Research

UIST 2026
Steerable Chatbots: Exploring Personalization Control Interfaces via LLM Activation Steering
Jessica Bo, Tianyu Xu, Ishan Chatterjee, Katrina Passarella-Ward, Achin Kulshrestha, D Shin

Leveraging activation steering to allow users to directly manipulate the personalization of their chatbot.

[paper]
EMNLP 2026 findings
Emulate or Estimate? The Divergent Strengths of Base and Post-Trained Language Models for Opinion Simulation
Seth Grief-Albert, Jessica Bo, Difan Jiao, Ashton Anderson

Base models are the stronger emulators, producing response distributions closer to human ground truth. Post-trained models are generally the stronger estimators, producing more accurate distributional predictions.

[paper]
AAAI AIES 2026
Studying People to Study AI: Expert Perspectives on the Epistemic Fit and Barriers of Human Research in AI Safety & Ethics
Jessica Bo, Paula Akemi Aoyagui, Shalaleh Rismani, Dipto Das, Syed Ishtiaque Ahmed, Ashton Anderson

Expert survey + interview to understand how AI Safety & Ethics experts conceptualize the value of human research, and what infrastructural and resource barriers they've faced in human research.

[paper]
FAccT 2026
Large Language Lovers: Lived Experiences of Negotiating Agency and Platform Control in AI Companionship
Patrick Yung Kang Lee, Jessica Bo, Zixin Zhao, Paula Akemi Aoyagui, Matthew Varona, Ashton Anderson, Anastasia Kuziminykh, Fanny Chevalier, Carolina Nobre

Triangulation study on how individuals with general purpose AI (e.g. chatGPT) companions form and maintain their relationship.

[paper] [project page]
CHI 2026 (honourable mention)
Invisible Saboteurs: Sycophantic LLMs Mislead Novices in Problem-Solving Tasks
Jessica Bo, Majeed Kazemitabaar, Mengqing Deng, Michael Inzlicht, Ashton Anderson

Investigation into how AI sycophancy causes task novices to reinforce their misconceptions, while being unnoticed by users.

[paper]
IASEAI 2026
Language Models Exhibit Inconsistent Biases Towards Algorithmic Agents and Human Experts
Jessica Bo*, Lillio Mok*, Ashton Anderson

Evaluating how LLMs express and reveal trust towards humanvs and towards other LLMs.

[paper]
CHI 2025 (honourable mention)
To Rely or Not to Rely? Evaluating Interventions for Appropriate Reliance on Large Language Models
Jessica Bo, Sophia Wan, Ashton Anderson

Benchmarking three different LLM reliance interventions on improving appropriate reliance.

[paper]
AAAI HCOMP 2024
Disclosures & Disclaimers: Investigating the Impact of Transparency Disclosures and Reliability Disclaimers on Learner-LLM Interactions
Jessica Bo*, Harsh Kumar*, Michael Liut, Ashton Anderson

Classroom experiment on how LLM transparency disclosures and reliance disclaimers affect students behaviours and perceptions.

[paper]
CHI 2024
Incremental XAI: Memorable Understanding of AI with Incremental Explanations
Bo J, Hao P, Lim B

Improving memorability and understanding of AI via incremental explanations.

[paper]
NeurIPS 2022 workshop (TS4H)
Pretraining ECG Data with Adversarial Masking Improves Model Generalizability for Data-Scarce Tasks
Jessica Bo, Hen-Wei Huang, Alvin Chan, Giovanni Traverso

Adversarial masking for ECG to improve model generalizability.

[paper] [code]
SIGGRAPH 2022 (best paper)
CLIPasso: Semantically-Aware Object Sketching
Yael Vinker, Ehsan Pajouheshgar, Jessica Bo, Roman Bachmann, Amit Haim Bermano, Daniel Cohen-Or, Amir Zamir, Ariel Shamir

Semantically guided sketch generation with a CLIP-based loss.

[paper] [project page]
IEEE EMBC 2022
ATTENTIV: Instrumented Peripheral Catheter for the Detection of Catheter Dislodgement in IV Infiltration
Jessica Bo, Kevin Ta, Rio Nishida, Gordon Yeh, Vivian Tsang, Megan Bolton, Manon Ranger, Konrad Walus

Sensor-embedded catheter for detecting IV failures in real-time.

[paper] [website]
made by © Jessica Bo 2025