Update README.md
Browse files
README.md
CHANGED
@@ -27,7 +27,7 @@ tags:
|
|
27 |
|
28 |
<p align="center">
|
29 |
<br>
|
30 |
-
<img src="assets/GoLLIE.png" style="height: 250px;">
|
31 |
<br>
|
32 |
<h2 align="center"><b>G</b>uideline f<b>o</b>llowing <b>L</b>arge <b>L</b>anguage Model for <b>I</b>nformation <b>E</b>xtraction</h2>
|
33 |
|
@@ -36,17 +36,17 @@ tags:
|
|
36 |
|
37 |
|
38 |
<p align="justify">
|
39 |
-
We present
|
40 |
|
41 |
- ๐ป Code: [https://github.com/osainz59/CoLLIE/](https://github.com/osainz59/CoLLIE/)
|
42 |
- ๐ Blog Post: [GoLLIE: Guideline-following Large Language Model for Information Extraction](docs/index.md)
|
43 |
- ๐ Paper: [GoLLIE: Annotation Guidelines improve Zero-Shot Information-Extraction]()
|
44 |
-
-
|
45 |
- ๐ Example Jupyter Notebooks: [GoLLIE Notebooks](notebooks/)
|
46 |
</p>
|
47 |
|
48 |
<p align="center">
|
49 |
-
<img src="assets/zero_shot_results.png">
|
50 |
</p>
|
51 |
|
52 |
|
@@ -112,7 +112,7 @@ result = [
|
|
112 |
|
113 |
## How to Get Started with the Model
|
114 |
|
115 |
-
Please read our [๐ Example Jupyter Notebooks](notebooks
|
116 |
|
117 |
The best way to load the model is using our custom `load_model` fuction. However, you can also load them using the AutoModelForCausalLM class.
|
118 |
|
@@ -135,17 +135,17 @@ model = AutoModelForCausalLM.from_pretrained("HiTZ/GoLLIE-7B", trust_remote_code
|
|
135 |
model.to("cuda")
|
136 |
```
|
137 |
|
138 |
-
Read our [๐ Example Jupyter Notebooks](notebooks
|
139 |
|
140 |
|
141 |
|
142 |
### Training Data
|
143 |
|
144 |
-
This is the list of task used for training and evaluating GoLLIE. However, as demonstrated in the ๐ [Create Custom Task notebook](notebooks/Create%20Custom%20Task.ipynb) GoLLIE can perform a wide range of unseen tasks.
|
145 |
For more info, read our [๐Paper]().
|
146 |
|
147 |
<p align="center">
|
148 |
-
<img src="assets/datasets.png">
|
149 |
</p>
|
150 |
|
151 |
|
@@ -162,9 +162,9 @@ For more info, read our [๐Paper]().
|
|
162 |
|
163 |
| Model | Hardware | FLOPs | Time (h) | CO<sup>2</sup>eq (kg) |
|
164 |
|----------------|-------------------|---------------------------|-------------------|-------------------------------------|
|
165 |
-
|
|
166 |
-
|
|
167 |
-
|
|
168 |
|
169 |
|
170 |
|
|
|
27 |
|
28 |
<p align="center">
|
29 |
<br>
|
30 |
+
<img src="https://github.com/hitz-zentroa/GoLLIE/raw/main/assets/GoLLIE.png" style="height: 250px;">
|
31 |
<br>
|
32 |
<h2 align="center"><b>G</b>uideline f<b>o</b>llowing <b>L</b>arge <b>L</b>anguage Model for <b>I</b>nformation <b>E</b>xtraction</h2>
|
33 |
|
|
|
36 |
|
37 |
|
38 |
<p align="justify">
|
39 |
+
We present GoLLIE, a Large Language Model trained to follow annotation guidelines. GoLLIE outperforms previous approaches on zero-shot Information Extraction and allows the user to perform inferences with annotation schemas defined on the fly. Different from previous approaches, GoLLIE is able to follow detailed definitions and does not only rely on the knowledge already encoded in the LLM. Code and models are publicly available.
|
40 |
|
41 |
- ๐ป Code: [https://github.com/osainz59/CoLLIE/](https://github.com/osainz59/CoLLIE/)
|
42 |
- ๐ Blog Post: [GoLLIE: Guideline-following Large Language Model for Information Extraction](docs/index.md)
|
43 |
- ๐ Paper: [GoLLIE: Annotation Guidelines improve Zero-Shot Information-Extraction]()
|
44 |
+
- GoLLIE Colection in the ๐คHuggingFace Hub: [HiTZ/gollie](https://huggingface.co/collections/HiTZ/gollie-651bf19ee315e8a224aacc4f)
|
45 |
- ๐ Example Jupyter Notebooks: [GoLLIE Notebooks](notebooks/)
|
46 |
</p>
|
47 |
|
48 |
<p align="center">
|
49 |
+
<img src="https://github.com/hitz-zentroa/GoLLIE/raw/main/assets/zero_shot_results.png">
|
50 |
</p>
|
51 |
|
52 |
|
|
|
112 |
|
113 |
## How to Get Started with the Model
|
114 |
|
115 |
+
Please read our [๐ Example Jupyter Notebooks](https://github.com/hitz-zentroa/GoLLIE/tree/main/notebooks) to get started with GoLLIE.
|
116 |
|
117 |
The best way to load the model is using our custom `load_model` fuction. However, you can also load them using the AutoModelForCausalLM class.
|
118 |
|
|
|
135 |
model.to("cuda")
|
136 |
```
|
137 |
|
138 |
+
Read our [๐ Example Jupyter Notebooks](https://github.com/hitz-zentroa/GoLLIE/tree/main/notebooks) to learn how to easily define guidelines, generate model inputs and parse the output!
|
139 |
|
140 |
|
141 |
|
142 |
### Training Data
|
143 |
|
144 |
+
This is the list of task used for training and evaluating GoLLIE. However, as demonstrated in the ๐ [Create Custom Task notebook](https://github.com/hitz-zentroa/GoLLIE/blob/main/notebooks/Create%20Custom%20Task.ipynb) GoLLIE can perform a wide range of unseen tasks.
|
145 |
For more info, read our [๐Paper]().
|
146 |
|
147 |
<p align="center">
|
148 |
+
<img src="https://github.com/hitz-zentroa/GoLLIE/raw/main/assets/datasets.png">
|
149 |
</p>
|
150 |
|
151 |
|
|
|
162 |
|
163 |
| Model | Hardware | FLOPs | Time (h) | CO<sup>2</sup>eq (kg) |
|
164 |
|----------------|-------------------|---------------------------|-------------------|-------------------------------------|
|
165 |
+
| GoLLIE 7B | 1xA100 | 11.9e<sup>18</sup> | 44.5 | 1.57 |
|
166 |
+
| GoLLIE 13B | 1xA100 | 22.7e<sup>18</sup> | 79.5 | 2.80 |
|
167 |
+
| GoLLIE 34B | 2xA100 | 55.8e<sup>18</sup> | 94.6 | 6.67 |
|
168 |
|
169 |
|
170 |
|