Butter-Bench: Evaluating LLM Controlled Robots for Practical Intelligence
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911230992580608 |
|---|---|
| author | Sharrock, Callum Petersson, Lukas Petersson, Hanna Backlund, Axel Wennström, Axel Nordström, Kristoffer Aronsson, Elias |
| author_facet | Sharrock, Callum Petersson, Lukas Petersson, Hanna Backlund, Axel Wennström, Axel Nordström, Kristoffer Aronsson, Elias |
| contents | We present Butter-Bench, a benchmark evaluating large language model (LLM) controlled robots for practical intelligence, defined as the ability to navigate the messiness of the physical world. Current state-of-the-art robotic systems use a hierarchical architecture with LLMs in charge of high-level reasoning, and a Vision Language Action (VLA) model for low-level control. Butter-Bench evaluates the LLM part in isolation from the VLA. Although LLMs have repeatedly surpassed humans in evaluations requiring analytical intelligence, we find humans still outperform LLMs on Butter-Bench. The best LLMs score 40% on Butter-Bench, while the mean human score is 95%. LLMs struggled the most with multi-step spatial planning and social understanding. We also evaluate LLMs that are fine-tuned for embodied reasoning and conclude that this training does not improve their score on Butter-Bench. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_21860 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Butter-Bench: Evaluating LLM Controlled Robots for Practical Intelligence Sharrock, Callum Petersson, Lukas Petersson, Hanna Backlund, Axel Wennström, Axel Nordström, Kristoffer Aronsson, Elias Robotics Artificial Intelligence We present Butter-Bench, a benchmark evaluating large language model (LLM) controlled robots for practical intelligence, defined as the ability to navigate the messiness of the physical world. Current state-of-the-art robotic systems use a hierarchical architecture with LLMs in charge of high-level reasoning, and a Vision Language Action (VLA) model for low-level control. Butter-Bench evaluates the LLM part in isolation from the VLA. Although LLMs have repeatedly surpassed humans in evaluations requiring analytical intelligence, we find humans still outperform LLMs on Butter-Bench. The best LLMs score 40% on Butter-Bench, while the mean human score is 95%. LLMs struggled the most with multi-step spatial planning and social understanding. We also evaluate LLMs that are fine-tuned for embodied reasoning and conclude that this training does not improve their score on Butter-Bench. |
| title | Butter-Bench: Evaluating LLM Controlled Robots for Practical Intelligence |
| topic | Robotics Artificial Intelligence |
| url | https://arxiv.org/abs/2510.21860 |