Update README.md
Browse files
README.md
CHANGED
|
@@ -25,6 +25,9 @@ This is the official huggingface repo for the paper "**Holistic Evaluation for I
|
|
| 25 |
- Run the inference on `interleaved_bench.json` with your model and get your model output (including text and image).
|
| 26 |
- Use the script in `src/interleavedeval_gpt4o.py` to perform evaluation.
|
| 27 |
|
|
|
|
|
|
|
|
|
|
| 28 |
|
| 29 |
One example in `interleaved_bench.json` is as follows:
|
| 30 |
|
|
|
|
| 25 |
- Run the inference on `interleaved_bench.json` with your model and get your model output (including text and image).
|
| 26 |
- Use the script in `src/interleavedeval_gpt4o.py` to perform evaluation.
|
| 27 |
|
| 28 |
+
### Important notes
|
| 29 |
+
- For image editing and subject-driven generation tasks, the scores on text-related aspects (text quality, text-image coherence) are directly set to 0. Please skip those scores when you compute the overall performance.
|
| 30 |
+
|
| 31 |
|
| 32 |
One example in `interleaved_bench.json` is as follows:
|
| 33 |
|