Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2410.19546 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908446925783040 |
|---|---|
| author | Wüst, Antonia Woydt, Tim Helff, Lukas Ibs, Inga Stammer, Wolfgang Dhami, Devendra S. Rothkopf, Constantin A. Kersting, Kristian |
| author_facet | Wüst, Antonia Woydt, Tim Helff, Lukas Ibs, Inga Stammer, Wolfgang Dhami, Devendra S. Rothkopf, Constantin A. Kersting, Kristian |
| contents | Recently, newly developed Vision-Language Models (VLMs), such as OpenAI's o1, have emerged, seemingly demonstrating advanced reasoning capabilities across text and image modalities. However, the depth of these advances in language-guided perception and abstract reasoning remains underexplored, and it is unclear whether these models can truly live up to their ambitious promises. To assess the progress and identify shortcomings, we enter the wonderland of Bongard problems, a set of classic visual reasoning puzzles that require human-like abilities of pattern recognition and abstract reasoning. With our extensive evaluation setup, we show that while VLMs occasionally succeed in identifying discriminative concepts and solving some of the problems, they frequently falter. Surprisingly, even elementary concepts that may seem trivial to humans, such as simple spirals, pose significant challenges. Moreover, when explicitly asked to recognize ground truth concepts, they continue to falter, suggesting not only a lack of understanding of these elementary visual concepts but also an inability to generalize to unseen concepts. We compare the results of VLMs to human performance and observe that a significant gap remains between human visual reasoning capabilities and machine cognition. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_19546 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Bongard in Wonderland: Visual Puzzles that Still Make AI Go Mad? Wüst, Antonia Woydt, Tim Helff, Lukas Ibs, Inga Stammer, Wolfgang Dhami, Devendra S. Rothkopf, Constantin A. Kersting, Kristian Artificial Intelligence Machine Learning Recently, newly developed Vision-Language Models (VLMs), such as OpenAI's o1, have emerged, seemingly demonstrating advanced reasoning capabilities across text and image modalities. However, the depth of these advances in language-guided perception and abstract reasoning remains underexplored, and it is unclear whether these models can truly live up to their ambitious promises. To assess the progress and identify shortcomings, we enter the wonderland of Bongard problems, a set of classic visual reasoning puzzles that require human-like abilities of pattern recognition and abstract reasoning. With our extensive evaluation setup, we show that while VLMs occasionally succeed in identifying discriminative concepts and solving some of the problems, they frequently falter. Surprisingly, even elementary concepts that may seem trivial to humans, such as simple spirals, pose significant challenges. Moreover, when explicitly asked to recognize ground truth concepts, they continue to falter, suggesting not only a lack of understanding of these elementary visual concepts but also an inability to generalize to unseen concepts. We compare the results of VLMs to human performance and observe that a significant gap remains between human visual reasoning capabilities and machine cognition. |
| title | Bongard in Wonderland: Visual Puzzles that Still Make AI Go Mad? |
| topic | Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2410.19546 |