Gaole He

Postdoc at NUS Ubicomp Lab, PhD at TU Delft @wisdelft, Human-centered AI.

[HCOMP 2026 @hcomp_conf]

🚨 Submit to the HCOMP 2026 Posters & Demos track!

Have late-breaking results, work in progress, an early proof-of-concept, or an interactive system ready to share?

Bring it to HCOMP and start a conversation with the community.

📅 Deadline: July 16

➡️ https://t.co/fIBzcNtlhz

[HCOMP 2026 @hcomp_conf]

🚨 The #HCOMP2026 submission site is OPEN!

Abstracts for papers are due June 1 (AoE). Full papers are due June 8.

Deadline is coming — get your abstract in soon!

🔗https://t.co/1lg1awjaqR

[Gaole He @HeGaole]

HCOMP is now an ACM conference focused on research at the intersection of human-AI complementarity and alignment.

Welcome submissions regarding human-AI complementarity, human-centered alignment and human contributions to AI system.

Please RT and share! 📢

[HCOMP 2026 @hcomp_conf]

📣 The CFP for #HCOMP2026 is now live!

HCOMP 2026, the ACM Conference on Human-AI Complementarity and Alignment, will be co-located with @ci_acm this year 🤝

📌 Abstract deadline: June 1, 2026
🔥 Full paper deadline: June 8, 2026

Check https://t.co/QgdnqhhNap

[HCOMP 2026 @hcomp_conf]

📣 The CFP for #HCOMP2026 is now live!

HCOMP 2026, the ACM Conference on Human-AI Complementarity and Alignment, will be co-located with @ci_acm this year 🤝

📌 Abstract deadline: June 1, 2026
🔥 Full paper deadline: June 8, 2026

Check https://t.co/QgdnqhhNap

[Gaole He @HeGaole]

4/ For more details about this work, please refer to our paper and code repo:

Paper: https://t.co/oRObdgwOOO

Code: https://t.co/VJWQfENT4n

[Gaole He @HeGaole]

1/ With high-level planning and external toolkits, LLM agents have shown good promises as assistants. In a recent #CHI2025 work, we conducted an empirical study with LLM agents as a daily assistant. This work is a collaboration with @eglu81 and @UjLaw.

[Gaole He @HeGaole]

3/ LLM agents can be a double-edged sword:
(1) they can work well when a high-quality plan and necessary user involvement in execution are available,
(2) users can easily mistrust the LLM agents with plans that seem plausible.

[Gaole He @HeGaole]

3/ We also provide promising future directions toward building effective conversational XAI interfaces. For more details, please check:

Paper: https://t.co/Z943QiecDI
Code: https://t.co/zeVD26m1tK

[Gaole He @HeGaole]

1/ Is Conversational XAI All You Need? Maybe not. In a recent #IUI2025 paper, we conducted an empirical study to compare conversational XAI interfaces against the XAI dashboard. This is a joint work with @UjLaw and @nilay_aishwarya at @wisdelft.

[Gaole He @HeGaole]

2/ Key takeaways:
1. Interactive XAI interface can amplify the illusion of explanatory depth
2. Enhancing the conversation quality (e.g., using LLM agents) further reinforced the over-reliance issues, and participants indicated worse objective feature understanding

[Gaole He @HeGaole]

A follow-up work based on our previous #HCOMP 2022 paper of analogy-based explanations was published in #JAIR. With a user study in skin cancer detection task and concept-level explanations, we show more insights about analogy in XAI

[J. AI Research-JAIR @JAIR_Editor]

New Article: "Opening the Analogical Portal to Explainability: Can Analogies Help Laypeople in AI-assisted Decision Making?" by He, Balayn, Buijsman, Yang, and Gadiraju https://t.co/kytmeuyHgf

[Gaole He @HeGaole]

@raquel_dmg I can help, feel free to include me

[Gaole He @HeGaole]

@damiano10 @wisdelft @RMITComputing @eglu81 @UQ_EAIT @SachinPC10 @OlegZendel @nattapat_bo @rankun203 It’s my pleasure to talk with you and discuss the amazing relevant topics😉

[Damiano Spina 🦋@damianospina.com @damiano10]

Great talk by @HeGaole from @wisdelft on Human-centred AI! Thanks for making the time to come to @RMITComputing while visiting @eglu81's group @UQ_EAIT.

Enjoy the rest of your time down under! 🐨
And thanks @SachinPC10 for making it happen!

[Gaole He @HeGaole]

@MilaNLProc @CurriedAmanda Thanks for your interest in our work!

[Gaole He @HeGaole]

Cognitive bias is an important human factor, which substantially affect human-AI collaboration. Please consider joining us for further discussion, CFP with CSCW 2023 workshop can be found in the tweet below.

[Nattapat Boonprakong @nattapat_bo]

Are you working or interested in Cognitive Biases? Join our hybrid #CSCW2023 workshop "Understanding and Mitigating Cognitive Biases in Human-AI Collaboration" where we will discuss how can we better understand, mitigate, and utilise cognitive biases. CFP: https://t.co/sHsovJyxCL

[Asterios Katsifodimos @kAsterios]

Please retweet!
I am searching for two PhD students and one postdoc for my VIDI grant. We are building a programming model & self-driving dataflow engine for Transactional Stateful Functions.

For PhD apply here by 15th July: https://t.co/SviNusp3FI

Postdoc: drop me an email!

[Gaole He @HeGaole]

Today I will be presenting our #CHI2023 paper on appropriate reliance in human-AI decision making. Come join us if you want to know more about appropriate reliance and how Dunning-Kruger effect impact human reliance on AI systems.

[Gaole He @HeGaole]

A good news to share, our work @wisdelft "Knowing About Knowing: An Illusion of Human Competence Can Hinder Appropriate Reliance on AI Systems" get accepted by #CHI2023 🥳. My pleasure to work with amazing coauthors Lucie Kuiper and @UjLaw. More details in the following threads: