# Beyond Mode Collapse: Generating Diverse Synthetic Expert Conversations via Generative Flow Networks

arXiv: https://arxiv.org/abs/2609.38359 · HTML: https://mlsanity.com/paper/2609.38359/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 26 references that are in the map.
- First posted on arXiv 29 Sep 2026.

## Builds on (most cited first, 18 of 18)

- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [LLMs Get Lost In Multi-Turn Conversation](https://mlsanity.com/paper/2505.06120/) (2025)
- [Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation](https://mlsanity.com/paper/2106.04399/) (2021)
- [Towards Emotional Support Dialog Systems](https://mlsanity.com/paper/2106.01144/) (2021)
- [Trajectory balance: Improved credit assignment in GFlowNets](https://mlsanity.com/paper/2201.13259/) (2022)
- [How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Models](https://mlsanity.com/paper/2102.08921/) (2021)
- [Turning Up the Heat: Min-p Sampling for Creative and Coherent LLM Outputs](https://mlsanity.com/paper/2407.01082/) (2024)
- [MathDial: A Dialogue Tutoring Dataset with Rich Pedagogical Properties Grounded in Math Reasoning Problems](https://mlsanity.com/paper/2305.14536/) (2023)
- [Synthetic Data for Deep Learning](https://mlsanity.com/paper/1909.11512/) (2019)
- [MT-Eval: A Multi-Turn Capabilities Evaluation Benchmark for Large Language Models](https://mlsanity.com/paper/2401.16745/) (2024)
- [Harnessing large-language models to generate private synthetic text](https://mlsanity.com/paper/2306.01684/) (2023)
- [Curated LLM: Synergy of LLMs and Data Curation for tabular augmentation in low-data regimes](https://mlsanity.com/paper/2312.12112/) (2023)
- [Training Proactive and Personalized LLM Agents](https://mlsanity.com/paper/2511.02208/) (2025)
- [Unmasking Implicit Bias: Evaluating Persona-Prompted LLM Responses in Power-Disparate Social Scenarios](https://mlsanity.com/paper/2503.01532/) (2025)
- [Detecting Mode Collapse in Language Models via Narration](https://mlsanity.com/paper/2402.04477/) (2024)
- [A Survey on Recent Advances in Conversational Data Generation](https://mlsanity.com/paper/2405.13003/) (2024)
- [SynthTextEval: Synthetic Text Data Generation and Evaluation for High-Stakes Domains](https://mlsanity.com/paper/2507.07229/) (2025)
- [Quality-Diversity Generative Sampling for Learning with Synthetic Data](https://mlsanity.com/paper/2312.14369/) (2023)

## Abstract (arXiv)

High quality synthetic data is central to post training LLMs for adaptive AI applications that represent the diverse expert strategies and decisions in conversations. Prompting LLMs directly or conditioning them on end use scenarios yields low diversity data that collapses onto dominant modes. We propose a method to generate diverse high quality synthetic data using Generative Flow Networks (GFlowNets). We show that training GFlowNets to generate latent conversation structure using a Gaussian mixture density over key interaction features (e.g., confusion episode dynamics, scaffolding directive balance) enables sampling expert strategies in proportion to their prevalence in the training data. Across two structurally distinct domains, tutoring and emotional support dialogues, our GFlow based synthetic data generation approach offers a better balance of fidelity, mode coverage and authenticity than reinforcement-learning and end to end LLM baselines, without copying training data. Evaluated on three downstream outcome prediction tasks, classifiers trained on synthetic GFlowNet generated conversations provide a stronger training signal than competitive synthesis baselines.