plainqafact 0.1.0__tar.gz

plainqafact-0.1.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

plainqafact-0.1.0/MANIFEST.in

@@ -0,0 +1,8 @@
+include README.md
+include LICENSE
+include requirements.txt
+include plainqafact/plainqafact/transformers_old_tokenizer-3.1.0-py3-none-any.whl
+recursive-include plainqafact/plainqafact *.py
+recursive-include models *
+recursive-include qaeval *
+recursive-include src * 

plainqafact-0.1.0/PKG-INFO

@@ -0,0 +1,167 @@
+Metadata-Version: 2.2
+Name: plainqafact
+Version: 0.1.0
+Summary: A framework for evaluating plain language summaries using question answering
+Home-page: https://github.com/zhiwenyou103/PlainQAFact
+Author: Zhiwen You
+Author-email: zhiweny2@illinois.edu
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: openai==1.64.0
+Requires-Dist: bert-score==0.3.13
+Requires-Dist: sentence-transformers==3.4.1
+Requires-Dist: edlib==1.3.9.post1
+Requires-Dist: overrides==7.7.0
+Requires-Dist: spacy==3.8.3
+Requires-Dist: sentencepiece==0.2.0
+Requires-Dist: sacremoses==0.1.1
+Requires-Dist: gdown
+Requires-Dist: faiss-cpu==1.7.4
+Requires-Dist: python-liquid==1.10.2
+Requires-Dist: openai==0.28.0
+Requires-Dist: langchain==0.0.345
+Requires-Dist: tqdm==4.66.1
+Requires-Dist: transformers==4.44.2
+Requires-Dist: sentence_transformers==2.2.2
+Requires-Dist: pyserini==0.22.1
+Requires-Dist: tiktoken==0.6.0
+Requires-Dist: datasets==2.16.1
+Requires-Dist: torch
+Requires-Dist: torchvision
+Requires-Dist: torchaudio
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
+
+# PlainQAFact
+
+## News
+- [2025.02.24] Our PlainFact dataset can be downloaded here: [`PlainFact`](https://drive.google.com/drive/folders/1mbb06BbZWogweoxc1I5AE7I7m13qhiRL?usp=sharing), including sentence-level and summary-level granularities.
+    - Target_Sentence: The plain language sentence/summary.
+    - Original_Abstract: The scientific abstract corresponding to each sentence/summary.
+    - External: Whether the sentence includes information does not explicitly present in the scientific abstract. ('yes': explanation, 'no': simplification)
+- [2025.02.24] Our fine-tuned Question Generation model is available on 🤗 Hugging Face: [`QG model`](https://huggingface.co/uzw/bart-large-question-generation) (or download it [here](https://drive.google.com/file/d/1-MA9dfOtCm38yTfiQN9Xm8sRvcRD_Cmc/view?usp=drive_link))
+
+> NOTE: This repo is heavily relied on [QAFactEval](https://github.com/salesforce/QAFactEval), [QAEval](https://github.com/danieldeutsch/qaeval), and [MedRAG](https://github.com/Teddy-XiongGZ/MedRAG).
+
+
+## Overall Framework
+<div align="center">
+  <img src="https://github.com/zhiwenyou103/PlainQAFact/blob/main/pics/system.jpg" height="500" width="750">
+</div>
+
+
+## Installation
+
+### Option 1: Install via pip (Recommended)
+```bash
+pip install plainqafact
+```
+
+After installation, you can use PlainQAFact directly in your Python code:
+```python
+from plainqafact import PlainQAFact
+
+# Initialize the framework
+qafact = PlainQAFact(
+    classifier_type='learned',
+    classifier_path='path/to/learned_classifier',
+    llm_model_path='meta-llama/Llama-3.1-8B-Instruct',
+    question_generation_model_path='uzw/bart-large-question-generation',
+    qa_answering_model_dir='models/answering',
+    knowledge_base='combined',
+    answer_selection_strategy='llm-keywords'
+)
+
+# Run evaluation
+results = qafact.evaluate(input_file='your_data.csv')
+```
+
+### Option 2: Install from source
+- First, create a new conda env: `conda create -n plainqafact python=3.9` and clone our repo.
+- `cd PlainQAFact`
+- Follow the instructions in [MedRAG](https://github.com/Teddy-XiongGZ/MedRAG?tab=readme-ov-file#requirements) to install PyTorch and other required packages.
+- Then, run the following command:
+    ```bash
+    conda install git
+    pip install -r requirements.txt
+    ```
+- Finally, install the old tokenizer package through:
+    ```bash
+    pip install transformers_old_tokenizer-3.1.0-py3-none-any.whl
+    ```
+
+## Model Downloading
+In PlainQAFact, we use [`Llama 3.1 8B Instruct`](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) for answer extraction, fine-tuned [`QG model`](https://huggingface.co/uzw/bart-large-question-generation), and the original question answering model from [QAFactEval](https://github.com/salesforce/QAFactEval).
+
+Download the pre-trained QA model and our pre-trained classifier through `bash download_question_answering.sh`.
+
+
+## Usage of PlainQAFact
+### Running through our PlainFact dataset
+Before running the following command, please download the question answering and learned classifier models through above instructions. 
+```bash
+python3 run.py \
+    --classifier_type learned \  # Options: 'learned', 'llama', 'gpt'
+    --input_file data/summary_level.csv \ # path of the input dataset 
+    --classifier_path path/to/learned_classifier \ # path of the classifier
+    --llm_model_path meta-llama/Llama-3.1-8B-Instruct \ # path of the answer extractor
+    --question_generation_model_path uzw/bart-large-question-generation \ # path of the question generation model
+    --qa_answering_model_dir models/answering \ # path of the question answering model
+    --knowledge_base combined \ # knowledge bases for retrieval
+    --answer_selection_strategy llm-keywords  # Options: 'llm-keywords', 'gpt-keywords', 'none'
+```
+
+### Running through your own data
+Please modify the [`default_config.py`](https://github.com/zhiwenyou103/PlainQAFact/blob/main/default_config.py#L17) file `Line 17-19` to indicate the heading/key names of your dataset. We currently support `.json`, `.txt`, and `.csv` file.
+```bash
+python3 run.py \
+    --classifier_type learned \
+    --input_file your_own_data.json \
+    --input_file_format json \
+    --classifier_path path/to/learned_classifier \
+    --llm_model_path meta-llama/Llama-3.1-8B-Instruct \
+    --question_generation_model_path uzw/bart-large-question-generation \
+    --qa_answering_model_dir models/answering \
+    --knowledge_base textbooks \
+    --answer_selection_strategy llm-keywords
+```
+
+### Easily replace the pre-trained classifier to OpenAI models or your own
+We provides options to easily replace our pre-trained classisifer tailored for the biomedical plain language summarization tasks to other tasks. You may simply set `--classifier_type` as `gpt` and provide your OpenAI API key in the [`default_config.py`](https://github.com/zhiwenyou103/PlainQAFact/blob/main/default_config.py#L26) file `Line 26` to run PlainQAFact.
+```bash
+python3 run.py \
+    --classifier_type gpt \
+    --input_file your_own_data.json \
+    --input_file_format json \
+    --llm_model_path meta-llama/Llama-3.1-8B-Instruct \
+    --question_generation_model_path uzw/bart-large-question-generation \
+    --qa_answering_model_dir models/answering \
+    --knowledge_base textbooks \
+    --answer_selection_strategy llm-keywords
+```
+
+### Using other Knowledge Bases for retrieval
+Currently, we only experiment with two KBs: Textbooks and StatPearls. You may want to use your customized KBs for more accurate retrieval. In PlainQAFact, we combine both Textbooks and StatPearls and concatenate with the scientific abstracts. Set `--knowledge_base textbooks` as `combined` to reproduce our results.
+
+
+> NOTE: Using Llama 3.1 8B model for both classification and answer extraction would take over 40 GB GPU memory. We recommend to use our pre-trained classifier or OpenAI models for classification if the GPU memory is limited.
+
+
+## Citation Information
+For the use of PlainQAFact and PlainFact benchmark, please cite:
+
+## Contact Information
+If you have any questions, please email `zhiweny2@illinois.edu`.

plainqafact-0.1.0/README.md

@@ -0,0 +1,120 @@
+# PlainQAFact
+
+## News
+- [2025.02.24] Our PlainFact dataset can be downloaded here: [`PlainFact`](https://drive.google.com/drive/folders/1mbb06BbZWogweoxc1I5AE7I7m13qhiRL?usp=sharing), including sentence-level and summary-level granularities.
+    - Target_Sentence: The plain language sentence/summary.
+    - Original_Abstract: The scientific abstract corresponding to each sentence/summary.
+    - External: Whether the sentence includes information does not explicitly present in the scientific abstract. ('yes': explanation, 'no': simplification)
+- [2025.02.24] Our fine-tuned Question Generation model is available on 🤗 Hugging Face: [`QG model`](https://huggingface.co/uzw/bart-large-question-generation) (or download it [here](https://drive.google.com/file/d/1-MA9dfOtCm38yTfiQN9Xm8sRvcRD_Cmc/view?usp=drive_link))
+
+> NOTE: This repo is heavily relied on [QAFactEval](https://github.com/salesforce/QAFactEval), [QAEval](https://github.com/danieldeutsch/qaeval), and [MedRAG](https://github.com/Teddy-XiongGZ/MedRAG).
+
+
+## Overall Framework
+<div align="center">
+  <img src="https://github.com/zhiwenyou103/PlainQAFact/blob/main/pics/system.jpg" height="500" width="750">
+</div>
+
+
+## Installation
+
+### Option 1: Install via pip (Recommended)
+```bash
+pip install plainqafact
+```
+
+After installation, you can use PlainQAFact directly in your Python code:
+```python
+from plainqafact import PlainQAFact
+
+# Initialize the framework
+qafact = PlainQAFact(
+    classifier_type='learned',
+    classifier_path='path/to/learned_classifier',
+    llm_model_path='meta-llama/Llama-3.1-8B-Instruct',
+    question_generation_model_path='uzw/bart-large-question-generation',
+    qa_answering_model_dir='models/answering',
+    knowledge_base='combined',
+    answer_selection_strategy='llm-keywords'
+)
+
+# Run evaluation
+results = qafact.evaluate(input_file='your_data.csv')
+```
+
+### Option 2: Install from source
+- First, create a new conda env: `conda create -n plainqafact python=3.9` and clone our repo.
+- `cd PlainQAFact`
+- Follow the instructions in [MedRAG](https://github.com/Teddy-XiongGZ/MedRAG?tab=readme-ov-file#requirements) to install PyTorch and other required packages.
+- Then, run the following command:
+    ```bash
+    conda install git
+    pip install -r requirements.txt
+    ```
+- Finally, install the old tokenizer package through:
+    ```bash
+    pip install transformers_old_tokenizer-3.1.0-py3-none-any.whl
+    ```
+
+## Model Downloading
+In PlainQAFact, we use [`Llama 3.1 8B Instruct`](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) for answer extraction, fine-tuned [`QG model`](https://huggingface.co/uzw/bart-large-question-generation), and the original question answering model from [QAFactEval](https://github.com/salesforce/QAFactEval).
+
+Download the pre-trained QA model and our pre-trained classifier through `bash download_question_answering.sh`.
+
+
+## Usage of PlainQAFact
+### Running through our PlainFact dataset
+Before running the following command, please download the question answering and learned classifier models through above instructions. 
+```bash
+python3 run.py \
+    --classifier_type learned \  # Options: 'learned', 'llama', 'gpt'
+    --input_file data/summary_level.csv \ # path of the input dataset 
+    --classifier_path path/to/learned_classifier \ # path of the classifier
+    --llm_model_path meta-llama/Llama-3.1-8B-Instruct \ # path of the answer extractor
+    --question_generation_model_path uzw/bart-large-question-generation \ # path of the question generation model
+    --qa_answering_model_dir models/answering \ # path of the question answering model
+    --knowledge_base combined \ # knowledge bases for retrieval
+    --answer_selection_strategy llm-keywords  # Options: 'llm-keywords', 'gpt-keywords', 'none'
+```
+
+### Running through your own data
+Please modify the [`default_config.py`](https://github.com/zhiwenyou103/PlainQAFact/blob/main/default_config.py#L17) file `Line 17-19` to indicate the heading/key names of your dataset. We currently support `.json`, `.txt`, and `.csv` file.
+```bash
+python3 run.py \
+    --classifier_type learned \
+    --input_file your_own_data.json \
+    --input_file_format json \
+    --classifier_path path/to/learned_classifier \
+    --llm_model_path meta-llama/Llama-3.1-8B-Instruct \
+    --question_generation_model_path uzw/bart-large-question-generation \
+    --qa_answering_model_dir models/answering \
+    --knowledge_base textbooks \
+    --answer_selection_strategy llm-keywords
+```
+
+### Easily replace the pre-trained classifier to OpenAI models or your own
+We provides options to easily replace our pre-trained classisifer tailored for the biomedical plain language summarization tasks to other tasks. You may simply set `--classifier_type` as `gpt` and provide your OpenAI API key in the [`default_config.py`](https://github.com/zhiwenyou103/PlainQAFact/blob/main/default_config.py#L26) file `Line 26` to run PlainQAFact.
+```bash
+python3 run.py \
+    --classifier_type gpt \
+    --input_file your_own_data.json \
+    --input_file_format json \
+    --llm_model_path meta-llama/Llama-3.1-8B-Instruct \
+    --question_generation_model_path uzw/bart-large-question-generation \
+    --qa_answering_model_dir models/answering \
+    --knowledge_base textbooks \
+    --answer_selection_strategy llm-keywords
+```
+
+### Using other Knowledge Bases for retrieval
+Currently, we only experiment with two KBs: Textbooks and StatPearls. You may want to use your customized KBs for more accurate retrieval. In PlainQAFact, we combine both Textbooks and StatPearls and concatenate with the scientific abstracts. Set `--knowledge_base textbooks` as `combined` to reproduce our results.
+
+
+> NOTE: Using Llama 3.1 8B model for both classification and answer extraction would take over 40 GB GPU memory. We recommend to use our pre-trained classifier or OpenAI models for classification if the GPU memory is limited.
+
+
+## Citation Information
+For the use of PlainQAFact and PlainFact benchmark, please cite:
+
+## Contact Information
+If you have any questions, please email `zhiweny2@illinois.edu`.

plainqafact-0.1.0/models/README.md

@@ -0,0 +1 @@
+Your downloaded model will be shown here.

plainqafact-0.1.0/plainqafact/plainqafact/__init__.py

@@ -0,0 +1,8 @@
+
+__version__ = '0.1.0'
+
+from .plainqafact import PlainQAFact
+from .classifier import Classifier
+from .default_config import DefaultConfig
+
+__all__ = ['PlainQAFact', 'Classifier', 'DefaultConfig'] 

plainqafact-0.1.0/plainqafact/plainqafact/bertscore_scorer.py

@@ -0,0 +1,41 @@
+from typing import Dict, List, Set
+import bert_score
+import numpy as np
+from qaeval.scoring.scorers.scorer import Scorer
+
+class BertScoreScorer(Scorer):
+    def __init__(self, cuda_device: int, batch_size: int = 8) -> None:
+        self.device = f'cuda:{cuda_device}' if cuda_device >= 0 else 'cpu'
+        self.batch_size = batch_size
+
+    def keys(self) -> Set[str]:
+        return {'bertscore'}
+
+    def _score_single_ref(
+        self,
+        context: str,
+        questions: List[str],
+        answers: List[str],
+        predictions: List[str],
+        probabilities: List[float],
+        null_probabilities: List[float]
+    ) -> List[Dict[str, float]]:
+        indices = []
+        valid_answers = []
+        valid_predictions = []
+        
+        for i, (answer, prediction, probability, null_probability) in \
+                enumerate(zip(answers, predictions, probabilities, null_probabilities)):
+            if probability > null_probability:
+                valid_answers.append(answer)
+                valid_predictions.append(prediction)
+                indices.append(i)
+
+        scores = [0.0] * len(questions)
+        if valid_predictions:
+            _, _, F1 = bert_score.score(valid_predictions, valid_answers, 
+                                      lang="en", device=self.device)
+            for idx, score in zip(indices, F1.tolist()):
+                scores[idx] = score
+
+        return [{'bertscore': s} for s in scores]

plainqafact-0.1.0/plainqafact/plainqafact/classifier.py

@@ -0,0 +1,113 @@
+import torch
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import re
+from transformers import pipeline
+from openai import OpenAI
+
+class LearnedClassifier:
+    def __init__(self, model_path, device):
+        self.device = torch.device(device if torch.cuda.is_available() else 'cpu')
+        self.tokenizer = AutoTokenizer.from_pretrained(model_path)
+        self.model = AutoModelForSequenceClassification.from_pretrained(model_path, use_safetensors=True)
+        self.model.to(self.device)
+        self.model.eval()
+
+    def predict(self, abstract, sentence, return_probability=False):
+        encoding = self.tokenizer.encode_plus(
+            text=abstract,
+            text_pair=sentence,
+            add_special_tokens=True,
+            max_length=512,
+            padding='max_length',
+            truncation=True,
+            return_attention_mask=True,
+            return_tensors='pt'
+        )
+
+        input_ids = encoding['input_ids'].to(self.device)
+        attention_mask = encoding['attention_mask'].to(self.device)
+
+        with torch.no_grad():
+            outputs = self.model(input_ids=input_ids, attention_mask=attention_mask)
+            probabilities = torch.nn.functional.softmax(outputs.logits, dim=1)
+
+            predicted_class = torch.argmax(probabilities, dim=1).item()
+            probabilities = probabilities[0].cpu().numpy()
+
+        label = 'yes' if predicted_class == 1 else 'no'
+
+        if return_probability:
+            return label, {'no': probabilities[0], 'yes': probabilities[1]}
+        else:
+            confidence = probabilities[predicted_class]
+            return label, confidence
+
+
+class LLMClassifier:
+    def __init__(self, classifier_type: str, model_path: str = None, device: int = None, openai_key: str = None):
+        self.max_tokens = 512
+        self.classifier_type = classifier_type
+
+        if classifier_type == 'llama':
+            self.generator = pipeline(
+                "text-generation",
+                model=model_path,
+                device=device
+            )
+        elif classifier_type == 'gpt':
+            self.client = OpenAI(api_key=openai_key)
+
+    def predict(self, abstract: str, sentence: str) -> tuple:
+        if self.classifier_type == 'llama':
+            label = self.llama_classifier(sentence, abstract)
+        else:
+            label = self.gpt_classifier(sentence, abstract)
+
+        confidence = 1.0  # placeholder
+        return label, confidence
+
+    def gpt_classifier(self, summary: str, abstract: str) -> str:
+        response = self.client.chat.completions.create(
+            model="gpt-4o-2024-11-20",
+            messages=[
+                {"role": "developer",
+                 "content": "Annotate whether a sentence or summary includes information not present in the original abstract. \n The sentence or summary contains external information that is not explicitly mentioned, paraphrased, or implied in the original abstract will be labeled as 'Yes'. \n The sentence or summary contains information that is explicitly stated or closely paraphrased from the original abstract will be labeled as 'No'."},
+                {"role": "user",
+                 "content": "Sentence or summary: " + summary + "\n Original abstract: " + abstract},
+            ],
+            max_tokens=self.max_tokens,
+            temperature=0,
+        )
+        if response.choices[0].message.content == '':
+            print('empty')
+            result = 'Yes'
+        else:
+            result = response.choices[0].message.content
+            result = re.sub(r'[^a-zA-Z0-9\s]', '', result)
+
+        return result.lower().strip()
+
+    def llama_classifier(self, summary: str, abstract: str) -> str:
+        messages = [
+            {"role": "system",
+             "content": "Annotate whether a summary includes information not present in the original abstract. \n The summary contains external information that is not explicitly mentioned, paraphrased, or implied in the original abstract will be labeled as 'Yes'. \n The summary contains information that is explicitly stated or closely paraphrased from the original abstract will be labeled as 'No'."},
+            {"role": "user",
+             "content": "Summary: " + summary + "\n Original abstract: " + abstract + "\n Your annotation: "}
+        ]
+
+        outputs = self.generator(messages, do_sample=True, temperature=0.01, max_new_tokens=self.max_tokens)
+        generated_text = outputs[0]['generated_text'][-1]['content']
+
+        pattern = r'annotation is:\s*(Yes|No)\b'
+        match = re.search(pattern, generated_text, flags=re.IGNORECASE)
+        if match:
+            final_answer = match.group(1)
+        else:
+            match = re.search(r'(Yes|No)\s*$', generated_text, flags=re.IGNORECASE)
+            if match:
+                final_answer = match.group(1)
+            else:
+                print("empty")
+                final_answer = 'Yes'
+
+        return final_answer.lower().strip()