analiza aiqbench
This paper addresses critical reproducibility challenges in multimodal affordance prediction by introducing the "Affordance Sheet," a standardized documentation framework for experimental protocols and model architectures. By enforcing consistent reporting across heterogeneous formulations, it enables fair benchmarking and reliable evaluation of AI agents operating in real-world scenarios with novel conditions.
Reproducible Multimodal Affordance Prediction
arXiv:2608.18317v1 Announce Type: cross Abstract: Affordance prediction is the identification of potential actions an agent can perform on a target object from multimodal inputs. Affordance prediction methods are difficult to evaluate and compare due to heterogeneous problem formulations, inconsistent dataset annotations, incomplete reporting of experimental protocols, and limited information about deployment conditions. These limitations challenge fair benchmarking and performance comparison. To promote transparency, we propose the Affordance Sheet, a documentation detailing task formulation with its input modalities, model architectures and training information, datasets, and experimental protocols. Affordance Sheets enable reproducible benchmarking and reliable evaluation of affordance models for real-world scenarios, including generalisation to novel conditions and human safety.
Abstract:Affordance prediction is the identification of potential actions an agent can perform on a target object from multimodal inputs. Affordance prediction methods are difficult to evaluate and compare due to heterogeneous problem formulations, inconsistent dataset annotations, incomplete reporting of experimental protocols, and limited information about deployment conditions. These limitations challenge fair benchmarking and performance comparison. To promote transparency, we propose the Affordance Sheet, a documentation detailing task formulation with its input modalities, model architectures and training information, datasets, and experimental protocols. Affordance Sheets enable reproducible benchmarking and reliable evaluation of affordance models for real-world scenarios, including generalisation to novel conditions and human safety.
Paper accepted to Workshop on Human-Centered Multimodal Intelligence in the Wild (HCMIW) in European Conference on Computer Vision (ECCV) 2026; 18 pages, 3 figures, 7 tables. Project webpage at this https URL
Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
arXiv:2608.18317 [cs.CV]
(or arXiv:2608.18317v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2608.18317 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 158 cuvinte
Subiecte conexe
Penn State scientists to lead three Genesis Mission projects
psu.edu · 24 iul. 2026
The Machines that Make the Machines
NVIDIA · 7 oct. 2026
When to Align, When to Predict: A Phase Diagram for Multimodal Learning
arXiv · 10 iun. 2026
Human-Guided Agentic AI for Multimodal Clinical Prediction: Lessons from the AgentDS Healthcare Benchmark
arXiv · 12 iun. 2026