Upload turing-motors/Heron-NVILA-Lite-2B
Browse files
README.md
CHANGED
@@ -118,7 +118,7 @@ I used [llm-jp-eval-mm](https://github.com/llm-jp/llm-jp-eval-mm) for this evalu
|
|
118 |
| Model | LLM Size | Heron-Bench overall LLM (%) | JA-VLM-Bench-In-the-Wild LLM (/5.0) | JA-VG-VQA-500 LLM (/5.0) |
|
119 |
|--------------------------------|----------|------------------------------|-------------------------------------|--------------------------|
|
120 |
| **Heron NVILA-Lite 2B** | 1.5B | 52.8 | 3.52 | 3.50 |
|
121 |
-
| **Heron NVILA-Lite 15B** | 14B | 59.6 | 4.2 | 3.82 |
|
122 |
| [LLaVA-CALM2-SigLIP](https://huggingface.co/cyberagent/llava-calm2-siglip) | 7B | 43.3 | 3.15 | 3.21 |
|
123 |
| [Llama-3-EvoVLM-JP-v2](https://huggingface.co/SakanaAI/Llama-3-EvoVLM-JP-v2) | 8B | 39.3 | 2.92 | 2.96 |
|
124 |
| [VILA-jp](https://huggingface.co/llm-jp/llm-jp-3-vila-14b) | 13B | 57.2 | 3.69 | 3.62 |
|
@@ -135,16 +135,9 @@ This model is experimental and has not been thoroughly calibrated for ethical co
|
|
135 |
- Model weights are licensed under [Apache License 2.0](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct/blob/main/LICENSE).
|
136 |
- Users must comply with [OpenAI terms of use](https://openai.com/policies/terms-of-use) due to the inclusion of GPT-4-generated synthetic data.
|
137 |
|
138 |
-
##
|
139 |
|
140 |
-
|
141 |
-
@misc{HeronNVILALite2B,
|
142 |
-
title = {Heron NVILA-Lite 2B},
|
143 |
-
author = {Shingo Yokoi},
|
144 |
-
year = {2025},
|
145 |
-
url = {https://huggingface.co/turing-motors/Heron-NVILA-Lite-2B},
|
146 |
-
}
|
147 |
-
```
|
148 |
|
149 |
## Citations
|
150 |
|
@@ -158,4 +151,8 @@ This model is experimental and has not been thoroughly calibrated for ethical co
|
|
158 |
primaryClass={cs.CV},
|
159 |
url={https://arxiv.org/abs/2412.04468},
|
160 |
}
|
161 |
-
```
|
|
|
|
|
|
|
|
|
|
118 |
| Model | LLM Size | Heron-Bench overall LLM (%) | JA-VLM-Bench-In-the-Wild LLM (/5.0) | JA-VG-VQA-500 LLM (/5.0) |
|
119 |
|--------------------------------|----------|------------------------------|-------------------------------------|--------------------------|
|
120 |
| **Heron NVILA-Lite 2B** | 1.5B | 52.8 | 3.52 | 3.50 |
|
121 |
+
| **[Heron NVILA-Lite 15B](https://huggingface.co/turing-motors/Heron-NVILA-Lite-15B)** | 14B | 59.6 | 4.2 | 3.82 |
|
122 |
| [LLaVA-CALM2-SigLIP](https://huggingface.co/cyberagent/llava-calm2-siglip) | 7B | 43.3 | 3.15 | 3.21 |
|
123 |
| [Llama-3-EvoVLM-JP-v2](https://huggingface.co/SakanaAI/Llama-3-EvoVLM-JP-v2) | 8B | 39.3 | 2.92 | 2.96 |
|
124 |
| [VILA-jp](https://huggingface.co/llm-jp/llm-jp-3-vila-14b) | 13B | 57.2 | 3.69 | 3.62 |
|
|
|
135 |
- Model weights are licensed under [Apache License 2.0](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct/blob/main/LICENSE).
|
136 |
- Users must comply with [OpenAI terms of use](https://openai.com/policies/terms-of-use) due to the inclusion of GPT-4-generated synthetic data.
|
137 |
|
138 |
+
## Acknowledgements
|
139 |
|
140 |
+
This model is based on the results obtained in the project, subsidized by the [Generative AI Accelerator Challenge (GENIAC)](https://www.meti.go.jp/policy/mono_info_service/geniac/index.html).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
141 |
|
142 |
## Citations
|
143 |
|
|
|
151 |
primaryClass={cs.CV},
|
152 |
url={https://arxiv.org/abs/2412.04468},
|
153 |
}
|
154 |
+
```
|
155 |
+
|
156 |
+
## Model Card Authors
|
157 |
+
|
158 |
+
Shingo Yokoi
|