Skip to content

Latest commit

 

History

History
85 lines (68 loc) · 5.56 KB

Non-task-Dialogue.md

File metadata and controls

85 lines (68 loc) · 5.56 KB
Raw

NRG (Neural Response Generation)

  • A Neural Conversational Mode ICLR15
    origin paper
  • Neural Responding Machine for Short-Text Conversation ACL15

Data

  • A Survey of Available Corpora for Building Data-Driven Dialogue Systems
  • Improved Deep Learning Baselines for Ubuntu Corpus Dialogs
  • LSDSCC: A Large Scale Domain-Specific Conversational Corpus for Response Generation with Diversity Oriented Evaluation Metrics NAACL18
  • Pchatbot: A Large-Scale Dataset for Personalized Chatbot paper

Metric

  • How NOT to evaluate your dialogue system: An empirical study of unsupervised evaluation metrics for dialogue response generation
  • Towards an automatic turing test: Learning to evaluate dialogue responses.
  • RUBER: An Unsupervised Method for Automatic Evaluation of Open-Domain Dialog Systems

Model

With knowledge

  • Emotional Chatting Machine: Emotional Conversation Generation with Internal and External Memory paper
  • Topic Aware Neural Response Generation AAAI16

Diversity

Inference

  • A Diversity-Promoting Objective Function for Neural Conversation Models NAACL16
  • Generating Long and Diverse Responses with Neural Conversation Models
  • A Simple, Fast Diverse Decoding Algorithm for Neural Generation

Model

  • Sequence to Backward and Forward Sequences: A Content-Introducing Approach to Generative Short-Text Conversation COLING16
  • Learning to Decode for Future Success
  • A Prospective-Performance Network to Alleviate Myopia in Beam Search for Response Generation COLING18
    1.predict larger beam search by a smaller one which promote the diversity a lot
  • Neural Response Generation with Dynamic Vocabularies AAAI18
  • Towards Less Generic Responses in Neural Conversation Models: A Statistical Re-weighting Method EMNLP18
    1.reweight by punish universal replies and short/long replies
    2.use tfidf similar responses to compute the coverage of a response (UR)

Latent

  • (VHRED) A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues
  • A Conditional Variational Framework for Dialog Generation ACL17
  • (CVAE) Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders ACL17 / code and data
  • Variational Autoregressive Decoder for Neural Response Generation EMNLP18
    1.increase the number of latent z from one to many;
    2.involving SBOW(predict future words in the sequence) auxiliary to training z.
  • Improving Variational Encoder-Decoders in Dialogue Generation AAAI18
    1.add a auto-encoder stagey (DAE) coo-work with CVAE stagey like GAN (to learn a good posterior), feed x->z' and sample z with using z' to stand x in a CVAE model;
    2.use a VAE/ecoder VAE/scheduled sampling to let the model consider more about the latent variable.
  • Topic-Guided Variational Autoencoders for Text Generation arxiv
  • A Semi-Supervised Stable Variational Network for Promoting Replier-Consistency in Dialogue Generation EMNLP19
    1.Von Mises-Fisher distribution
  • Dirichlet Latent Variable Hierarchical Recurrent Encoder-Decoder in Dialogue Generation EMNLP19
    1.Dirichlet

Analysis

  • Why are Sequence-to-Sequence Models So Dull? Understanding the Low-Diversity Problem of Chatbots EMNLP18 workshop

Beam Search Optimization

  • Sequence-to-Sequence Learning as Beam-Search Optimization
  • Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks

RL

  • A Deep Reinforcement Learning Chatbot paper
  • Deep Reinforcement Learning for Dialogue Generation ACL16

GAN

  • Neural Response Generation via GAN with an Approximate Embedding Layer EMNLP17
    Approximate (weighted averaged embedding by softmax) Embedding to solve the dispersion.

Persona

  • A Persona-Based Neural Conversation Model ACL16
  • Learning Personas from Dialogue with Attentive Memory Networks EMNLP18
  • Exploring Personalized Neural Conversational Models ICJAI17
  • Content-Oriented User Modeling for Personalized Response Ranking in Chatbots TASLP
  • Assigning Personality/Profile to a Chatting Machine for Coherent Conversation Generation IJCAI18
    1.involve profile into NRG (attention)
  • Steering Output Style and Topic in Neural Response Generation EMNLP17
  • Personalizing Dialogue Agents: I have a dog, do you have pets too? ACL18 / code & data

Retrieval

Analysis & Common