PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
July 28, 20241 citationsOpen Access

Visual Riddles: a Commonsense and World Knowledge Challenge for Large Vision and Language Models

View Full Paper
NBNitzan Bitton-GuettaASAviv SlobodkinAMAviya Maimon

Key Points

Key points are not available for this paper at this time.

Abstract

Imagine observing someone scratching their arm; to understand why, additional context would be necessary. However, spotting a mosquito nearby would immediately offer a likely explanation for the person's discomfort, thereby alleviating the need for further information. This example illustrates how subtle visual cues can challenge our cognitive skills and demonstrates the complexity of interpreting visual scenarios. To study these skills, we present Visual Riddles, a benchmark aimed to test vision and language models on visual riddles requiring commonsense and world knowledge. The benchmark comprises 400 visual riddles, each featuring a unique image created by a variety of text-to-image models, question, ground-truth answer, textual hint, and attribution. Human evaluation reveals that existing models lag significantly behind human performance, which is at 82\% accuracy, with Gemini-Pro-1.5 leading with 40\% accuracy. Our benchmark comes with automatic evaluation tasks to make assessment scalable. These findings underscore the potential of Visual Riddles as a valuable resource for enhancing vision and language models' capabilities in interpreting complex visual scenarios.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Bitton-Guetta et al. (2024) studied this question.

synapsesocial.com/papers/68e5ec41b6db643587581010https://doi.org/10.48550/arxiv.2407.19474
Ask AI
Helpful
Bookmark
Share
View Full Paper

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models2024 · 6 citations
  2. 2Are Language Models Puzzle Prodigies? Algorithmic Puzzles Unveil Serious Challenges in Multimodal Reasoning2024 · 3 citations
  3. 3VisualOverload: Probing Visual Understanding of VLMs in Really Dense Scenes2025
  4. 4PuzzleVQA: Diagnosing Multimodal Reasoning Challenges of Language Models with Abstract Visual Patterns2024 · 1 citations
  5. 5WinoViz: Probing Visual Properties of Objects Under Different States2024