[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121614-en":3,"doc-seo-121614-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},121614,1099513958607,"Jiven","https://ap-avatar.wpscdn.com/avatar/100002390cf8733938c?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778829742770036399",8,"Research & Report","TalkToModel - Explaining Machine Learning Models with Interactive Natural Language Conversations - Research","Machine Learning (ML) models increasingly drive high-stakes decisions, yet their growing complexity makes them difficult to interpret. The document introduces TalkToModel, an interactive dialogue system that explains ML model predictions through natural language conversations. It combines a conversational interface, a dialogue engine that maps user utterances to suitable explanations for tabular models and datasets, and an execution component that generates the explanation outputs. Extensive quantitative and human evaluations show strong accuracy on novel models and datasets, and real-world human studies indicate high willingness to use it over baseline tools, especially for healthcare explainability tasks.","TalkToModel: Explaining Machine Learning Models with Interactive Natural Language Conversations  \nDylan Slack UC Irvine [dslack@uci. edu](dslack@uci. edu)  \nSatyapriya Krishna Harvard University[skrishna@g. harvard. edu](skrishna@g. harvard. edu)  \nHimabindu Lakkaraju∗  \nHarvard University [hlakkaraju@hbs. edu](hlakkaraju@hbs. edu)  \narXiv :2207 .04154v3 [ cs .LG] 20 Sep 2022  \nSameer Singh∗  \nUC Irvine / AI2  \n[sameer@uci. edu](sameer@uci. edu)  \nAbstract  \nMachine Learning (ML) models are increasingly used to make critical decisions in real-world applications, yet they have become more complex, making them harder to understand. To this end, researchers have proposed several techniques to explain model predictions. However, practitioners struggle to use these explainability techniques because they often do not know which one to choose and how to interpret the results of the explanations. In this work, we address these challenges by introducing TalkToModel: an interactive dialogue system for explaining machine learning models through conversations. Specifically, TalkToModel comprises of three key components: 1) a natural language interface for engaging in conversations, making ML model explainability highly accessible, 2) a dialogue engine that adapts to any tabular model and dataset, interprets natural language, maps it to appropriate explanations, and generates text responses, and 3) an execution component that constructs the explanations. We carried out extensive quantitative and human subject evaluations of TalkToModel. Overall, we found the conversational system understands user inputs on novel datasets and models with high accuracy, demonstrating the system's capacity to generalize to new situations. In real-world evaluations with humans, 73% of healthcare workers (e.g., doctors and nurses) agreed they would use TalkToModel over baseline point-and-click systems for explainability in a disease prediction task, and 85% of ML professionals agreed TalkToModel was easier to use for computing explanations. Our findings demonstrate that TalkToModel is more effective for model explainability than existing systems, introducing a new category of explainability tools for practitioners.1  \nMachine learning (ML) models are being deployed to make consequential decisions in several critical domains such as healthcare, finance, and law, due to their strong predictive performance. However, state-of-the-art ML models, such as deep neural networks, have also become more complex and, therefore, hard to understand. This dynamic poses challenges in real-world applications for the model stakeholders who need to understand why models make predictions and whether to trust the predictions. Consequently, there are an increasing number of techniques that explain the predictions of ML models. Such explanation techniques help users understand predictions of ML models in terms of why they occur and the interventions needed to get different outcomes, thus helping establish when to trust model predictions. However, recent work suggests practitioners often have difficulty interpreting the results of the explanations and determining which ones to run [35 , 31] . Further, model understanding  \n∗ Equal Advising  \n1 Code & demo released here: [https://github.com/dylan-slack/TalkToModel](https://github.com/dylan-slack/TalkToModel)  \n| filter applicant 358 feature importance previous filter\u003Cbr>counterfactual explanation |  |\n| --- | --- |\n\nTalkToModel parses inputs  \nto executable form 2  \n|  |\n| --- |\n| \u003Cbr> |\n\nFigure 1: An overview of TalkToModel: Instead of writing code or using a dashboard, users engage in dynamic open ended dialogues with TalkToModel to understand models. First (1) , users supply natural language inputs using the interface. Next (2), the dialogue engine parses the input into an executable representation of the question. After, (3), the execution engine runs the operations. Finally, the dialogue engine formats the results into a response and prov","cbCaicrskY3j1TNh","https://ap.wps.com/l/cbCaicrskY3j1TNh","pdf",2320722,1,34,"English","en",105,"# Abstract\n## Explainability challenges and practitioner needs\n## TalkToModel system design and components\n## Dialogue workflow and explanation execution\n## Evaluation results and adoption feedback","[{\"question\":\"What problem does TalkToModel address in machine learning explainability?\",\"answer\":\"TalkToModel targets the gap between available explainability techniques and practitioners’ difficulty choosing and interpreting them, especially when models are complex and stakeholders need understandable reasons for predictions.\"},{\"question\":\"How does TalkToModel generate explanations during a conversation?\",\"answer\":\"It uses a dialogue engine to parse natural-language questions into executable representations, an execution component to run the required explanation operations, and then formats the results into a user-facing response.\"},{\"question\":\"What evidence does the document provide about TalkToModel’s effectiveness?\",\"answer\":\"The document reports extensive quantitative and human-subject evaluations, including real-world feedback where a majority of healthcare workers and ML professionals indicated they would use TalkToModel and found it easier for computing explanations.\"}]","TalkToModel - Explaining Machine Learning Models with Interactive Natural Language Conversations - Research | PDF",1785805664,86,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"talktomodel-explaining-machine-learning-models-with-interactive-natural-language-conversations-research","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/talktomodel-explaining-machine-learning-models-with-interactive-natural-language-conversations-research/121614/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does TalkToModel address in machine learning explainability?","Question",{"text":75,"@type":76},"TalkToModel targets the gap between available explainability techniques and practitioners’ difficulty choosing and interpreting them, especially when models are complex and stakeholders need understandable reasons for predictions.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does TalkToModel generate explanations during a conversation?",{"text":80,"@type":76},"It uses a dialogue engine to parse natural-language questions into executable representations, an execution component to run the required explanation operations, and then formats the results into a user-facing response.",{"name":82,"@type":73,"acceptedAnswer":83},"What evidence does the document provide about TalkToModel’s effectiveness?",{"text":84,"@type":76},"The document reports extensive quantitative and human-subject evaluations, including real-world feedback where a majority of healthcare workers and ML professionals indicated they would use TalkToModel and found it easier for computing explanations.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]