# Dialog2Flow: Pre-training Soft-Contrastive Action-Driven Sentence Embeddings for Automatic Dialog Flow Extraction

arXiv: https://arxiv.org/abs/2410.18481 · HTML: https://mlsanity.com/paper/2410.18481/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 50 of its 63 references that are in the map.
- First posted on arXiv 24 Oct 2024.

## Builds on (most cited first, 30 of 50)

- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Prototypical Networks for Few-shot Learning](https://mlsanity.com/paper/1703.05175/) (2017)
- [Sparks of Artificial General Intelligence: Early experiments with GPT-4](https://mlsanity.com/paper/2303.12712/) (2023)
- [SimCSE: Simple Contrastive Learning of Sentence Embeddings](https://mlsanity.com/paper/2104.08821/) (2021)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [Don't Stop Pretraining: Adapt Language Models to Domains and Tasks](https://mlsanity.com/paper/2004.10964/) (2020)
- [Contrastive Multiview Coding](https://mlsanity.com/paper/1906.05849/) (2019)
- [Learning deep representations by mutual information estimation and maximization](https://mlsanity.com/paper/1808.06670/) (2018)
- [BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Models](https://mlsanity.com/paper/2104.08663/) (2021)
- [MPNet: Masked and Permuted Pre-training for Language Understanding](https://mlsanity.com/paper/2004.09297/) (2020)
- [DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation](https://mlsanity.com/paper/1911.00536/) (2019)
- [How Contextual are Contextualized Word Representations? Comparing the Geometry of BERT, ELMo, and GPT-2 Embeddings](https://mlsanity.com/paper/1909.00512/) (2019)
- [MultiWOZ -- A Large-Scale Multi-Domain Wizard-of-Oz Dataset for Task-Oriented Dialogue Modelling](https://mlsanity.com/paper/1810.00278/) (2018)
- [Data-Efficient Image Recognition with Contrastive Predictive Coding](https://mlsanity.com/paper/1905.09272/) (2019)
- [Universal Sentence Encoder](https://mlsanity.com/paper/1803.11175/) (2018)
- [Large Dual Encoders Are Generalizable Retrievers](https://mlsanity.com/paper/2112.07899/) (2021)
- [Towards Scalable Multi-domain Conversational Agents: The Schema-Guided Dialogue Dataset](https://mlsanity.com/paper/1909.05855/) (2019)
- [Supervised Learning of Universal Sentence Representations from Natural Language Inference Data](https://mlsanity.com/paper/1705.02364/) (2017)
- [Text and Code Embeddings by Contrastive Pre-Training](https://mlsanity.com/paper/2201.10005/) (2022)
- [Skip-Thought Vectors](https://mlsanity.com/paper/1506.06726/) (2015)
- [MultiWOZ 2.2 : A Dialogue Dataset with Additional Annotation Corrections and State Tracking Baselines](https://mlsanity.com/paper/2007.12720/) (2020)
- [TOD-BERT: Pre-trained Natural Language Understanding for Task-Oriented Dialogue](https://mlsanity.com/paper/2004.06871/) (2020)
- [Neural Belief Tracker: Data-Driven Dialogue State Tracking](https://mlsanity.com/paper/1606.03777/) (2016)
- [Taskmaster-1: Toward a Realistic and Diverse Dialog Dataset](https://mlsanity.com/paper/1909.05358/) (2019)
- [Frames: A Corpus for Adding Memory to Goal-Oriented Dialogue Systems](https://mlsanity.com/paper/1704.00057/) (2017)

## Built on by (most cited first, 8 of 8)

- [Turning Conversations into Workflows: A Framework to Extract and Evaluate Dialog Workflows for Service AI Agents](https://mlsanity.com/paper/2502.17321/) (2025)
- [SDialog: A Python Toolkit for End-to-End Agent Building, User Simulation, Dialog Generation, and Evaluation](https://mlsanity.com/paper/2506.10622/) (2025)
- [METRO: Towards Strategy Induction from Expert Dialogue Transcripts for Non-collaborative Dialogues](https://mlsanity.com/paper/2604.11427/) (2026)
- [Text-to-SQL Task-oriented Dialogue Ontology Construction](https://mlsanity.com/paper/2507.23358/) (2025)
- [PrefixGuard: From LLM-Agent Traces to Online Failure-Warning Monitors](https://mlsanity.com/paper/2605.06455/) (2026)
- [Factors affecting the in-context learning abilities of LLMs for dialogue state tracking](https://mlsanity.com/paper/2506.08753/) (2025)
- [SDialog: A Python Toolkit for End-to-End Agent Building, User Simulation, Dialog Generation, and Evaluation](https://mlsanity.com/paper/2512.09142/) (2025)
- [Mining Workflow Graphs for Black-Box Boundary Testing of Conversational LLM Agents](https://mlsanity.com/paper/2607.06873/) (2026)

## Abstract (arXiv)

Efficiently deriving structured workflows from unannotated dialogs remains an underexplored and formidable challenge in computational linguistics. Automating this process could significantly accelerate the manual design of workflows in new domains and enable the grounding of large language models in domain-specific flowcharts, enhancing transparency and controllability. In this paper, we introduce Dialog2Flow (D2F) embeddings, which differ from conventional sentence embeddings by mapping utterances to a latent space where they are grouped according to their communicative and informative functions (i.e., the actions they represent). D2F allows for modeling dialogs as continuous trajectories in a latent space with distinct action-related regions. By clustering D2F embeddings, the latent space is quantized, and dialogs can be converted into sequences of region/action IDs, facilitating the extraction of the underlying workflow. To pre-train D2F, we build a comprehensive dataset by unifying twenty task-oriented dialog datasets with normalized per-turn action annotations. We also introduce a novel soft contrastive loss that leverages the semantic information of these actions to guide the representation learning process, showing superior performance compared to standard supervised contrastive loss. Evaluation against various sentence embeddings, including dialog-specific ones, demonstrates that D2F yields superior qualitative and quantitative results across diverse domains.