intensive. Considering this knowledge's breadth and dynamic nature, existing knowledge-sharing solutions are inefcient and resource intensive. Conversational user interfaces are an efcient way to convey information that mimics how humans share knowledge; however, we know little about how to design them specifcally for knowledge sharing, especially regarding tacit knowledge. In this work, we present an intelligent assistant that we have developed to support the elicitation of tacit knowledge from workers through systematic refection. The system can interact with workers by voice or text and generate visualizations of shop foor data to support refective prompts. CCS CONCEPTS• Human-centered computing → Empirical studies in HCI; Interactive systems and tools; Natural language interfaces.

show abstract

Section: Introductionmentioning

confidence: 55%

Tacit Knowledge Elicitation for Shop-floor Workers with an Intelligent Assistant

Freire

Wang

Ruiz-Arenas

et al. 2023

Extended Abstracts of the 2023 CHI Conference on Human Factors in Computing Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…Efforts were also made to integrate multimodal inputs, enabling LLMs to process and respond to a combination of text, image, and voice data [21][22][23]. Another significant development was the customization of LLMs for specific industries, such as legal or medical, tailoring their outputs to suit domain-specific needs [24,25]. Lastly, the integration of real-time data streams opened new avenues for LLMs, allowing them to provide responses that were not only contextually relevant but also current and timely [7,26].…”

Section: Advancements In Large Language Modelsmentioning

confidence: 99%

Probabilistic Inference Layer Integration in Mistral LLM for Accurate Information Retrieval

Wang,

Ouyang

2024

Preprint

View full text Add to dashboard Cite

This study introduces a novel integration of a Probabilistic Inference Layer (PIL) into the Mistral Large Language Model (LLM), aiming to address the critical challenge of accurate and reliable information retrieval in natural language processing. By employing advanced statistical models within the PIL, the enhanced Mistral LLM demonstrates a marked improvement in information retrieval accuracy, context understanding, and bias reduction. The PIL's application of Bayesian networks and sophisticated mathematical constructs, such as matrix calculus and principles akin to Bernoulli's and Lorentz transformations, enables the Mistral LLM to process information with a higher degree of accuracy and reliability. The study's results indicate significant advancements in the model's performance across various tests, particularly in discerning context and intent, reducing biases, and handling complex logical operations. Despite its computational demands and the ongoing challenge to completely eliminate biases, the PIL integration establishes a new benchmark for LLMs and opens avenues for future research. This study contributes to the field by demonstrating the potential of probabilistic methods in enhancing the capabilities of generative AI models, thus paving the way for more sophisticated and reliable AI-driven information systems.

show abstract

“…However, we believe that it is crucial for cognitive assistants' long-term success on the shop floor. Technological probes have shown that supporting workers to think aloud can result in the creation of high-quality reports from which knowledge can be extracted [17]. Fenoglio et al [11] suggest a system for capturing explicit and tacit knowledge (through best practices) from experienced industrial workers, employing a role-playing game where a virtual agent interacts with human experts and knowledge engineers to iteratively extract and represent knowledge.…”

Section: Leveraging Cognitive Assistants For Knowledge Sharing In Man...mentioning

confidence: 99%

Harnessing Large Language Models for Cognitive Assistants in Factories

Freire

Foosherian²,

Wang

et al. 2023

Proceedings of the 5th International Conference on Conversational User Interfaces

Self Cite

View full text Add to dashboard Cite

As agile manufacturing expands and workforce mobility increases, the importance of efficient knowledge transfer among factory workers grows. Cognitive Assistants (CAs) with Large Language Models (LLMs), like GPT-3.5, can bridge knowledge gaps and improve worker performance in manufacturing settings. This study investigates the opportunities, risks, and user acceptance of LLM-powered CAs in two factory contexts: textile and detergent production. Several opportunities and risks are identified through a literature review, proof-of-concept implementation, and focus group sessions. Factory representatives raise concerns regarding data security, privacy, and the reliability of LLMs in high-stake environments. By following design guidelines regarding persistent memory, realtime data integration, security, privacy, and ethical concerns, LLMpowered CAs can become valuable assets in manufacturing settings and other industries. CCS CONCEPTS• Human-centered computing → Human computer interaction (HCI); Natural language interfaces; • Computing methodologies → Natural language generation; • Information systems → Users and interactive retrieval; • Applied computing → Industry and manufacturing.

show abstract

A Conversational User Interface for Instructional Maintenance Reports

Cited by 8 publications

References 19 publications

Tacit Knowledge Elicitation for Shop-floor Workers with an Intelligent Assistant

Tacit Knowledge Elicitation for Shop-floor Workers with an Intelligent Assistant

Probabilistic Inference Layer Integration in Mistral LLM for Accurate Information Retrieval

Harnessing Large Language Models for Cognitive Assistants in Factories

Contact Info

Product

Resources

About