karroyan commited on
Commit
023588a
·
1 Parent(s): 03e60fb

doc(lxy): modify readme

Browse files
Files changed (1) hide show
  1. README.md +17 -8
README.md CHANGED
@@ -1,4 +1,21 @@
1
  # HUMOR-RM (Keye-VL Version)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
 
3
  <div align="center">
4
 
@@ -12,14 +29,6 @@
12
 
13
  This specific version is fine-tuned on **Keye-VL**, utilizing a dataset of pairwise meme comparisons (ranked by human annotators). It takes two memes (sharing the same template) as input and predicts which one is funnier, providing a consistent proxy for human preference.
14
 
15
- ## Model Details
16
-
17
- * **Framework:** LLaMA-Factory
18
- * **Base Model:** [Keye-VL](https://huggingface.co/Kwai-Keye/Keye-VL-8B-Preview)
19
- * **Task:** Pairwise Classification / Reward Modeling
20
- * **Input:** Image Pair + Text Prompt
21
- * **Output:** Preference Score (Logits indicating )
22
-
23
  ## Requirements
24
 
25
  This model is built using the **LLaMA-Factory** framework structure. To run inference, you must have `llamafactory` installed.
 
1
  # HUMOR-RM (Keye-VL Version)
2
+ ---
3
+ language:
4
+ - en
5
+ - zh
6
+ license: apache-2.0
7
+ base_model: Kwai-Kolors/Keye-VL
8
+ tags:
9
+ - vision
10
+ - image-classification
11
+ - reward-model
12
+ - reinforcement-learning
13
+ - multimodal
14
+ - llama-factory
15
+ pipeline_tag: image-classification
16
+ library_name: transformers
17
+ ---
18
+
19
 
20
  <div align="center">
21
 
 
29
 
30
  This specific version is fine-tuned on **Keye-VL**, utilizing a dataset of pairwise meme comparisons (ranked by human annotators). It takes two memes (sharing the same template) as input and predicts which one is funnier, providing a consistent proxy for human preference.
31
 
 
 
 
 
 
 
 
 
32
  ## Requirements
33
 
34
  This model is built using the **LLaMA-Factory** framework structure. To run inference, you must have `llamafactory` installed.