Evaluating Contextual Intelligence in Recyclability: A Comprehensive Study of Image-Based Reasoning Systems

TL;DR

This study evaluates advanced vision-language models' ability to assess recyclability of items from images, considering context-specific factors, and highlights their progress and limitations in supporting sustainable recycling practices.

Contribution

It introduces a comprehensive evaluation of state-of-the-art vision-language models for recyclability prediction, emphasizing their contextual reasoning capabilities and identifying current shortcomings.

Findings

Models outperform previous versions in contextual understanding.

Performance varies across different recycling scenarios.

Identifies key areas for future model improvements.

Abstract

While the importance of efficient recycling is widely acknowledged, accurately determining the recyclability of items and their proper disposal remains a complex task for the general public. In this study, we explore the application of cutting-edge vision-language models (GPT-4o, GPT-4o-mini, and Claude 3.5) for predicting the recyclability of commonly disposed items. Utilizing a curated dataset of images, we evaluated the models' ability to match objects to appropriate recycling bins, including assessing whether the items could physically fit into the available bins. Additionally, we investigated the models' performance across several challenging scenarios: (i) adjusting predictions based on location-specific recycling guidelines; (ii) accounting for contamination or structural damage; and (iii) handling objects composed of multiple materials. Our findings highlight the significant advancements in contextual understanding offered by these models compared to previous iterations, while also identifying areas where they still fall short. The continued refinement of context-aware models is crucial for enhancing public recycling practices and advancing environmental sustainability.