---
format: "aidr-story-markdown/v1"
id: "4472cef530d68270a59fcc50b148c461c75e0b5561a3cc6828c6cadafc09a468"
canonical_url: "https://aidr.today/4472cef5?lang=en"
title: "RoTA-Bench: Evaluating 3D Spatial Reasoning through Interactive Projection Alignment"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-16T00:00:00.000Z"
category: "Research"
topics: ["reasoning"]
source_urls: ["https://huggingface.co/papers/2610.17665"]
summary: "RoTA-Bench evaluates 3D spatial reasoning in vision-language models by having them align rotated 3D objects with target silhouettes through successive rotations within a limited action budget."
---

# RoTA\-Bench: Evaluating 3D Spatial Reasoning through Interactive Projection Alignment

> [Open the canonical story](<https://aidr.today/4472cef5?lang=en>)

**Published:** 2026-10-16T00:00:00.000Z
**Category:** Research
**Topics:** reasoning

## Summary

RoTA\-Bench evaluates 3D spatial reasoning in vision\-language models by having them align rotated 3D objects with target silhouettes through successive rotations within a limited action budget\.

## Sources

- [Story source](<https://huggingface.co/papers/2610.17665>)