Home  |  Activities  |  Awards  |  Funding  |  Members  |  NewsLog  |  Patents  |  Projects  |  Publications |
|
|
We present ElectionBot-SC, a tool designed to provide reliable and accessible election information, featuring three response engines: SafeChat (explained below), Google Search, and a free-tier Large Language Model (LLM): Mixtral 8x7b.
[Paper] [Demo Video]This website organizes election-related questions by country, highlighting diverse geopolitical contexts and allowing users to submit questions specific to their nation.
[Webpage]We analyze traffic accidents over time and across various regions, focusing on economic impact, road conditions, and incident hotspots. The insights provided help in understanding patterns and mitigating future risks.
[Demo] [More Details]In this demonstration, we introduce the first PDDL formulation for the 3-dimension RC and solve it with an off-the-shelf Fast-Downward planner. We also create a plan executor and visualizer to show how the plan achieves the intended goal.
[Paper] [More Details]We present a demonstration of integrating Automated Planning with opinion dynamics, enabling users to visualize, simulate, and strategically influence opinion evolution in networks. This approach offers a practical solution for understanding and guiding information spread.
[Paper] [Demo Video]As Artificial Intelligence (AI) systems are becoming powerful, there is a growing concern about these data-driven systems with respect to trust issues like exhibiting biasing which may adversely impact their large-scale adoption. Here bias may be with respect to protected attributes such as well-studied gender, race, and age. In this paper, we introduce ARC, a tool to rate AI systems for bias through a causal lens. The main objective of the tool is to assist developers in building better models and aid end-users in making informed decisions based on the available data. The tool is extensible and currently supports three different AI tasks: binary classification ,sentiment analysis, and group recommendation. It gives users the option of choosing data for a task and rating AI systems for bias with respect to different protected attributes present in the data. The rating method is system-independent and the ratings given by the algorithm are causally interpretable .These ratings help the user make an informed decision based on the data in hand.
[Demo Video] [Tool]We introduce ULTRA, a novel AI-based system for assisting team formation when researchers respond to calls for proposals from funding agencies.
[Paper] [Demo Video] [Tool] [BibTex] [More Details]We introduce Plansformer; an LLM fine-tuned on planning problems and capable of generating plans with favorable behavior in terms of correctness and length with reduced knowledge-engineering efforts.
[Paper] [BibTex] [More Details]Here, we demonstrate how a chatbot built with the SafeChat architecture and using official FAQs (questions and their answers) from South Carolina will answer to any user question. The chatbot should only answer when it is confident of the question. The chatbot can make answers consumable but will not cook up (hallucinate) new answers.
[Paper] [Tool] [BibTex]Here, we demonstrate how a chatbot built with the SafeChat architecture and using official FAQs (questions and their answers) from Mississippi will answer to any user question. The chatbot should only answer when it is confident of the question. The chatbot can make answers consumable but will not cook up (hallucinate) new answers.
[Paper] [Tool] [BibTex]We demonstrate ALLURE, an educational AI system for learning to solve the Rubik’s cube that is designed to help students improve their problem solving skills.
[Paper] [Demo Video] [BibTex] [More Details]We present an unsupervised system, called KITE, for exploring textual data which can generate insights from a general as well as domaindependent perspective consisting of holistic views, entity-centric view, events view, domain-specific interpretation using industry taxonomies and a detailed full-text view transparently connecting the document to insight elements.
[Paper] [Demo Video] [Tool] [BibTex]Sentiment Analysis Systems (SASs) are data-driven Artificial Intelligence (AI) systems that assign a score conveying the sentiment and emotion intensity when a piece of text is given as input. Like other AI, and especially machine learning (ML) based systems, they have also exhibited instability in their values when inputs are perturbed with respect to gender and race, which can be interpreted as biased behavior. In this demonstration paper, we present ROSE, a resource for understanding the behavior of SAS systems with respect to gender. It consists of data consisting of input text and output sentiment scores and a visualization tool to explore the behavior of SAS. We calculated the output sentiment scores using off-the-shelf SASs and our deep-learning-based implementations based on published architectures.
[Paper] [Demo Video] [Tool] [BibTex]We present a generic approach for dialogs for information retrieval based on automated planning within a reinforcement learning (RL)-based platform, ParlAI.
[Paper] [Github] [BibTex] [More Details]