model update

README.md

@@ -26,7 +26,7 @@ widget:
 - text: "extract answers: Iran === Landwirtschaft === Die landwirtschaftliche Nutzfläche beträgt trotz zahlreicher Gebirge und Wüsten 10 % der Landesfläche, wobei ein Drittel künstlich bewässert wird. Die Landwirtschaft ist einer der größten Arbeitgeber des Landes. Wichtige Produkte sind Pistazien, Weizen, Reis, Zucker, Baumwolle, Früchte, Nüsse, Datteln, Wolle und Kaviar. Seit der Revolution von 1979 wurde der Anbau von Weintrauben wegen des islamischen Alkoholverbots auf den 200.000 Hektar Rebfläche fast vollständig auf Tafeltrauben und Rosinen umgestellt. Bei Rosinen ist <hl> der Iran <hl> inzwischen nach der Türkei der zweitgrößte Exporteur der Welt, bei Safran mit ungefähr 90 % Marktanteil des globalen Bedarfs mit Abstand der größte."
   example_title: "Answer Extraction Example 2" 
 model-index:
-- name: lmqg/mt5-base-dequad-multitask
+- name: lmqg/mt5-base-dequad-qg-ae
   results:
   - task:
       name: Text2text Generation
@@ -51,34 +51,49 @@ model-index:
     - name: MoverScore (Question Generation)
       type: moverscore_question_generation
       value: 53.77
-    - name: QAAlignedF1Score-BERTScore
-      type: qa_aligned_f1_score_bertscore
+    - name: QAAlignedF1Score-BERTScore (Question & Answer Generation)
+      type: qa_aligned_f1_score_bertscore_question_answer_generation
       value: 6.11
-    - name: QAAlignedRecall-BERTScore
-      type: qa_aligned_recall_bertscore
+    - name: QAAlignedRecall-BERTScore (Question & Answer Generation)
+      type: qa_aligned_recall_bertscore_question_answer_generation
       value: 5.95
-    - name: QAAlignedPrecision-BERTScore
-      type: qa_aligned_precision_bertscore
+    - name: QAAlignedPrecision-BERTScore (Question & Answer Generation)
+      type: qa_aligned_precision_bertscore_question_answer_generation
       value: 6.3
-    - name: QAAlignedF1Score-MoverScore
-      type: qa_aligned_f1_score_moverscore
+    - name: QAAlignedF1Score-MoverScore (Question & Answer Generation)
+      type: qa_aligned_f1_score_moverscore_question_answer_generation
       value: 4.24
-    - name: QAAlignedRecall-MoverScore
-      type: qa_aligned_recall_moverscore
+    - name: QAAlignedRecall-MoverScore (Question & Answer Generation)
+      type: qa_aligned_recall_moverscore_question_answer_generation
       value: 4.15
-    - name: QAAlignedPrecision-MoverScore
-      type: qa_aligned_precision_moverscore
+    - name: QAAlignedPrecision-MoverScore (Question & Answer Generation)
+      type: qa_aligned_precision_moverscore_question_answer_generation
       value: 4.34
+    - name: BLEU4 (Answer Extraction)
+      type: bleu4_answer_extraction
+      value: 0.85
+    - name: ROUGE-L (Answer Extraction)
+      type: rouge_l_answer_extraction
+      value: 3.93
+    - name: METEOR (Answer Extraction)
+      type: meteor_answer_extraction
+      value: 5.34
+    - name: BERTScore (Answer Extraction)
+      type: bertscore_answer_extraction
+      value: 63.01
+    - name: MoverScore (Answer Extraction)
+      type: moverscore_answer_extraction
+      value: 47.54
     - name: AnswerF1Score (Answer Extraction)
-      type: answer_f1_score_answer_extraction
+      type: answer_f1_score__answer_extraction
       value: 8.63
     - name: AnswerExactMatch (Answer Extraction)
       type: answer_exact_match_answer_extraction
       value: 0.32
 ---
 
-# Model Card of `lmqg/mt5-base-dequad-multitask`
-This model is fine-tuned version of [google/mt5-base](https://huggingface.co/google/mt5-base) for question generation task and answer extraction jointly on the [lmqg/qg_dequad](https://huggingface.co/datasets/lmqg/qg_dequad) (dataset_name: default) via [`lmqg`](https://github.com/asahi417/lm-question-generation).
+# Model Card of `lmqg/mt5-base-dequad-qg-ae`
+This model is fine-tuned version of [google/mt5-base](https://huggingface.co/google/mt5-base) for question generation and answer extraction jointly on the [lmqg/qg_dequad](https://huggingface.co/datasets/lmqg/qg_dequad) (dataset_name: default) via [`lmqg`](https://github.com/asahi417/lm-question-generation).
 
 
 ### Overview
@@ -95,7 +110,7 @@ This model is fine-tuned version of [google/mt5-base](https://huggingface.co/goo
 from lmqg import TransformersQG
 
 # initialize model
-model = TransformersQG(language="de", model="lmqg/mt5-base-dequad-multitask")
+model = TransformersQG(language="de", model="lmqg/mt5-base-dequad-qg-ae")
 
 # model prediction
 question_answer_pairs = model.generate_qa("das erste weltweit errichtete Hermann Brehmer 1855 im niederschlesischen ''Görbersdorf'' (heute Sokołowsko, Polen).")
@@ -106,7 +121,7 @@ question_answer_pairs = model.generate_qa("das erste weltweit errichtete Hermann
 ```python
 from transformers import pipeline
 
-pipe = pipeline("text2text-generation", "lmqg/mt5-base-dequad-multitask")
+pipe = pipeline("text2text-generation", "lmqg/mt5-base-dequad-qg-ae")
 
 # answer extraction
 answer = pipe("generate question: Empfangs- und Sendeantenne sollen in ihrer Polarisation übereinstimmen, andernfalls <hl> wird die Signalübertragung stark gedämpft. <hl>")
@@ -119,7 +134,7 @@ question = pipe("extract answers: Sommerzeit <hl> Frühling <hl>: Umstellung von
 ## Evaluation
 
 
-- ***Metric (Question Generation)***: [raw metric file](https://huggingface.co/lmqg/mt5-base-dequad-multitask/raw/main/eval/metric.first.sentence.paragraph_answer.question.lmqg_qg_dequad.default.json) 
+- ***Metric (Question Generation)***: [raw metric file](https://huggingface.co/lmqg/mt5-base-dequad-qg-ae/raw/main/eval/metric.first.sentence.paragraph_answer.question.lmqg_qg_dequad.default.json) 
 
 |            |   Score | Type    | Dataset                                                          |
 |:-----------|--------:|:--------|:-----------------------------------------------------------------|
@@ -133,7 +148,7 @@ question = pipe("extract answers: Sommerzeit <hl> Frühling <hl>: Umstellung von
 | ROUGE_L    |    8.58 | default | [lmqg/qg_dequad](https://huggingface.co/datasets/lmqg/qg_dequad) |
 
 
-- ***Metric (Question & Answer Generation)***:  [raw metric file](https://huggingface.co/lmqg/mt5-base-dequad-multitask/raw/main/eval/metric.first.answer.paragraph.questions_answers.lmqg_qg_dequad.default.json)
+- ***Metric (Question & Answer Generation)***:  [raw metric file](https://huggingface.co/lmqg/mt5-base-dequad-qg-ae/raw/main/eval/metric.first.answer.paragraph.questions_answers.lmqg_qg_dequad.default.json)
 
 |                                 |   Score | Type    | Dataset                                                          |
 |:--------------------------------|--------:|:--------|:-----------------------------------------------------------------|
@@ -145,7 +160,7 @@ question = pipe("extract answers: Sommerzeit <hl> Frühling <hl>: Umstellung von
 | QAAlignedRecall (MoverScore)    |    4.15 | default | [lmqg/qg_dequad](https://huggingface.co/datasets/lmqg/qg_dequad) |
 
 
-- ***Metric (Answer Generation)***: [raw metric file](https://huggingface.co/lmqg/mt5-base-dequad-multitask/raw/main/eval/metric.first.answer.paragraph_sentence.answer.lmqg_qg_dequad.default.json)
+- ***Metric (Answer Extraction)***: [raw metric file](https://huggingface.co/lmqg/mt5-base-dequad-qg-ae/raw/main/eval/metric.first.answer.paragraph_sentence.answer.lmqg_qg_dequad.default.json)
 
 |                  |   Score | Type    | Dataset                                                          |
 |:-----------------|--------:|:--------|:-----------------------------------------------------------------|
@@ -181,7 +196,7 @@ The following hyperparameters were used during fine-tuning:
  - gradient_accumulation_steps: 2
  - label_smoothing: 0.15
 
-The full configuration can be found at [fine-tuning config file](https://huggingface.co/lmqg/mt5-base-dequad-multitask/raw/main/trainer_config.json).
+The full configuration can be found at [fine-tuning config file](https://huggingface.co/lmqg/mt5-base-dequad-qg-ae/raw/main/trainer_config.json).
 
 ## Citation
 ```