Task 1: Dialogue Response Generation - Elevating Conversational AI
This task is an invitation for you to advance conversational AI by crafting responses that transcend standard interactions, aiming to improve the response quality and user engagement.
Possess a Preliminary Chatbot Model? If your current model is adept at generating basic responses, you’re on the right trajectory. We invite you to refine it in accordance with our submission standards.
Is Your AI Capable of Persona-Grounded Dialogue? Our challenge calls for models that account for the dialogue history and personas. If your model is equipped for this, we encourage you to adapt it to our specified format and submit.
Aspiring to Integrate Commonsense Knowledge? We are looking for models that seamlessly integrate personal insights with a broader commonsense understanding. If your model is not there yet, consider our baseline model as a foundation. Your innovative solutions could pave the way for more dynamic and engaging AI conversations.
Embarking on Dialogue Generation? For those new to this domain, we suggest exploring our resource section or investigating foundational texts in open-domain dialogue generation. Start with a foundational model and progressively enhance its capabilities.
Task 2: Knowledge Linking - Advancing AI’s Conversational Contextualization
This task involves training AI to adeptly correlate dialogue with pertinent real-world knowledge, thereby enriching the informational and contextual quality of conversations.
Focusing on Binary Classification? If your model is proficient in discerning relevant from irrelevant information, this task is suitable for your research. Utilize our baseline model’s training dataset as a starting point.
Expertise in Knowledge Graphs? We encourage the development of models that can effectively associate knowledge segments with dialogue elements. We have prepared a set of potential knowledge-dialogue pairs for your AI to analyze and link appropriately.
Initiating Research in Knowledge Linking? For newcomers in this field, our “Baseline” section and the ComFact paper serve as comprehensive introductory resources. This task offers an excellent avenue to delve into the intricate interplay between AI and knowledge graphs.