Add library_name: transformers to metadata
#1
by
nielsr
HF staff
- opened
README.md
CHANGED
@@ -1,5 +1,6 @@
|
|
1 |
---
|
2 |
-
|
|
|
3 |
datasets:
|
4 |
- QCRI/LlamaLens-English
|
5 |
- QCRI/LlamaLens-Arabic
|
@@ -8,8 +9,11 @@ language:
|
|
8 |
- ar
|
9 |
- en
|
10 |
- hi
|
11 |
-
|
12 |
-
|
|
|
|
|
|
|
13 |
pipeline_tag: text-generation
|
14 |
tags:
|
15 |
- Social-Media
|
@@ -17,12 +21,10 @@ tags:
|
|
17 |
- Summarization
|
18 |
- offensive-language
|
19 |
- News-Genre
|
20 |
-
|
21 |
-
- accuracy
|
22 |
-
- f1
|
23 |
-
- rouge
|
24 |
---
|
25 |
-
|
|
|
26 |
|
27 |
## Overview
|
28 |
LlamaLens is a specialized multilingual LLM designed for analyzing news and social media content. It focuses on 18 NLP tasks, leveraging 52 datasets across Arabic, English, and Hindi.
|
@@ -188,7 +190,6 @@ Below, we present the performance of **L-Lens: LlamaLens** , where *"Eng"* refe
|
|
188 |
| Sarcasm Detection | News-Headlines-Dataset-For-Sarcasm-Detection | Acc | 0.897 | 0.668 | 0.936 | 0.947 | 0.039 |
|
189 |
| Sentiment Classification | NewsMTSC-dataset | Ma-F1 | 0.817 | 0.628 | 0.751 | 0.748 | -0.066 |
|
190 |
| Subjectivity Detection | clef2024-checkthat-lab | Ma-F1 | 0.744 | 0.535 | 0.642 | 0.628 | -0.102 |
|
191 |
-
|
|
192 |
|
193 |
---
|
194 |
|
@@ -203,12 +204,10 @@ Below, we present the performance of **L-Lens: LlamaLens** , where *"Eng"* refe
|
|
203 |
| News Summarization | xlsum | R-2 | 0.136 | 0.078 | 0.171 | 0.170 | 0.035 |
|
204 |
| Offensive Language Detection | Offensive Speech Detection | Mi-F1 | 0.723 | 0.621 | 0.862 | 0.865 | 0.139 |
|
205 |
| Cyberbullying Detection | MC_Hinglish1 | Acc | 0.609 | 0.233 | 0.625 | 0.627 | 0.016 |
|
206 |
-
| Sentiment Classification | Sentiment Analysis | Acc | 0.697 | 0.552 | 0.647 | 0.654 | -0.050
|
207 |
|
208 |
## Paper
|
209 |
-
For an in-depth understanding, refer to our paper: [
|
210 |
-
|
211 |
-
|
212 |
|
213 |
|
214 |
# License
|
|
|
1 |
---
|
2 |
+
base_model:
|
3 |
+
- meta-llama/Llama-3.1-8B-Instruct
|
4 |
datasets:
|
5 |
- QCRI/LlamaLens-English
|
6 |
- QCRI/LlamaLens-Arabic
|
|
|
9 |
- ar
|
10 |
- en
|
11 |
- hi
|
12 |
+
license: cc-by-nc-sa-4.0
|
13 |
+
metrics:
|
14 |
+
- accuracy
|
15 |
+
- f1
|
16 |
+
- rouge
|
17 |
pipeline_tag: text-generation
|
18 |
tags:
|
19 |
- Social-Media
|
|
|
21 |
- Summarization
|
22 |
- offensive-language
|
23 |
- News-Genre
|
24 |
+
library_name: transformers
|
|
|
|
|
|
|
25 |
---
|
26 |
+
|
27 |
+
# LlamaLens: Specialized Multilingual LLM for Analyzing News and Social Media Content
|
28 |
|
29 |
## Overview
|
30 |
LlamaLens is a specialized multilingual LLM designed for analyzing news and social media content. It focuses on 18 NLP tasks, leveraging 52 datasets across Arabic, English, and Hindi.
|
|
|
190 |
| Sarcasm Detection | News-Headlines-Dataset-For-Sarcasm-Detection | Acc | 0.897 | 0.668 | 0.936 | 0.947 | 0.039 |
|
191 |
| Sentiment Classification | NewsMTSC-dataset | Ma-F1 | 0.817 | 0.628 | 0.751 | 0.748 | -0.066 |
|
192 |
| Subjectivity Detection | clef2024-checkthat-lab | Ma-F1 | 0.744 | 0.535 | 0.642 | 0.628 | -0.102 |
|
|
|
193 |
|
194 |
---
|
195 |
|
|
|
204 |
| News Summarization | xlsum | R-2 | 0.136 | 0.078 | 0.171 | 0.170 | 0.035 |
|
205 |
| Offensive Language Detection | Offensive Speech Detection | Mi-F1 | 0.723 | 0.621 | 0.862 | 0.865 | 0.139 |
|
206 |
| Cyberbullying Detection | MC_Hinglish1 | Acc | 0.609 | 0.233 | 0.625 | 0.627 | 0.016 |
|
207 |
+
| Sentiment Classification | Sentiment Analysis | Acc | 0.697 | 0.552 | 0.647 | 0.654 | -0.050 |
|
208 |
|
209 |
## Paper
|
210 |
+
For an in-depth understanding, refer to our paper: [LlamaLens: Specialized Multilingual LLM for Analyzing News and Social Media Content](https://arxiv.org/pdf/2410.15308).
|
|
|
|
|
211 |
|
212 |
|
213 |
# License
|