Update README.md

README.md

@@ -24,10 +24,11 @@ p{color:Black !important;}
 <!--  <img src="./figs/Example.png" width=70%/> -->
 </p>
 
+<div align="left">
+    <span style="font-size:16px;">Wikipedia contradict benchmark is a dataset consisting of 253 high-quality, human-annotated instances designed to assess LLM performance when augmented with retrieved passages containing real-world knowledge conflicts. The dataset was created intentionally with that task in mind, focusing on a benchmark consisting of high-quality, human-annotated instances.</span>
+</div>
 
 
-Wikipedia contradict benchmark is a dataset consisting of 253 high-quality, human-annotated instances designed to assess LLM performance when augmented with retrieved passages containing real-world knowledge conflicts. The dataset was created intentionally with that task in mind, focusing on a benchmark consisting of high-quality, human-annotated instances.
-
 <!--  Note that, in the dataset viewer, there are 130 valid-tag instances, but each instance can contain more that one question and its respective two answers. Then, the total number of questions and answers is 253. -->
 
 <!-- This dataset card has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/datasetcard_template.md?plain=1). --> 
@@ -38,9 +39,22 @@ Wikipedia contradict benchmark is a dataset consisting of 253 high-quality, huma
 
 <!-- Provide a longer summary of what this dataset is. -->
 
+<div align="left">
+    <span style="font-size:16px;">
 Wikipedia contradict benchmark is a QA-based benchmark consisting of 253 human-annotated instances that cover different types of real-world knowledge conflicts. 
+<br><br>      
+    </span>
+</div>
+
 
+<div align="left">
+    <span style="font-size:16px;">
 Each instance consists of a question, a pair of contradictory passages extracted from Wikipedia, and two distinct answers, each derived from on the passages. The pair is annotated by a human annotator who identify where the conflicted information is and what type of conflict is observed. The annotator then produces a set of questions related to the passages with different answers reflecting the conflicting source of knowledge.
+    </span>
+</div>
+
+
+
 
 - **Curated by:** Yufang Hou, Alessandra Pascale, Javier Carnerero-Cano, Tigran Tchrakian, Radu Marinescu, Elizabeth Daly, Inkit Padhi, and Prasanna Sattigeri. All authors are employed by IBM Research.
 <!-- - **Funded by [optional]:** There was no associated grant. -->
@@ -64,41 +78,66 @@ Each instance consists of a question, a pair of contradictory passages extracted
 
 <!-- This section describes suitable use cases for the dataset. -->
 
+
+<div align="left">
+    <span style="font-size:16px;">
 The dataset has been used in the paper to assess LLMs performance when augmented with retrieved passages containing real-world knowledge conflicts.
+<br><br>   
+      The following figure illustrates the evaluation process:
+    </span>
+</div>
 
-The following figure illustrates the evaluation process:
 
 <p align="center">
   <img src="https://github.com/javiccano/Wikipedia_contradict_benchmark/blob/main/figs/Evaluation.png?raw=true" width=70%/>
 <!--  <img src="./figs/Evaluation.png" width=70%/> -->
 </p>
 
+
+<div align="left">
+    <span style="font-size:16px;">
 And the following table shows the performance of five LLMs (Mistral-7b-inst, Mixtral-8x7b-inst, Llama-2-70b-chat, Llama-3-70b-inst, and GPT-4) on the Wikipedia Contradict Benchmark based on rigorous human evaluations on a subset of answers for 55 instances, which corresponds to 1,375 LLM responses in total.
+    </span>
+</div>
+
 
 <p align="center">
   <img src="https://github.com/javiccano/Wikipedia_contradict_benchmark/blob/main/figs/table2.png?raw=true" width=70%/> 
   <!-- <img src="./figs/table2.png" width=70%/> -->
 </p>
 
-Notes: “C”, “PC” and “IC” stand for “Correct”, “Partially correct”, “Incorrect”, respectively. “all”, “exp”, and “imp” represent for instance
-types: all instances, instances with explicit conflicts, and instances with implicit conflicts. The
-numbers represent the ratio of responses from each LLM that were assessed as “Correct, “Partially
-correct, or “Incorrect for each instance type under a prompt template. The bold numbers highlight
-the best models that correctly answer questions for each type and prompt template.
+
+<div align="left">
+    <span style="font-size:16px;">
+Notes: “C”, “PC” and “IC” stand for “Correct”, “Partially correct”, “Incorrect”, respectively. “all”, “exp”, and “imp” represent for instance types: all instances, instances with explicit conflicts, and instances with implicit conflicts. The numbers represent the ratio of responses from each LLM that were assessed as “Correct, “Partially correct, or “Incorrect for each instance type under a prompt template. The bold numbers highlight the best models that correctly answer questions for each type and prompt template.
+    </span>
+</div>
+
+
 
 ### Out-of-Scope Use
 
 <!-- This section addresses misuse, malicious use, and uses that the dataset will not work well for. -->
 
+<div align="left">
+    <span style="font-size:16px;">
 N/A.
+    </span>
+</div>
 
 ## Dataset Structure
 
 <!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. -->
 
+
+<div align="left">
+    <span style="font-size:16px;">
 Wikipedia contradict benchmark is given in CSV format to store the corresponding information, so researchers can easily use our data. There are 253 instances in total.
+    <br><br>   
+      The description of each field (when the instance contains two questions) is as follows:
+    </span>
+</div>
 
-The description of each field (when the instance contains two questions) is as follows:
 
 
 - **question_ID:** ID of question.
@@ -118,8 +157,12 @@ The description of each field (when the instance contains two questions) is as f
 
 ## Usage of the Dataset
 
-We provide the following starter code. Please refer to the [GitHub repository](https://github.com/) for more information about the functions ```load_testingdata``` and ```generateAnswers_bam_models```.
 
+<div align="left">
+    <span style="font-size:16px;">
+We provide the following starter code. Please refer to the [GitHub repository](https://github.com/) for more information about the functions ```load_testingdata``` and ```generateAnswers_bam_models```.
+    </span>
+</div>
 
 ```python
 from genai import Client, Credentials
@@ -178,7 +221,12 @@ generateAnswers_bam_models(testingUnits)
 
 <!-- Motivation for the creation of this dataset. -->
 
+<div align="left">
+    <span style="font-size:16px;">
 Retrieval-augmented generation (RAG) has emerged as a promising solution to mitigate the limitations of large language models (LLMs), such as hallucinations and outdated information. However, it remains unclear how LLMs handle knowledge conflicts arising from different augmented retrieved passages, especially when these passages originate from the same source and have equal trustworthiness. In this regard, the motivation of Wikipedia Contradict Benchmark is to comprehensively evaluate LLM-generated answers to questions that have varying answers based on contradictory passages from Wikipedia, a dataset widely regarded as a high-quality pre-training resource for most LLMs.
+    </span>
+</div>
+
 
 ### Source Data
 
@@ -188,13 +236,24 @@ Retrieval-augmented generation (RAG) has emerged as a promising solution to miti
 
 <!-- This section describes the data collection and processing process such as data selection criteria, filtering and normalization methods, tools and libraries used, etc. -->
 
+
+<div align="left">
+    <span style="font-size:16px;">
 The data was mostly observable as raw text. The raw data was retrieved from Wikipedia articles containing inconsistent, self-contradictory, and contradict-other tags. The first two tags denote contradictory statements within the same article, whereas the third tag highlights instances where the content of one article contradicts that of another article. In total, we collected around 1,200 articles that contain these tags through the Wikipedia maintenance category “Wikipedia articles with content issues”. Given a content inconsistency tag provided by Wikipedia editors, the annotators verified whether the tag is valid by checking the relevant article content, the editor’s comment, as well as the information in the edit history and the article’s talk page if necessary.
+    </span>
+</div>
 
 #### Who are the source data producers?
 
 <!-- This section describes the people or systems who originally created the data. It should also include self-reported demographic or identity information for the source data creators if this information is available. -->
 
+
+<div align="left">
+    <span style="font-size:16px;">
 Wikipedia contributors.
+    </span>
+</div>
+
 
 ### Annotations
 
@@ -204,44 +263,79 @@ Wikipedia contributors.
 
 <!-- This section describes the annotation process such as annotation tools used in the process, the amount of data annotated, annotation guidelines provided to the annotators, interannotator statistics, annotation validation, etc. -->
 
+<div align="left">
+    <span style="font-size:16px;">
 The annotation interface was developed using [Label Studio](https://labelstud.io/).
-
+<br><br>   
 The annotators were required to slightly modify the original passages to make them stand-alone (decontextualization). Normally, this requires resolving the coreference anaphors or the bridging anaphors in the first sentence (see annotation guidelines). In Wikipedia, oftentimes the antecedents for these anaphors are the article titles themselves.
-
+<br><br>   
 For further information, see the annotation guidelines of the paper.
+    </span>
+</div>
+
+
 
 #### Who are the annotators?
 
 <!-- This section describes the people or systems who created the annotations. -->
 
+
+<div align="left">
+    <span style="font-size:16px;">
 Yufang Hou, Alessandra Pascale, Javier Carnerero-Cano, Tigran Tchrakian, Radu Marinescu, Elizabeth Daly, Inkit Padhi
+    </span>
+</div>
+
 
 #### Personal and Sensitive Information
 
 <!-- State whether the dataset contains data that might be considered personal, sensitive, or private (e.g., data that reveals addresses, uniquely identifiable names or aliases, racial or ethnic origins, sexual orientations, religious beliefs, political opinions, financial or health data, etc.). If efforts were made to anonymize the data, describe the anonymization process. -->
 
+
+<div align="left">
+    <span style="font-size:16px;">
 N/A.
+    </span>
+</div>
+
 
 ## Bias, Risks, and Limitations
 
 <!-- This section is meant to convey both technical and sociotechnical limitations. -->
 
+<div align="left">
+    <span style="font-size:16px;">
 Each annotation instance contains at least one question and two possible answers, but some instances may contain more than one question (and the corresponding two possible answers for each question). Some instances may not contain a value for **paragraphA_clean**, **tagDate**, and **tagReason**.
+    </span>
+</div>
+
 
 ### Recommendations
 
 <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
 
+<div align="left">
+    <span style="font-size:16px;">
 Our data is downloaded from Wikipedia. As such, the data is biased towards the original content and sources. Given that human data annotation involves some degree of subjectivity we created a comprehensive 17-page annotation guidelines document to clarify important cases during the annotation process.  The annotators were explicitly instructed not to take their personal feeling about the particular topic. Nevertheless, some degree of intrinsic subjectivity might have impacted the techniques picked up by the annotators during the annotation.
-
+   <br><br>  
 Since our dataset requires manual annotation, annotation noise is inevitably introduced. 
+    </span>
+</div>
+
+
 
 
 ## Citation
 
 <!-- If there is a paper or blog post introducing the dataset, the APA and Bibtex information for that should go in this section. -->
 
+
+<div align="left">
+    <span style="font-size:16px;">
 If this dataset is utilized in your research, kindly cite the following paper:
+    </span>
+</div>
+
 
 **BibTeX:**
 
@@ -256,7 +350,12 @@ If this dataset is utilized in your research, kindly cite the following paper:
 
 **APA:**
 
+<div align="left">
+    <span style="font-size:16px;">
 Hou, Y., Pascale, A., Carnerero-Cano, J., Tchrakian, T., Marinescu, R., Daly, E., Padhi, I., & Sattigeri, P. (2024). WikiContradict: A Benchmark for Evaluating LLMs on Real-World Knowledge Conflicts from Wikipedia. *arXiv preprint arXiv:2406.13805*.
+    </span>
+</div>
+
 
 <!-- ## Glossary [optional] -->
 
@@ -270,8 +369,17 @@ Hou, Y., Pascale, A., Carnerero-Cano, J., Tchrakian, T., Marinescu, R., Daly, E.
 
 ## Dataset Card Authors
 
+<div align="left">
+    <span style="font-size:16px;">
 Yufang Hou, Alessandra Pascale, Javier Carnerero-Cano, Tigran Tchrakian, Radu Marinescu, Elizabeth Daly, Inkit Padhi, and Prasanna Sattigeri.
+    </span>
+</div>
+
 
 ## Dataset Card Contact
 
-Yufang Hou (yhou@ie.ibm.com), Alessandra Pascale (apascale@ie.ibm.com), Javier Carnerero-Cano (javier.cano@ibm.com), Tigran Tchrakian (tigran@ie.ibm.com), Radu Marinescu (radu.marinescu@ie.ibm.com), Elizabeth Daly (elizabeth.daly@ie.ibm.com), Inkit Padhi (inkpad@ibm.com), and Prasanna Sattigeri (psattig@us.ibm.com).
+<div align="left">
+    <span style="font-size:16px;">
+Yufang Hou (yhou@ie.ibm.com), Alessandra Pascale (apascale@ie.ibm.com), Javier Carnerero-Cano (javier.cano@ibm.com), Tigran Tchrakian (tigran@ie.ibm.com), Radu Marinescu (radu.marinescu@ie.ibm.com), Elizabeth Daly (elizabeth.daly@ie.ibm.com), Inkit Padhi (inkpad@ibm.com), and Prasanna Sattigeri (psattig@us.ibm.com).    </span>
+</div>
+