dean-sh / conv-ai-papers

Papers in the field of Open Domain, Personalized Conversational AI.

Open Domain, Personalized Conversational AI Papers

• This is repository contains papers for of Open Domain, Personalized Conversational AI.
• The papers tackle the challenges in building a long-term memory, personalized conversational AI that is consistent (with itself), knowledgeable, correct and engaging.

• Learning towards conversational AI: A survey, 2022. [paper]
• Retrieval Augmentation Reduces Hallucination in Conversation, 2021. [paper]
• A Survey on Response Selection for Retrieval-based Dialogues, 2021. [paper]
• What you need is retrieval - Current Limitations of Language Models, 2020. [paper]

• Augmenting Pre-trained Language Models with QA-Memory for Open-Domain Question Answering, 2022. [paper]
• Standing on the Shoulders of Giant Frozen Language Models, 2022. [paper]
• Memorizing Transformers, 2022. [paper][code]
• ∞-former - Infinite Memory Transformer, 2022. [paper] [code]
• Relational Memory Augmented Language Models, 2022. [paper]
• Dialogue History Matters! Personalized Response Selection in Multi-Turn Retrieval-Based Chatbots, 2021. [paper]
• Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering, 2021. [paper]
• RetGen - A Joint framework for Retrieval and Grounded Text Generation Modeling, 2021. [paper]
• Beyond Goldfish Memory - Long-Term Open-Domain Conversation, 2021. [paper]
• RETRO - Improving language models by retrieving from trillions of tokens, 2021. [paper][code]
• REALM - Retrieval-Augmented Language Model Pre-Training, 2020. [paper][HuggingFace]
• Multi-Attentive Transformer - A Response Retrieval Approach for Dialogue Using a Multi-Attentive Transformer, 2020. [paper]
• Hybrid Generative-Retrieval Transformers for Dialogue Domain Adaptation, 2020. [paper]
• Augmenting Transformers with KNN-Based Composite Memory for Dialogue, 2020. [paper]
• RAG - Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks, 2020. [paper][HuggingFace]
• TransformerXL - Attentive Language Models Beyond a Fixed-Length Context, 2019. [paper]

• Internet-augmented language models through few-shot prompting for open-domain question answering, 2022. [paper]
• SeeKeR - Language Models that Seek for Knowledge: Modular Search & Generation for Dialogue and Prompt Completion, 2022. [paper][code]
• LaMDA - Language Models for Dialog Applications, 2022. [paper]
• Internet-augmented Dialogue Generation, 2021. [paper]
• WebGPT - Browser-assisted question-answering with human feedback, 2021. [paper]

• Long Time No See! Open-Domain Conversation with Long-Term Persona Memory, 2022. [paper]
• EmpTransfo - A Multi-head Transformer Architecture for Creating Empathetic Dialog Systems, 2020. [paper] [code]
• Meena - Towards a Human-like Open-Domain Chatbot, 2020. [paper]

• BLOOM, 2022. [HuggingFace]
• OPT, 2022. [HuggingFace]
• GPT-Neo-X, 2022. [HuggingFace]
• DialoGPT, 2019. [HuggingFace]