Add pipeline tag, library name, paper, and code links to model card
Browse filesThis PR enhances the model card for Omni-Reward by:
* Adding `library_name: transformers` to correctly enable the "how to use" widget, as the model's file structure (e.g., `config.json`, `tokenizer_config.json`) indicates compatibility with the Hugging Face Transformers library.
* Setting `pipeline_tag: any-to-any` to accurately reflect its omni-modal capabilities (handling text, image, video, audio, and 3D data) and improve discoverability on the Hub.
* Including a direct link to the paper ([Omni-Reward: Towards Generalist Omni-Modal Reward Modeling with Free-Form Preferences](https://huggingface.co/papers/2510.23451)) and the official GitHub repository (`https://github.com/HongbangYuan/OmniReward`) in the introductory link section.
* Correcting a minor `</a></a>` typo in the existing benchmark link.
These updates improve the model's discoverability and provide users with comprehensive information and usage guidance.
|
@@ -1,18 +1,20 @@
|
|
| 1 |
---
|
| 2 |
-
license: cc-by-nc-4.0
|
| 3 |
-
datasets:
|
| 4 |
-
- jinzhuoran/OmniRewardData
|
| 5 |
base_model:
|
| 6 |
- openbmb/MiniCPM-o-2_6
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
---
|
| 8 |
|
| 9 |
-
|
| 10 |
-
|
| 11 |
# Omni-Reward: Towards Generalist Omni-Modal Reward Modeling with Free-Form Preferences
|
| 12 |
|
| 13 |
|
| 14 |
<p align="center">
|
| 15 |
-
<a href="https://huggingface.co/
|
|
|
|
|
|
|
| 16 |
<a href="https://hf.co/datasets/jinzhuoran/OmniRewardData"> π€ Dataset</a> |
|
| 17 |
<a href="https://hf.co/jinzhuoran/OmniRewardModel"> π€ Model</a> |
|
| 18 |
<a href="https://omnireward.github.io/"> π Homepage</a>
|
|
@@ -102,5 +104,4 @@ bash scripts/eval_ti2t_tie.sh
|
|
| 102 |
|
| 103 |
- `--eval_dataset`: Specifies the evaluation dataset (e.g., `omni_t2t`, `omni_t2i`, `omni_t2v`, etc.).
|
| 104 |
|
| 105 |
-
- `--eval_tie`: Enables w/ Ties evaluation.
|
| 106 |
-
|
|
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
| 2 |
base_model:
|
| 3 |
- openbmb/MiniCPM-o-2_6
|
| 4 |
+
datasets:
|
| 5 |
+
- jinzhuoran/OmniRewardData
|
| 6 |
+
license: cc-by-nc-4.0
|
| 7 |
+
library_name: transformers
|
| 8 |
+
pipeline_tag: any-to-any
|
| 9 |
---
|
| 10 |
|
|
|
|
|
|
|
| 11 |
# Omni-Reward: Towards Generalist Omni-Modal Reward Modeling with Free-Form Preferences
|
| 12 |
|
| 13 |
|
| 14 |
<p align="center">
|
| 15 |
+
<a href="https://huggingface.co/papers/2510.23451"> π Paper</a> |
|
| 16 |
+
<a href="https://github.com/HongbangYuan/OmniReward"> π» Code</a> |
|
| 17 |
+
<a href="https://huggingface.co/datasets/HongbangYuan/OmniRewardBench"> π€ Benchmark</a> |
|
| 18 |
<a href="https://hf.co/datasets/jinzhuoran/OmniRewardData"> π€ Dataset</a> |
|
| 19 |
<a href="https://hf.co/jinzhuoran/OmniRewardModel"> π€ Model</a> |
|
| 20 |
<a href="https://omnireward.github.io/"> π Homepage</a>
|
|
|
|
| 104 |
|
| 105 |
- `--eval_dataset`: Specifies the evaluation dataset (e.g., `omni_t2t`, `omni_t2i`, `omni_t2v`, etc.).
|
| 106 |
|
| 107 |
+
- `--eval_tie`: Enables w/ Ties evaluation.
|
|
|