Unsolvable Problem Detection: Evaluating Trustworthiness of Vision
  Language Models

Atsuyuki Miyai; Jingkang Yang; Jingyang Zhang; Yifei Ming; Qing Yu; Go Irie; Yixuan Li; Hai Li; Ziwei Liu; Kiyoharu Aizawa

Paper · arXiv 2403.20331

Unsolvable Problem Detection: Evaluating Trustworthiness of Vision Language Models

Atsuyuki Miyai, Jingkang Yang, Jingyang Zhang, Yifei Ming, Qing Yu, Go Irie, Yixuan Li, Hai Li, Ziwei Liu, Kiyoharu Aizawa

15 upvotesMarch 29, 2024arXiv 预印本

AI 摘要

Unsolvable Problem Detection evaluates Vision Language Models' ability to identify unsolvable questions in Visual Question Answering through various settings, revealing areas for improvement and offering both training-free and training-based solutions.

Vision Language ModelsUnsolvable Problem DetectionVisual Question AnsweringAbsent Answer DetectionIncompatible Answer Set DetectionIncompatible Visual Question DetectionGPT-4VLLaVA-Next-34B

Abstract

This paper introduces a novel and significant challenge for Vision Language Models (VLMs), termed Unsolvable Problem Detection (UPD). UPD examines the VLM's ability to withhold answers when faced with unsolvable problems in the context of Visual Question Answering (VQA) tasks. UPD encompasses three distinct settings: Absent Answer Detection (AAD), Incompatible Answer Set Detection (IASD), and Incompatible Visual Question Detection (IVQD). To deeply investigate the UPD problem, extensive experiments indicate that most VLMs, including GPT-4V and LLaVA-Next-34B, struggle with our benchmarks to varying extents, highlighting significant room for the improvements. To address UPD, we explore both training-free and training-based solutions, offering new insights into their effectiveness and limitations. We hope our insights, together with future efforts within the proposed UPD settings, will enhance the broader understanding and development of more practical and reliable VLMs.