# GFlowRL: Scaling Distribution-Matching RL to Large Language Models

arXiv: https://arxiv.org/abs/2607.13394 · HTML: https://mlsanity.com/paper/2607.13394/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 46 of its 56 references that are in the map.
- First posted on arXiv 15 Jul 2026.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)

## Builds on (most cited first, 30 of 46)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [Solving Quantitative Reasoning Problems with Language Models](https://mlsanity.com/paper/2206.14858/) (2022)
- [Kimi k1.5: Scaling Reinforcement Learning with LLMs](https://mlsanity.com/paper/2501.12599/) (2025)
- [OlympiadBench: A Challenging Benchmark for Promoting AGI with Olympiad-Level Bilingual Multimodal Scientific Problems](https://mlsanity.com/paper/2402.14008/) (2024)
- [Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation](https://mlsanity.com/paper/2106.04399/) (2021)
- [Great, Now Write an Article About That: The Crescendo Multi-Turn LLM Jailbreak Attack](https://mlsanity.com/paper/2404.01833/) (2024)
- [GFlowNet Foundations](https://mlsanity.com/paper/2111.09266/) (2021)
- [Trajectory balance: Improved credit assignment in GFlowNets](https://mlsanity.com/paper/2201.13259/) (2022)
- [REINFORCE++: Stabilizing Critic-Free Policy Optimization with Global Advantage Normalization](https://mlsanity.com/paper/2501.03262/) (2025)
- [Learning GFlowNets from partial episodes for improved convergence and stability](https://mlsanity.com/paper/2209.12782/) (2022)
- [X-Teaming: Multi-Turn Jailbreaks and Defenses with Adaptive Multi-Agents](https://mlsanity.com/paper/2504.13203/) (2025)
- [Generative Flow Networks for Discrete Probabilistic Modeling](https://mlsanity.com/paper/2202.01361/) (2022)
- [Backpropagation through the Void: Optimizing control variates for black-box gradient estimation](https://mlsanity.com/paper/1711.00123/) (2017)
- [GFlowNets and variational inference](https://mlsanity.com/paper/2210.00580/) (2022)
- [Amortizing intractable inference in large language models](https://mlsanity.com/paper/2310.04363/) (2023)
- [LLMs know their vulnerabilities: Uncover Safety Gaps through Natural Distribution Shifts](https://mlsanity.com/paper/2410.10700/) (2024)
- [The Optimal Reward Baseline for Gradient-Based Reinforcement Learning](https://mlsanity.com/paper/1301.2315/) (2013)
- [Solving high-dimensional Hamilton-Jacobi-Bellman PDEs using neural networks: perspectives from the theory of controlled diffusions and measures on path space](https://mlsanity.com/paper/2005.05409/) (2020)
- [FlipAttack: Jailbreak LLMs via Flipping](https://mlsanity.com/paper/2410.02832/) (2024)
- [REBAR: Low-variance, unbiased gradient estimates for discrete latent variable models](https://mlsanity.com/paper/1703.07370/) (2017)
- [Chain of Attack: a Semantic-Driven Contextual Multi-Turn attacker for LLM](https://mlsanity.com/paper/2405.05610/) (2024)

## Built on by (most cited first, 2 of 2)

- [EvoSteer: Online Self-Evolving Graph Orchestration via Reference-Anchored Credit Assignment](https://mlsanity.com/paper/2609.38661/) (2026)
- [RL Post-Training Diffusion Language Models Along Their Own Generation Paths](https://mlsanity.com/paper/2610.17266/) (2026)

## Abstract (arXiv)

Generative Flow Networks (GFlowNets) offer a promising alternative to reward-maximizing reinforcement learning (RL) for large reasoning models, encouraging diverse reasoning paths by matching reward distributions rather than collapsing to dominant modes. Recent work shows promise on math and code, but scaling GFlowNet-style RL to modern post-training pipelines remains difficult: as model size, rollout horizon, reward noise, and distributed-systems complexity grow together, a learned prompt-conditional partition function becomes a source of gradient instability and engineering overhead rather than a useful normalizer. Through systematic analysis, we find that the learned partition function, previously treated as essential, can be replaced by an in-batch Monte Carlo estimate computed from the rollout group already required for training. We propose GFlowRL, a streamlined GFlowNet-style RL algorithm that removes the auxiliary partition network entirely while preserving the reward-distribution-matching objective, completed by two stabilizers: importance-sampling correction for rollout/trainer drift and asymmetric flow-gap clipping for outlier residuals. GFlowRL exceeds all counterparts on math, code, and adversarial red-teaming benchmarks, reaching a Codeforces rating of 2048 at the 14B scale (within 25 Elo of o3-mini) and attaining the highest average ASR@1 on AdvBench and HarmBench, outperforming the previous SOTA multi-turn attacker in a regime where FlowRL, a prior GFlowNet-style method, diverges. The same recipe transfers to all evaluated MoE configurations up to 235B parameters, where FlowRL again fails to converge. To our knowledge, GFlowRL is the first GFlowNet-style RL algorithm to scale stably across both dense and sparse architectures. Code will be at: https://github.com/microsoft/gflowrl