Anh (Totti) Nguyen
@anh-ng8.bsky.social
In search of an intelligent and explainable AI. Machine Learning, Human-Computer Interaction, and Javascript. Associate Professor at Auburn U.
🌐 https://anhnguyen.me/
🐦 https://x.com/anh_ng8
🌐 https://anhnguyen.me/
🐦 https://x.com/anh_ng8
Pinned
🧵 Vision Language Models are ⚠️ biased
Q: Count the legs of this animal?
🤖: 4 ❌
Same problem:
- w/ 5 best VLMs: GPT-4.1, o3, o4-mini, Gemini 2.5 Pro, Sonnet 3.7
- on 7 domains: animals, logos, flags, chess, boardgames, optical illusions, patterned grids
code, paper, data: vlmsarebiased.github.io
Q: Count the legs of this animal?
🤖: 4 ❌
Same problem:
- w/ 5 best VLMs: GPT-4.1, o3, o4-mini, Gemini 2.5 Pro, Sonnet 3.7
- on 7 domains: animals, logos, flags, chess, boardgames, optical illusions, patterned grids
code, paper, data: vlmsarebiased.github.io
🧵 Vision Language Models are ⚠️ biased
Q: Count the legs of this animal?
🤖: 4 ❌
Same problem:
- w/ 5 best VLMs: GPT-4.1, o3, o4-mini, Gemini 2.5 Pro, Sonnet 3.7
- on 7 domains: animals, logos, flags, chess, boardgames, optical illusions, patterned grids
code, paper, data: vlmsarebiased.github.io
Q: Count the legs of this animal?
🤖: 4 ❌
Same problem:
- w/ 5 best VLMs: GPT-4.1, o3, o4-mini, Gemini 2.5 Pro, Sonnet 3.7
- on 7 domains: animals, logos, flags, chess, boardgames, optical illusions, patterned grids
code, paper, data: vlmsarebiased.github.io
June 5, 2025 at 7:28 PM
🧵 Vision Language Models are ⚠️ biased
Q: Count the legs of this animal?
🤖: 4 ❌
Same problem:
- w/ 5 best VLMs: GPT-4.1, o3, o4-mini, Gemini 2.5 Pro, Sonnet 3.7
- on 7 domains: animals, logos, flags, chess, boardgames, optical illusions, patterned grids
code, paper, data: vlmsarebiased.github.io
Q: Count the legs of this animal?
🤖: 4 ❌
Same problem:
- w/ 5 best VLMs: GPT-4.1, o3, o4-mini, Gemini 2.5 Pro, Sonnet 3.7
- on 7 domains: animals, logos, flags, chess, boardgames, optical illusions, patterned grids
code, paper, data: vlmsarebiased.github.io