AFUN: Towards an Affordance Foundation Model for Functionality Understanding
Summary
AFUN proposes an affordance foundation model that predicts functional masks and 3D motion curves from RGB-D observations and language descriptions, enabling generalizable robot manipulation across diverse environments. The model outperforms baselines on multiple benchmarks and can be deployed for real-world tasks without fine-tuning.
View Cached Full Text
Cached at: 06/02/26, 07:33 PM
Paper page - AFUN: Towards an Affordance Foundation Model for Functionality Understanding
Source: https://huggingface.co/papers/2606.02551
Abstract
Affordance understanding model predicts functional masks and 3D motion curves from RGB-D observations and language descriptions, enabling generalizable robot manipulation across diverse environments.
Affordance understanding bridges visual perception and physical action, serving as an explainable interface for robot manipulation in open and unstructured real-world environments. Yet, building anaffordance foundation modelthat not only understands where and how the interaction should happen, but also generalizes across diverse environments, objects, and tasks, remains a long-standing research challenge. Existing methods typically address only part of this challenge, either localizing task-relevant regions without specifying executable motion, or predicting motion but with limited scalability. In this paper, we present ourmodel, a step towards anaffordance foundation modelfor functionality understanding. From a singleRGB-D observationand alanguage task description, ourmodel predicts atask-conditional functional mask(where to interact) and a3D post-contact motion curve(how to interact). To supportopen-world generalization, we build a large-scale standardized data pipeline that converts heterogeneous robot, human, simulation, and real-world scan data into a shared affordance schema with language, masks, and object-centric 3D motion labels. We evaluate ourmodel from three aspects: for affordance segmentation, ourmodel outperforms all baselines by a large margin across 8 test sets from 4 benchmarks, improving mean gIoU/cIoU by +23.9/+26.3; forcontact-point prediction, it predicts substantially more accurate points, with a 12.7--61.3% hit-rate gain over the best baseline; and for 3D motion, it achieves the best performance on all three test sets. ourmodel can be deployed for real-world robot manipulation without finetuning for robot embodiment or using task-specific heuristics, demonstrating the ability to adapt to open-world affordance tasks. Project page: https://www.zhaoningwang.com/AFUN
View arXiv pageView PDFProject pageGitHub0Add to collection
Get this paper in your agent:
hf papers read 2606\.02551
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2606.02551 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2606.02551 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2606.02551 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
What Objects Enable, Not What They Are: Functional Latent Spaces for Affordance Reasoning
This paper introduces A4D, a framework that maps visual observations into a shared latent space structured around affordances (e.g., 'movable') for robot planning. It achieves 94% inference accuracy on existing affordances, outperforming state-of-the-art by 15%, and enables 100x faster inference with superior generalization to unseen object functionalities.
AffordanceVLA: A Vision-Language-Action Model Empowering Action Generation through Affordance-Aware Understanding
AffordanceVLA introduces a unified framework using structured affordance forecasting as an intermediate representation to improve perception-action mapping in robotic manipulation, leveraging vision-language models and a Mixture-of-Transformer architecture.
AFFORDANCE20Q: Evaluating Affordance Reasoning from Physical Properties
Affordance20Q is a benchmark that evaluates LLMs' ability to reason about object affordances from physical properties without revealing object identity, using a 20-Questions format. Experiments show a ~20 point gap between LLMs and humans, and a proposed pipeline KARI improves open-source LLMs by up to 15.2 points.
ABot-N1: Toward a General Visual Language Navigation Foundation Model
ABot-N1 is a visual language navigation foundation model that decouples cognition from control using a slow-fast architecture with dual visual-language signals, achieving new state-of-the-art results across indoor and outdoor navigation tasks.
Foresight: Failure Detection for Long-Horizon Robotic Manipulation with Action-Conditioned World Model Latents
Foresight is a failure detection framework for long-horizon robotic manipulation that uses action-conditioned world model latents and functional conformal prediction to monitor trajectories, trained only with final task labels. It demonstrates state-of-the-art performance across simulation and real robot tasks.