Source author record

Haojun Li

Haojun Li appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Computation and Language Artificial Intelligence Machine Learning math-ph math.MP math.NA physics.comp-ph physics.optics

Catalog footprint

What is connected

4works

8topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2023arXiv

Neural Generation Meets Real People: Building a Social, Informative Open-Domain Dialogue Agent

We present Chirpy Cardinal, an open-domain social chatbot. Aiming to be both informative and conversational, our bot chats with users in an authentic, emotionally intelligent way. By integrating controlled neural generation with scaffolded, hand-written dialogue, we let both the user and bot take turns driving the conversation, producing an engaging and socially fluent experience. Deployed in the fourth iteration of the Alexa Prize Socialbot Grand Challenge, Chirpy Cardinal handled thousands of conversations per day, placing second out of nine bots with an average user rating of 3.58/5.

preprint2020arXiv

Neural Generation Meets Real People: Towards Emotionally Engaging Mixed-Initiative Conversations

We present Chirpy Cardinal, an open-domain dialogue agent, as a research platform for the 2019 Alexa Prize competition. Building an open-domain socialbot that talks to real people is challenging - such a system must meet multiple user expectations such as broad world knowledge, conversational style, and emotional connection. Our socialbot engages users on their terms - prioritizing their interests, feelings and autonomy. As a result, our socialbot provides a responsive, personalized user experience, capable of talking knowledgeably about a wide variety of topics, as well as chatting empathetically about ordinary life. Neural generation plays a key role in achieving these goals, providing the backbone for our conversational and emotional tone. At the end of the competition, Chirpy Cardinal progressed to the finals with an average rating of 3.6/5.0, a median conversation duration of 2 minutes 16 seconds, and a 90th percentile duration of over 12 minutes.

preprint2015arXiv

A Gaussian Particle Filter Approach for Sensors to Track Multiple Moving Targets

In a variety of problems, the number and state of multiple moving targets are unknown and are subject to be inferred from their measurements obtained by a sensor with limited sensing ability. This type of problems is raised in a variety of applications, including monitoring of endangered species, cleaning, and surveillance. Particle filters are widely used to estimate target state from its prior information and its measurements that recently become available, especially for the cases when the measurement model and the prior distribution of state of interest are non-Gaussian. However, the problem of estimating number of total targets and their state becomes intractable when the number of total targets and the measurement-target association are unknown. This paper presents a novel Gaussian particle filter technique that combines Kalman filter and particle filter for estimating the number and state of total targets based on the measurement obtained online. The estimation is represented by a set of weighted particles, different from classical particle filter, where each particle is a Gaussian distribution instead of a point mass.

preprint2012arXiv

Adaptive Wavelet Collocation Method for Simulation of Time Dependent Maxwell's Equations

This paper investigates an adaptive wavelet collocation time domain method for the numerical solution of Maxwell's equations. In this method a computational grid is dynamically adapted at each time step by using the wavelet decomposition of the field at that time instant. In the regions where the fields are highly localized, the method assigns more grid points; and in the regions where the fields are sparse, there will be less grid points. On the adapted grid, update schemes with high spatial order and explicit time stepping are formulated. The method has high compression rate, which substantially reduces the computational cost allowing efficient use of computational resources. This adaptive wavelet collocation method is especially suitable for simulation of guided-wave optical devices.