# Behavioral Foundation Models for Quality Diversity

arXiv: https://arxiv.org/abs/2609.35615 · HTML: https://mlsanity.com/paper/2609.35615/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 40 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## Builds on (most cited first, 27 of 27)

- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [Exploration by Random Network Distillation](https://mlsanity.com/paper/1810.12894/) (2018)
- [Illuminating search spaces by mapping elites](https://mlsanity.com/paper/1504.04909/) (2015)
- [The CMA Evolution Strategy: A Tutorial](https://mlsanity.com/paper/1604.00772/) (2016)
- [Successor Features for Transfer in Reinforcement Learning](https://mlsanity.com/paper/1606.05312/) (2016)
- [The Role of Permutation Invariance in Linear Mode Connectivity of Neural Networks](https://mlsanity.com/paper/2110.06296/) (2021)
- [A Survey on Vision-Language-Action Models for Embodied AI](https://mlsanity.com/paper/2405.14093/) (2024)
- [OGBench: Benchmarking Offline Goal-Conditioned RL](https://mlsanity.com/paper/2410.20092/) (2024)
- [URLB: Unsupervised Reinforcement Learning Benchmark](https://mlsanity.com/paper/2110.15191/) (2021)
- [Learning One Representation to Optimize All Rewards](https://mlsanity.com/paper/2103.07945/) (2021)
- [Universal Successor Features Approximators](https://mlsanity.com/paper/1812.07626/) (2018)
- [The Laplacian in RL: Learning Representations with Efficient Approximations](https://mlsanity.com/paper/1810.04586/) (2018)
- [Does Zero-Shot Reinforcement Learning Exist?](https://mlsanity.com/paper/2209.14935/) (2022)
- [Covariance Matrix Adaptation for the Rapid Illumination of Behavior Space](https://mlsanity.com/paper/1912.02400/) (2019)
- [Learning Successor States and Goal-Dependent Values: A Mathematical Viewpoint](https://mlsanity.com/paper/2101.07123/) (2021)
- [Zero-Shot Whole-Body Humanoid Control via Behavioral Foundation Models](https://mlsanity.com/paper/2504.11054/) (2025)
- [Illuminating Mario Scenes in the Latent Space of a Generative Adversarial Network](https://mlsanity.com/paper/2007.05674/) (2020)
- [TD-JEPA: Latent-predictive Representations for Zero-Shot Reinforcement Learning](https://mlsanity.com/paper/2510.00739/) (2025)
- [Fast Adaptation with Behavioral Foundation Models](https://mlsanity.com/paper/2504.07896/) (2025)
- [Policy Manifold Search: Exploring the Manifold Hypothesis for Diversity-based Neuroevolution](https://mlsanity.com/paper/2104.13424/) (2021)
- [MAP-Elites with Descriptor-Conditioned Gradients and Archive Distillation into a Single Policy](https://mlsanity.com/paper/2303.03832/) (2023)
- [QDax: A Library for Quality-Diversity and Population-based Algorithms with Hardware Acceleration](https://mlsanity.com/paper/2308.03665/) (2023)
- [Self-Predictive Representations for Combinatorial Generalization in Behavioral Cloning](https://mlsanity.com/paper/2506.10137/) (2025)
- [Understanding the Synergies between Quality-Diversity and Deep Reinforcement Learning](https://mlsanity.com/paper/2303.06164/) (2023)
- [Synergizing Quality-Diversity with Descriptor-Conditioned Reinforcement Learning](https://mlsanity.com/paper/2401.08632/) (2023)
- [Improving Zero-Shot Offline RL via Behavioral Task Sampling](https://mlsanity.com/paper/2604.25496/) (2026)

## Abstract (arXiv)

Behavioral Foundation Models (BFMs) are an emerging paradigm in reinforcement learning, playing a role analogous to large language models in natural language processing: they have shown remarkable versatility, enabling zero-shot performance, fast imitation, and online adaptation, all by exploiting the structure of a latent space. In this work, we investigate whether the latent behavioral space induced by BFMs can serve as an effective search space to discover large repertoires of behaviorally diverse and high-performing policies through Quality-Diversity (QD) methods. While QD methods generally search directly in high-dimensional policy parameter space, in this paper, we present BFM-QD, a framework that performs QD search in the compact latent space of a BFM. We further show that the BFM-QD framework provides a closed-form, gradient-free policy improvement operator that approximates a policy gradient update, but requires no critic training and no backpropagation. Across continuous-control benchmarks spanning dense locomotion, sparse navigation, and contact-rich manipulation, BFM-QD consistently outperforms parameter-space baselines, with particularly stark gains in sparse and deceptive settings, where all tested parameter-space QD methods collapse to near-zero performance. These results show the effectiveness of the BFM-QD framework, benefiting from the synergy between dimensionality reduction of the search space and offline pretraining from diverse behavioral data. This positions BFMs as a general-purpose backbone for QD optimization, extending their utility beyond zero-shot task solving to the discovery of diverse behavioral repertoires.