Hugging Face's logo

MILVLG
/
imp-v1-3b

Text Generation
Transformers
Safetensors
imp
custom_code
Model card Files Community
11
MILVLG commited on
Commit
a6da6b8
•
1 Parent(s): 5843de5

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -51,7 +51,7 @@ print(tokenizer.decode(output_ids[input_ids.shape[1]:], skip_special_tokens=True
51
  ## Model evaluation
52
  We conduct evaluation on 9 commonly-used benchmarks, including 5 academic VQA benchmarks and 4 popular MLLM benchmarks, to compare our Imp model with LLaVA (7B) and existing MSLMs with similar sizes.
53
 
54
- | Models | Size | VQAv2 | GQA |VisWiz | SQA (IMG) | TextVQA | POPE | MME | MMB |MM-Vet|
55
  |:--------:|:-----:|:----:|:----:|:-------------:|:--------:|:-----:|:----:|:-------:|:-------:|:-------:|
56
  | [LLaVA-v1.5-lora](https://huggingface.co/liuhaotian/llava-v1.5-7b) | 7B |79.10 | **63.00** |47.80 | 68.40 |58.20| 86.40 | **1476.9** | 66.10 |30.2|
57
  | [TinyGPT-V](https://huggingface.co/Tyrannosaurus/TinyGPT-V) | 3B | - | 33.60 | 24.80 | - | - | -| - | - |-|
 
51
  ## Model evaluation
52
  We conduct evaluation on 9 commonly-used benchmarks, including 5 academic VQA benchmarks and 4 popular MLLM benchmarks, to compare our Imp model with LLaVA (7B) and existing MSLMs with similar sizes.
53
 
54
+ | Models | Size | VQAv2 | GQA |VizWiz | SQA (IMG) | TextVQA | POPE | MME | MMB |MM-Vet|
55
  |:--------:|:-----:|:----:|:----:|:-------------:|:--------:|:-----:|:----:|:-------:|:-------:|:-------:|
56
  | [LLaVA-v1.5-lora](https://huggingface.co/liuhaotian/llava-v1.5-7b) | 7B |79.10 | **63.00** |47.80 | 68.40 |58.20| 86.40 | **1476.9** | 66.10 |30.2|
57
  | [TinyGPT-V](https://huggingface.co/Tyrannosaurus/TinyGPT-V) | 3B | - | 33.60 | 24.80 | - | - | -| - | - |-|