# MINERVA: How Small Can a Manipulation Policy Be and Still Solve LIBERO?

arXiv: https://arxiv.org/abs/2609.03715 · HTML: https://mlsanity.com/paper/2609.03715/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Failure recovery and robustness, one of 13 directions in Vision-language-action models.
- Builds on 27 of its 29 references that are in the map.
- First posted on arXiv 3 Sep 2026.

## Placed in

- Vision-language-action models / [Failure recovery and robustness](https://mlsanity.com/vla/failure-recovery-and-robustness/)

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

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [Octo: An Open-Source Generalist Robot Policy](https://mlsanity.com/paper/2405.12213/) (2024)
- [MLP-Mixer: An all-MLP Architecture for Vision](https://mlsanity.com/paper/2105.01601/) (2021)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success](https://mlsanity.com/paper/2502.19645/) (2025)
- [End-to-End Training of Deep Visuomotor Policies](https://mlsanity.com/paper/1504.00702/) (2015)
- [FAST: Efficient Action Tokenization for Vision-Language-Action Models](https://mlsanity.com/paper/2501.09747/) (2025)
- [SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics](https://mlsanity.com/paper/2506.01844/) (2025)
- [TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulation](https://mlsanity.com/paper/2409.12514/) (2024)
- [LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Models](https://mlsanity.com/paper/2510.13626/) (2025)
- [Real-Time Execution of Action Chunking Flow Policies](https://mlsanity.com/paper/2506.07339/) (2025)
- [LIBERO-PRO: Towards Robust and Fair Evaluation of Vision-Language-Action Models Beyond Memorization](https://mlsanity.com/paper/2510.03827/) (2025)
- [Deployment-Efficient Reinforcement Learning via Model-Based Offline Optimization](https://mlsanity.com/paper/2006.03647/) (2020)
- [DeeR-VLA: Dynamic Inference of Multimodal Large Language Models for Efficient Robot Execution](https://mlsanity.com/paper/2411.02359/) (2024)
- [Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Sampling](https://mlsanity.com/paper/2408.17355/) (2024)
- [Policy Information Capacity: Information-Theoretic Measure for Task Complexity in Deep Reinforcement Learning](https://mlsanity.com/paper/2103.12726/) (2021)
- [DiT-Air: Revisiting the Efficiency of Diffusion Model Architecture Design in Text to Image Generation](https://mlsanity.com/paper/2503.10618/) (2025)
- [TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM](https://mlsanity.com/paper/2607.27205/) (2026)

## Built on by (most cited first, 3 of 3)

- [Beyond Single-Axis Testing: Paired Evaluation of Compound Robustness in Vision-Language-Action Policies](https://mlsanity.com/paper/2609.15940/) (2026)
- [Decoupling Vision, Language, and Action for Efficient Multi-Task Robot Policies](https://mlsanity.com/paper/2609.18374/) (2026)
- [Screw Attention: Rigid-Body Algebra Inside a Transformer](https://mlsanity.com/paper/2610.00904/) (2026)

## Abstract (arXiv)

Vision-language-action (VLA) models with billions of parameters now dominate the LIBERO manipulation benchmark, but the model capacity actually required by the benchmark remains unclear. We introduce MINERVA (MINimal Efficient Robotic Vision-Action policy), a family of deliberately compact visuomotor policies designed to measure this task-specific capacity floor. A 0.54M-parameter policy achieves 95.1% average success over 2,000 rollouts on the four standard LIBERO suites, only 2.4 points below the reported LeRobot π0.5 result despite using 7,700× fewer parameters. Performance saturates near 1M parameters and collapses below 0.25M. Across broad architectural, training, and inference sweeps, only action-chunk length and vision capacity consistently exceed a ±1-point training-seed band. Flow matching provides no detectable advantage over direct L1 regression across three seeds, while regression is up to 3.8× faster on GPU. A task-ID permutation probe shows that standard LIBERO instruction conditioning primarily selects among memorized tasks: changing only the task-ID mapping reduces success to near chance. The same recipe achieves 94.6% success across 89 LIBERO-90 tasks, while LIBERO-Plus perturbations reduce performance to 46--56%, with near-zero robustness to photometric shifts. The 0.54M policy replans every control step in 5--9 ms per chunk on a laptop CPU, 113× faster than SmolVLA and 1,400× faster than π0.5, without a GPU. These results establish a first empirical estimate of LIBERO's task-specific capacity floor and motivate capacity-aware design and distillation for deployment-efficient robot policies.