Search papers, labs, and topics across Lattice.
This paper introduces the Blind-Spots-Bench, a new benchmark designed to identify persistent blind spots in multimodal AI models through tasks that are trivial for humans but challenging for AI. By analyzing 235 carefully curated tasks, the authors reveal that closed-source models significantly outperform open-weight models by approximately 10% on these tasks, despite similar performance on traditional benchmarks. The findings underscore the necessity of using targeted assessments like Blind-Spots-Bench to diagnose specific weaknesses in AI systems that are not captured by existing evaluation methods.
Closed-source AI models can outperform open-weight counterparts by 10% on seemingly simple tasks, revealing hidden vulnerabilities in multimodal systems.
Modern AI models achieve strong performance on many established benchmarks, yet they still fail on tasks that humans find almost trivial, such as manipulating a string or drawing a dog with five legs. These examples suggest that existing benchmarks may under-measure persistent blind spots in current systems. We introduce $\texttt{blind-spots-bench}$, a benchmark designed to expose such blind spots through tasks that appear simple for humans but remain challenging for modern AI. We collect raw questions from students in an AI course, clean and annotate them with structured reference solutions, and propose a task taxonomy tailored to the resulting dataset of 235 samples. We further develop an automated grading pipeline to evaluate a wide range of models, including open-weight and closed-source language, vision-language, and image-generation models. Our analysis on $\texttt{blind-spots-bench}$ reveals that closed-source frontier models can substantially outperform open-weight models with even $\approx10\%$ gap, even when they attain comparable performance on existing benchmarks. A more fine-grained analysis shows that no single model dominates across all task types, and that some tasks remain challenging for all evaluated models. These results highlight the value of $\texttt{blind-spots-bench}$ as a diagnostic stress test for identifying concrete weaknesses in current modern models.
