Header
ELOQUENCE
Multilingual and Cross-cultural interactions for context-aware, and bias-controlled dialogue systems for safety-critical applications
ELOQUENCE is focused on the research and development of innovative technologies for collaborative voice/chat bots. Voice assistant-powered
dialogue engines have previously been deployed in a number of commercial and governmental technological pipelines, with a diverse level of
complexity. In our concept, such a complexity can be understood as a problem of analysing unstructured dialogues. ELOQUENCE’s key objective is
to better comprehend those unstructured dialogues and translate them into explainable, safe, knowledge-grounded, trustworthy and bias-controlled
language models. We envision to develop a technology capable of learning by its own, by adapting from a very data-limited corpora to efficiently
support most of the EU languages; from a sustainable computational framework to efficient and green-power architectures and, in essence, that may
serve as a guidance for all European citizens whilst being respectful and showing the best of our European values, specifically supporting safety-
critical applications by involving humans-in-the-loop.
Overall, ELOQUENCE’s project considers building on top and to improve of prior achievements in the domain of conversational agents, e.g. recently
launched and public-domain Large Language Models (LLMs), such as chatGPT (e.g., more recent versions), or LaMDa most of them developed in
non-EU countries. While including key industrial enterprises from Europe (i.e., Omilia, Telefonica, Synelixis), ELOQUENCE will validate the
developed technology through (i) safety-critical scenarios with human-in-the-loop for security-critical applications (i.e., emergency services in call
centres) and (ii) smart home assistants via information retrieval and fact-checking against an online knowledge base for lesser risky autonomous
systems (i.e., home-assistants). ELOQUENCE will target the R&D of these novel conversational AI technologies in multilingual and multimodal
environments and demonstrated in several pilots.
Business Categories
Duration
-
ELOQUENCE is dedicated to advancing collaborative voice/chat bots through innovative technology research and development. We aim to understand and transform unstructured dialogues into explainable, safe, knowledge-grounded, trustworthy, and bias-controlled language models. Our technology goals include self-learning, adaptability across languages and use-cases, sustainability with new computational frameworks, and serving as a guide for European citizens, particularly in safety-critical applications.
Our project leverages prior successes in conversational agents, including non-EU-developed Large Language Models (LLMs). We collaborate with European enterprises and validate our technology in safety-critical scenarios like emergency services as well as smart home assistants for less risky autonomous systems.
ELOQUENCE focuses on context-aware conversational AI in diverse settings, with an interdisciplinary team including experts in various fields. Our project emphasizes sustainability, open science, and addressing ethical and societal concerns, including bias mitigation.
O1: Advanced SLU technologies for safety-critical applications
O2: Hybrid LLMs combining contextual knowledge toward explainable and decision making in complex semi-structured and unstructured dialogues
O3: Enhanced conversational agents toward un-biased and trustworthy dialogues with end-users
O4: Definition of a framework to assess the conversational AI methods in various scenarios
O5: Definition and execution of pilots to validate ELOQUENCE technologies
O6: ValidatIion ELOQUENCE technologies against legal, ethical and societal requirements, demonstrate their compatibility with European values, awareness of achievements
ELOQUENCE is committed to achieving greater inclusiveness in technology development while aligning with European values and sustainability. This commitment is realized through an agile software development approach that facilitates continuous dialogue and assessment between technology developers and pilot study leaders. The project comprises five work packages, each addressing specific aspects of technology development, including multilingual conversational AI systems, scalable decision-making agents, resource-efficient models for low-resourced languages, and ethical and legal considerations.
ELOQUENCE's key objectives include reducing bias in deployed models by 50% to align with European AI regulations. The project also plans to showcase its virtual agent's capabilities in call center safety-critical scenarios and smart home environments through pilot validations. These results will be shared with industrial partners and stakeholders to promote responsible AI development and deployment.
ELOQUENCE is committed to promoting sustainable and high-quality job creation by addressing skills gaps and empowering workers, including those at risk of social exclusion. The project recognizes the ethical considerations associated with technological progress and its impact on job markets. To achieve this, specific work packages (WP3 and WP5) focus on developing AI-based solutions for complex problem-solving in both high-risk and low-risk scenarios. These solutions are expected to lead to more sustainable and high-quality jobs, particularly in industries requiring complex decision-making.
Key results include reducing the quantity of training data and energy consumption by 50%, which will make AI technologies more accessible and environmentally friendly. Additionally, ELOQUENCE aims to achieve an 80% correlation between automatic responses generated by AI models and those from human agents, enhancing reliability and trust in AI-powered technologies. This will facilitate their acceptance by end-users and increase their adoption across various industries.