Academic Poster Content Evaluation on 30 academic posters 1.0 (test)
3.97LayeringPosterGen
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| PosterGenVLM Judge=Claude Sonnet 42025.08 | 3.97 | 3.43 | 4.27 | 4.8 | 4.6 | 4.87 | 4.53 | 4.35 | |
| Human-designedVLM Judge=Claude Sonnet 42025.08 | 3.73 | 3.33 | 4.13 | 4.47 | 4.4 | 4.27 | 3.73 | 4.01 | |
| Human-designedVLM Judge=GPT-4.12025.08 | 3.47 | 3.83 | 3.97 | 4.5 | 4.53 | 4.47 | 4.47 | 4.18 | |
| PosterGenVLM Judge=GPT-4.12025.08 | 3.27 | 3.93 | 3.83 | 4.7 | 4.87 | 4.9 | 4.83 | 4.33 | |
| PosterAgentVLM Judge=Claude Sonnet 42025.08 | 3.03 | 3.07 | 3.5 | 4.8 | 4.23 | 4.3 | 3.9 | 3.83 | |
| PosterAgentVLM Judge=GPT-4.12025.08 | 3 | 3.33 | 3.33 | 4.8 | 4.07 | 4.17 | 4.07 | 3.82 | |
| P2PVLM Judge=Claude Sonnet 42025.08 | 2.9 | 3.13 | 3.5 | 4.73 | 4.27 | 4.27 | 3.77 | 3.8 | |
| P2PVLM Judge=GPT-4.12025.08 | 2.63 | 3.33 | 3.17 | 3.83 | 4.67 | 4.57 | 4.53 | 3.82 | |
| GPT-4o-ImageVLM Judge=GPT-4.12025.08 | 2.23 | 2.1 | 2.2 | 2.2 | 2.73 | 2.87 | 2.57 | 2.41 | |
| GPT-4o-ImageVLM Judge=Claude Sonnet 42025.08 | 2.23 | 2.03 | 2.87 | 2.1 | 2.83 | 2.6 | 2.13 | 2.4 |