Jiachen Li 李家琛
Google scholar           LinkedIn           X(Twitter)   

Keywords: Health, HCI, UbiComp, AI/LLM/ML, Aging Populations, Conversational Agent ...        

NEWs

Oct 2023  
Present my first CSCW paper at Minneapolis!

Aug 2022  
In Boston! Starting my CS PhD with Dr, Beth Mynatt and Dr. Varun Mishra.

Apr 2021  
Join Everyday Computing Lab advised by Dr. Beth Mynatt.

Jan 2021  
Start working on research using EMG signal to enhance yoga learning with Elaine.

Nov 2020  
Join Ubicomp & COSMOS lab, working on research about computaional materials and robotic actuators with Tingyu.

Jul 20  
Finally in Atlanta!

FUN

See my Instagram @jasl_is_jasl

CONTACT

[email protected]

[email protected]

Download my CV

I am Jiachen Li, a current CS PhD student at Northeastern University. Prior to it, I have a Master's degree in Digital Media at Georgia Tech, and a Bachelor's Degree in Electronic Engineering.

My research centers on examining the role of AI assistants in people's everyday experiences, with a particular emphasis on understanding the gap between current technology and human expectations in healthcare settings. My primary focus has been on the contexts of daily routine management and healthcare & wellness for older adults (OAs) with the help of passive personal tracking and LLM.

Recent:

1) Sense-making for multi-modal personal health sensing data with visualization and LLM (CHI submissions)
2) Creating LLM agents that can simulate daily behaviors of OAs    
   -  Profile, schedule & routine, raw sensor data, social interactions
3) Providing support to OAs and their care networks with the help of LLM  
  -  Daily summaries, JIT interventions, Caregivers collaboration
4) Visual and voice AI assistant in assisting cooking tasks and routines for OAs with cognitive impairment.

I had my bachelor's degree in Electronics Information Technology and decided to pursue a master’s degree in Digital Media after graduation. Due to my previous experience, I highly value creativity and intuition in the design and research process more than many engineers, however, I still regard myself as a logical and rational person. As a result, I would like to apply scientific and epistemological methods as the main part of my future research and combine them with design synthesis and abductive reasoning to generate valuable insights.

To know more about me, I'm a doll maker, a figure maker and collector. I work with clay, mostly modeling clay and polymer clay. I also do figure coating, moldering and casting. In my spare time, I travel, read and photograph. I love cats, have been living with six cats before (four of them are my cats: Leo, Lily, Jacob and Echo).

Experience

2022 - present
Northeastern University       Ph.D. in Computer Science
Graduate Researcher - AI-Caring Institute
Beth Mynatt (ECL), Varun Mishra (Ubiwell)
2023
Accenture, Digital Experiences       intern
Summer research specialist
Manaswi Saha, Jordan Ackerman, Michael Kuniavsky 
2020 - 2022
Georgia Institute of Technology       M.S. in Digital Media
Graduate Researcher
Beth Mynatt (ECL, Institute for People and Technology)
Graduate Researcher
Gregory Abowd, Hyun Joo Oh (UbiComp COSMOS Lab)
2018
University of California, Berkeley      Concurrent Student
2015 - 2019
Beijing University of Posts and Telecommunications
B. Eng in Electronic Information Science and Technology

Recent Research

Activity Annotation Tool for Multi-Modal Health and Sensing Data through LLM

Data collection + Demo + User testing

(In progress)

Simulated LLM Agents as Older Adults using Multi-Modal Health and Sensing Data

Data collection + Agent simulation

(In progress)

Intervention through agent

(In progress)

Cancer Patient-Provider Telecommunication through LLM-powered Voice Assistant

Interview + Qualitative Analysis:

“I Wish There Were an AI”: Challenges and AI Potential in Cancer Patient-Provider Communication

(In submission)

System Design:

(In submission)

Deployment:

(In progress)

LLM + Conversational Agent in Cooking Scenarios

Exploratory User Study:

"Mango Mango, How to Let The Lettuce Dry Without A Spinner?'': Exploring User Perceptions of Using An LLM-Based Conversational Assistant Toward Cooking Partner

(In submission)

Comprehensive analysis (video coding):

(In submission)

Visual Cooking Partner for Older Adults through Digital Sticky Board

User study + Demo + Testing

To be added

(In submission)

Deployment

(In progress)

Smart Home Sensors Data Visualization and Sharing for Older Adults

User study + Demo + Testing:

Privacy vs Awareness: learning the tradeoffs between seniors and their children when sharing indoor localization data

(CSCW 2023)

Deployment:

(In progress)

Medication Management for Older Adults using Conversational Agent

User study + Demo + Deployment:

A Collaborative Approach to Support Medication Management in Older Adults with Mild Cognitive Impairment Using Conversational Assistants (CAs)

(ASSETS 2022 *Best Paper Award*)

Gen AI vs. Q&A Platform for Developer Support

Qualitative content analysis:

"Always Nice and Confident, Sometimes wrong": Developer's Experiences Engaging Generative AI Chatbots Versus Human-Powered Q&A Platforms

(In submission)

Detect and Close Knowledge Gap in Online Meeting through LLM

Demo

knowGap: An LLM based System for Data-driven Collaborative Support to Detect and Close Knowledge Gaps

(In submission)

Remote Communication through Computational Materials

Phase Transition Actuator

PITAS: Sensing and Actuating Embedded Robotic Sheet for Physical Information Communication

(CHI 2022)

Activity Annotation Tool for Multi-Modal Health and Sensing Data through LLM

Data collection + Demo + User testing

(In progress)

Simulated LLM Agents as Older Adults using Multi-Modal Health and Sensing Data

Data collection + Agent simulation

(In progress)

Intervention through agent

(In progress)

Cancer Patient-Provider Telecommunication through LLM-powered Voice Assistant

Interview + Qualitative Analysis:

“I Wish There Were an AI”: Challenges and AI Potential in Cancer Patient-Provider Communication

(In submission)

System Design:

(In submission)

Deployment:

(In progress)

Visual Cooking Partner for Older Adults through Digital Sticky Board

User study + Demo + Testing

To be added

(In submission)

Deployment

(In progress)

Smart Home Sensors Data Visualization and Sharing for Older Adults

User study + Demo + Testing:

Privacy vs Awareness: learning the tradeoffs between seniors and their children when sharing indoor localization data

(CSCW 2023)

Deployment:

(In progress)

Medication Management for Older Adults using Conversational Agent

User study + Demo + Deployment:

A Collaborative Approach to Support Medication Management in Older Adults with Mild Cognitive Impairment Using Conversational Assistants (CAs)

(ASSETS 2022 *Best Paper Award*)

Activity Annotation Tool for Multi-Modal Health and Sensing Data through LLM

Data collection + Demo + User testing

(In progress)

Simulated LLM Agents as Older Adults using Multi-Modal Health and Sensing Data

Data collection + Agent simulation

(In progress)

Intervention through agent

(In progress)

Cancer Patient-Provider Telecommunication through LLM-powered Voice Assistant

Interview + Qualitative Analysis:

“I Wish There Were an AI”: Challenges and AI Potential in Cancer Patient-Provider Communication

(In submission)

System Design:

(In submission)

Deployment:

(In progress)

LLM + Conversational Agent in Cooking Scenarios

Exploratory User Study:

"Mango Mango, How to Let The Lettuce Dry Without A Spinner?'': Exploring User Perceptions of Using An LLM-Based Conversational Assistant Toward Cooking Partner

(In submission)

Comprehensive analysis (video coding):

(In submission)

Gen AI vs. Q&A Platform for Developer Support

Qualitative content analysis:

"Always Nice and Confident, Sometimes wrong": Developer's Experiences Engaging Generative AI Chatbots Versus Human-Powered Q&A Platforms

(In submission)

Detect and Close Knowledge Gap in Online Meeting through LLM

Demo

knowGap: An LLM based System for Data-driven Collaborative Support to Detect and Close Knowledge Gaps

(In submission)

Cancer Patient-Provider Telecommunication through LLM-powered Voice Assistant

Interview + Qualitative Analysis:

“I Wish There Were an AI”: Challenges and AI Potential in Cancer Patient-Provider Communication

(In submission)

System Design:

(In submission)

Deployment:

(In progress)

LLM + Conversational Agent in Cooking Scenarios

Exploratory User Study:

"Mango Mango, How to Let The Lettuce Dry Without A Spinner?'': Exploring User Perceptions of Using An LLM-Based Conversational Assistant Toward Cooking Partner

(In submission)

Comprehensive analysis (video coding):

(In submission)

Medication Management for Older Adults using Conversational Agent

User study + Demo + Deployment:

A Collaborative Approach to Support Medication Management in Older Adults with Mild Cognitive Impairment Using Conversational Assistants (CAs)

(ASSETS 2022 *Best Paper Award*)

Activity Annotation Tool for Multi-Modal Health and Sensing Data through LLM

Data collection + Demo + User testing

(In progress)

Simulated LLM Agents as Older Adults using Multi-Modal Health and Sensing Data

Data collection + Agent simulation

(In progress)

Intervention through agent

(In progress)

LLM + Conversational Agent in Cooking Scenarios

Exploratory User Study:

"Mango Mango, How to Let The Lettuce Dry Without A Spinner?'': Exploring User Perceptions of Using An LLM-Based Conversational Assistant Toward Cooking Partner

(In submission)

Comprehensive analysis (video coding):

(In submission)

Visual Cooking Partner for Older Adults through Digital Sticky Board

User study + Demo + Testing

To be added

(In submission)

Deployment

(In progress)

Smart Home Sensors Data Visualization and Sharing for Older Adults

User study + Demo + Testing:

Privacy vs Awareness: learning the tradeoffs between seniors and their children when sharing indoor localization data

(CSCW 2023)

Deployment:

(In progress)

Medication Management for Older Adults using Conversational Agent

User study + Demo + Deployment:

A Collaborative Approach to Support Medication Management in Older Adults with Mild Cognitive Impairment Using Conversational Assistants (CAs)

(ASSETS 2022 *Best Paper Award*)

Simulated LLM Agents as Older Adults using Multi-Modal Health and Sensing Data

Data collection + Agent simulation

(In progress)

Intervention through agent

(In progress)

Cancer Patient-Provider Telecommunication through LLM-powered Voice Assistant

Interview + Qualitative Analysis:

“I Wish There Were an AI”: Challenges and AI Potential in Cancer Patient-Provider Communication

(In submission)

System Design:

(In submission)

Deployment:

(In progress)

Smart Home Sensors Data Visualization and Sharing for Older Adults

User study + Demo + Testing:

Privacy vs Awareness: learning the tradeoffs between seniors and their children when sharing indoor localization data

(CSCW 2023)

Deployment:

(In progress)

Detect and Close Knowledge Gap in Online Meeting through LLM

Demo

knowGap: An LLM based System for Data-driven Collaborative Support to Detect and Close Knowledge Gaps

(In submission)

Remote Communication through Computational Materials

Phase Transition Actuator

PITAS: Sensing and Actuating Embedded Robotic Sheet for Physical Information Communication

(CHI 2022)

Previous Research

Let's play Chinese Abacus (2019)

Abacus, as a traditional Chinese physical calculator, has an irreplaceable position in pre-k and primary school mathematics education. We designed and developed a physical interaction system that enabled children to study the principle of mathematical calculation through the process of abacus learning.

Familyship Face Videos in the Wild (2018)

In this paper, we investigate the problem of video-based parent-child relationship prediction via human face analysis. Most of the existing kinship verification methods are based on single images; these approaches cannot effectively utilize videos of the face for kinship verification. We propose our own dataset, Familyship Fa...

Unintentional Touch Detection (2019)

Designed unintentional touch detection algorithm based on 125Hz capacitive sensing signal on phone.

Used flood-fill algorithm and ellipse fitting techniques to detect touch, and performed FFT analysis and generated the frequency spectrum.

Clay Archive Study (2021)    

Inspired by ancient cuneiform that uses imprints on the clay for documentation, we built a tangible user interface using clay boards, conductive paints and digital animation with narratives to distance the physical from the digital.

Using EMG signal to enhance Yoga learning (2021)

We designed a tangible system that used biological signals (EMG, heart rate, etc.) to increase body awareness in yoga practice and created an interactive figurative animation as representation and visualization of bodily data.

Projects

Link to Figma Prototype

Empatea Box (2021)

We aims to create a solution to enhance empathy between people with MCI (Mild Cognitive Impairment) and those around them, and we provide a service that promotes activities between the MCI patients and caregivers using subscription boxes called the “Empatea Box”, which will be regularly delivered to the home address of the MCI patient's family and involve loose lea...

Link to Figma Prototype

CODE-Crafters Website Design (2021)

The project aims to study how to design a research project website that enables different users to find a variety of information and accomplish different tasks. We designed a project website for Code-Crafter, a research project that investigates connections between quilting and computational thinking.

Clay Photo Lab (2021)

"Whenever clay encounters with other objects, there would be traces and imprints left on the surface.”

Have you ever thought about taking a picture using clay?

https://digital-sidewalk.herokuapp.com     Link to Figma Prototype

Digital Sidewalk (2020)

Inspired by the way students communicate their thoughts with chalk on sidewalk tiles on campus, we hope to create a digital sidewalk in the form of an interactive web page, where a student can share their personal stories during COVID(and beyond) on a sidewalk tile and have an opportunity to empathize, resonate, and reconnect with each other. We aim to provide people a way of communication and connection through the format of digital craft.

https://www.jiachen-li.com/newscity2

Newscity (2020)

Newscity is a word processor that retrieves the text entered by the user and outputs the same content in different sizes and colors according to the data from NY Times API, including the number of related articles, categories and other data. By continuously inputting different vocabulary, users can work with NY Times to build a unique 'news city' that belongs to them but also influenced by external information.

Empathy Helmet (2020)

A device was created to allow people to experience an overwhelming negative interaction of another person using a helmet with vibration motors, a resistive heater, LEDs, and a servo motor. Users can have a better und...

https://youtu.be/_XLRpbq6DNo

Cat Mirror (2020)

People spent a long time in front of the mirror but never noticed it. I hope to make people aware of the existence of the mirror, and furthermore, not only regard it as an object but more as a creature with life and soul. Cat Mirror is a unique mirror that can perform differently ac...

https://itch.io/embed-upload/3824458?color=333333

EduFirst (2020)

EduFirst is a game that simulates an education-oriented society. At the beginning of the game, the player is told that the goal of judging whether the game is successful is whether your child can enter a good university, and this is closely related to his various points on “Knowledge”, “Patience” and “Concentration”. In this game, players consume their “Ability” points to ask their children to co...

TextEx (2018)

Every semester, college students need to buy textbooks for their classes. At the end of the semester, they often no longer need the books. TextEx is an online platform for UC Berkeley students to buy and sell their used textbooks and engage in textbook exchange with other students.

Stress App Design (2018)

Designed UI and wearable for an app that detects the stress level of people and helps them release stress. Used different features such as layout, weight, typography and color to distinguish two stress levels.

Link to Figma Prototype

Empatea Box (2021)

We aims to create a solution to enhance empathy between people with MCI (Mild Cognitive Impairment) and those around them, and we provide a service that promotes activities between the MCI patients and caregivers using subscription boxes called the “Empatea Box”, which will be regularly delivered to the home address of the MCI patient's family and involve loose lea...

Link to Figma Prototype

CODE-Crafters Website Design (2021)

The project aims to study how to design a research project website that enables different users to find a variety of information and accomplish different tasks. We designed a project website for Code-Crafter, a research project that investigates connections between quilting and computational thinking.

TextEx (2018)

Every semester, college students need to buy textbooks for their classes. At the end of the semester, they often no longer need the books. TextEx is an online platform for UC Berkeley students to buy and sell their used textbooks and engage in textbook exchange with other students.

Stress App Design (2018)

Designed UI and wearable for an app that detects the stress level of people and helps them release stress. Used different features such as layout, weight, typography and color to distinguish two stress levels.

Clay Photo Lab (2021)

"Whenever clay encounters with other objects, there would be traces and imprints left on the surface.”

Have you ever thought about taking a picture using clay?

http://www.jiachen-li.com/newscity/

Newscity (2020)

Newscity is a word processor that retrieves the text entered by the user and outputs the same content in different sizes and colors according to the data from NY Times API, including the number of related articles, categories and other data. By continuously inputting different vocabulary, users can work with NY Times to build a unique 'news city' that belongs to them but also influenced by external information.

Empathy Helmet (2020)

A device was created to allow people to experience an overwhelming negative interaction of another person using a helmet with vibration motors, a resistive heater, LEDs, and a servo motor. Users can have a better und...

https://youtu.be/_XLRpbq6DNo

Cat Mirror (2020)

People spent a long time in front of the mirror but never noticed it. I hope to make people aware of the existence of the mirror, and furthermore, not only regard it as an object but more as a creature with life and soul. Cat Mirror is a unique mirror that can perform differently ac...

https://digital-sidewalk.herokuapp.com     Link to Figma Prototype

Digital Sidewalk (2020)

Inspired by the way students communicate their thoughts with chalk on sidewalk tiles on campus, we hope to create a digital sidewalk in the form of an interactive web page, where a student can share their personal stories during COVID(and beyond) on a sidewalk tile and have an opportunity to empathize, resonate, and reconnect with each other. We aim to provide people a way of communication and connection through the format of digital craft.

https://www.jiachen-li.com/newscity2

Newscity (2020)

Newscity is a word processor that retrieves the text entered by the user and outputs the same content in different sizes and colors according to the data from NY Times API, including the number of related articles, categories and other data. By continuously inputting different vocabulary, users can work with NY Times to build a unique 'news city' that belongs to them but also influenced by external information.

https://itch.io/embed-upload/3824458?color=333333

EduFirst (2020)

EduFirst is a game that simulates an education-oriented society. At the beginning of the game, the player is told that the goal of judging whether the game is successful is whether your child can enter a good university, and this is closely related to his various points on “Knowledge”, “Patience” and “Concentration”. In this game, players consume their “Ability” points to ask their children to co...

Publications

Google scholar
*****in submission*****
[1] Jiachen Li, Justin Steinberg, Elizabeth Mynatt, Varun Mishra, Navigating the Paradox: Challenges and Strategies of University Students Managing Mental Health Medication in Real-World Practices, (arXiv ’24) (submitted)

[2] Z Yang, X Xu, B Yao, Jiachen Li, J Bagdasarian, G Gao, D Wang, “I Wish There Were an AI”: Challenges and AI Potential in Cancer Patient-Provider Communication, (arXiv ’24) (submitted)

[3] Szeyi Chan*, Jiachen Li*, Bingsheng Yao, Amama Mahmood, Chien-Ming Huang, Holly Jimison, Elizabeth D Mynatt, Dakuo Wang "Mango Mango, How to Let The Lettuce Dry Without A Spinner?'': Exploring User Perceptions of Using An LLM-Based Conversational Assistant Toward Cooking Partner, (arXiv ’24) (submitted)

[4] Szeyi Chan, Shihan Fu, Jiachen Li, Bingsheng Yao, Smit Desai, Mirjana Prpa, Dakuo Wang, Human and LLM-Based Voice Assistant Interaction: An Analytical Framework for User Verbal and Nonverbal Behaviors, (arXiv ’24) (submitted)

[5] Manaswi Saha, Jiachen Li, Jordan Ackerman, Michael Kuniavsky, knowGap: An LLM based System for Data-driven Collaborative Support to Detect and Close Knowledge Gaps (submitted)

*****accepted*****
[1] Jiachen Li, Justin Steinberg, Xiwen Li, Bingsheng Yao, Dakuo Wang, Elizabeth Mynatt, Varun Mishra, Understanding the Daily Lives of Older Adults: Integrating Multi-modal Personal Health Tracking Data through Visualization and Large Language Models (AAAI AiP symposium' 24) (1 related CHI'25 submission)

[2] Jiachen Li, Varun Mishra, Elizabeth Mynatt, Jonathan Bell, "Always Nice and Confident, Sometimes wrong": Developer's Experiences Engaging Generative AI Chatbots Versus Human-Powered Q&A Platforms, (CSCW ’24)

[3] JiachenLi, Bingrui Zong, Tingyu Cheng, Yunzhi Li, Elizabeth D Mynatt, Ashutosh Dhekne, “Privacy vs. Awareness: Relieving the Tension between Older Adults andAdult Children When Sharing In-home Activity Data”, (CSCW ’23)

[4] Bingsheng Yao, Guiming Chen, Ruishi Zou, Yuxuan Lu, Jiachen Li, Shao Zhang, Sijia Liu, James Hendler, Dakuo Wang, More Samples or More Prompts? Exploring Effective Few-Shot In-Context Learning for LLMs with In-Context Sampling, (NAACL ’24)

[5] Niharika Mathur, Kunal Dhodapkar, Tamara Zubatiy, Jiachen Li, Brian Jones, and Elizabeth Mynatt. 2022. A Collaborative Approach to Support Medication Management in Older Adults with Mild Cognitive Impairment Using Conversational Assistants (CAs). In Proceedings of the 24th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS '22). Association for Computing Machinery, New York, NY, USA, Article 42, 1–14. https://doi.org/10.1145/3517428.3544830 (Best Paper Award)

[6] Tingyu Cheng, Jung Wook Park*, Jiachen Li*, Charles Ramey, Hongnan Lin, Gregory D. Abowd, Carolina Brum Medeiros, HyunJoo Oh, and Marcello Giordano. 2022. PITAS: Sensing and Actuating Embedded Robotic Sheet for Physical Information Communication. In Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems (CHI '22). Association for Computing Machinery, New York, NY, USA, Article 172, 1–16. https://doi.org/10.1145/3491102.3517532. (Honorable mention in Fast Company’s 2022 Innovation by DesignAwards, Student notable interaction award in 2022 Core77 Design Award)

[7] Ying Sun*, Jiachen Li*, Yiwen Wei, Haibin Yan. Video-based Parent-Child Relationship Prediction, (IEEE VCIP' 18)
(*: contributed equally; submitted: submitted to other peer-reviewed top conferences; anonymous: some papers in review process that were not on arXiv)

Gallery