# Generative Conversational Networks

arXiv: https://arxiv.org/abs/2106.08484 · HTML: https://mlsanity.com/paper/2106.08484/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 32 references that are in the map.
- First posted on arXiv 15 Jun 2021.

## Builds on (most cited first, 26 of 26)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Fine-Tuning Language Models from Human Preferences](https://mlsanity.com/paper/1909.08593/) (2019)
- [Meta-Learning in Neural Networks: A Survey](https://mlsanity.com/paper/2004.05439/) (2020)
- [Texygen: A Benchmarking Platform for Text Generation Models](https://mlsanity.com/paper/1802.01886/) (2018)
- [Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfaces](https://mlsanity.com/paper/1805.10190/) (2018)
- [Transferable Multi-Domain State Generator for Task-Oriented Dialogue Systems](https://mlsanity.com/paper/1905.08743/) (2019)
- [Leveraging Pre-trained Checkpoints for Sequence Generation Tasks](https://mlsanity.com/paper/1907.12461/) (2019)
- [Cross-Lingual Transfer Learning for Multilingual Task Oriented Dialog](https://mlsanity.com/paper/1810.13327/) (2018)
- [Generative Teaching Networks: Accelerating Neural Architecture Search by Learning to Generate Synthetic Training Data](https://mlsanity.com/paper/1912.07768/) (2019)
- [Few-shot Natural Language Generation for Task-Oriented Dialog](https://mlsanity.com/paper/2002.12328/) (2020)
- [Few-shot Slot Tagging with Collapsed Dependency Transfer and Label-enhanced Task-adaptive Projection Network](https://mlsanity.com/paper/2006.05702/) (2020)
- [Zero-Shot Transfer Learning with Synthesized Data for Multi-Domain Dialogue State Tracking](https://mlsanity.com/paper/2005.00891/) (2020)
- [Neural Data Augmentation via Example Extrapolation](https://mlsanity.com/paper/2102.01335/) (2021)
- [Few-Shot NLG with Pre-Trained Language Model](https://mlsanity.com/paper/1904.09521/) (2019)
- [Span-ConveRT: Few-shot Span Extraction for Dialog with Pretrained Conversational Representations](https://mlsanity.com/paper/2005.08866/) (2020)
- [Robust Zero-Shot Cross-Domain Slot Filling with Example Values](https://mlsanity.com/paper/1906.06870/) (2019)
- [Coach: A Coarse-to-Fine Approach for Cross-domain Slot Filling](https://mlsanity.com/paper/2004.11727/) (2020)
- [Meta-Learning for Low-resource Natural Language Generation in Task-oriented Dialogue Systems](https://mlsanity.com/paper/1905.05644/) (2019)
- [Few Shot Dialogue State Tracking using Meta-learning](https://mlsanity.com/paper/2101.06779/) (2021)
- [C2C-GenDA: Cluster-to-Cluster Generation for Data Augmentation of Slot Filling](https://mlsanity.com/paper/2012.07004/) (2020)
- [Reward Shaping with Recurrent Neural Networks for Speeding up On-Line Policy Learning in Spoken Dialogue Systems](https://mlsanity.com/paper/1508.03391/) (2015)
- [Data Augmentation for Spoken Language Understanding via Pretrained Language Models](https://mlsanity.com/paper/2004.13952/) (2020)
- [ProtoDA: Efficient Transfer Learning for Few-Shot Intent Classification](https://mlsanity.com/paper/2101.11753/) (2021)
- [Template Guided Text Generation for Task-Oriented Dialogue](https://mlsanity.com/paper/2004.15006/) (2020)
- [Adversarial Domain Adaptation for Variational Neural Language Generation in Dialogue Systems](https://mlsanity.com/paper/1808.02586/) (2018)

## Built on by (most cited first, 11 of 11)

- [CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society](https://mlsanity.com/paper/2303.17760/) (2023)
- [State-of-the-art generalisation research in NLP: A taxonomy and review](https://mlsanity.com/paper/2210.03050/) (2022)
- [PLACES: Prompting Language Models for Social Conversation Synthesis](https://mlsanity.com/paper/2302.03269/) (2023)
- [Weakly Supervised Data Augmentation Through Prompting for Dialogue Understanding](https://mlsanity.com/paper/2210.14169/) (2022)
- [LINGUIST: Language Model Instruction Tuning to Generate Annotated Utterances for Intent Classification and Slot Tagging](https://mlsanity.com/paper/2209.09900/) (2022)
- [Selective In-Context Data Augmentation for Intent Detection using Pointwise V-Information](https://mlsanity.com/paper/2302.05096/) (2023)
- [NormDial: A Comparable Bilingual Synthetic Dialog Dataset for Modeling Social Norm Adherence and Violation](https://mlsanity.com/paper/2310.14563/) (2023)
- [Revisit Few-shot Intent Classification with PLMs: Direct Fine-tuning vs. Continual Pre-training](https://mlsanity.com/paper/2306.05278/) (2023)
- [Knowledge-Grounded Conversational Data Augmentation with Generative Conversational Networks](https://mlsanity.com/paper/2207.11363/) (2022)
- [Training Conversational Agents with Generative Conversational Networks](https://mlsanity.com/paper/2110.08383/) (2021)
- [A Longitudinal Study on Different Annotator Feedback Loops in Complex RAG Tasks](https://mlsanity.com/paper/2510.11897/) (2025)

## Abstract (arXiv)

Inspired by recent work in meta-learning and generative teaching networks, we propose a framework called Generative Conversational Networks, in which conversational agents learn to generate their own labelled training data (given some seed data) and then train themselves from that data to perform a given task. We use reinforcement learning to optimize the data generation process where the reward signal is the agent's performance on the task. The task can be any language-related task, from intent detection to full task-oriented conversations. In this work, we show that our approach is able to generalise from seed data and performs well in limited data and limited computation settings, with significant gains for intent detection and slot tagging across multiple datasets: ATIS, TOD, SNIPS, and Restaurants8k. We show an average improvement of 35% in intent detection and 21% in slot tagging over a baseline model trained from the seed data. We also conduct an analysis of the novelty of the generated data and provide generated examples for intent detection, slot tagging, and non-goal oriented conversations.