Kevin Pu
@kevpjk
Followers
535
Following
458
Media
12
Statuses
86
CS PhD student @dgpToronto @UofT | HCI Research | Former @UMich Wolverine | prev @Meta @RealityLabs @allen_ai @SemanticScholar
Joined August 2022
I am presenting IdeaSynth in the last paper session at #CHI2025 right now! .Feel free to come by G314-315 to learn about how we utilize LLM to provide literature-grounded assistance for research idea development! The talk is happening at around 10:12 AM.
🔬Research ideation is hard: After the spark of a brilliant initial idea, much work is still needed to further develop it into a well-thoughtout project by iteratively expanding and refining the initial idea and grounding it to relevant literature. How can we better support this?
1
8
45
This is happening tomorrow at 2:10 PM at the Coding and Development session at G418-419! . Come by and check out how a proactive agent with its own cursor editing the code with you impacts AI programming usage and developer experience!.
🤖👩🏻💻Proactive AI tools like @GitHubCopilot @cursor_ai @allhands_ai promise to assist developers by anticipating their needs and automating engineering processes—but do they truly help?.We evaluated three design probes to explore the trade-offs of proactive AI programming support
1
2
23
RT @IanArawjo: How should we update AI memory of user intent as intent evolves? 🤖💭 How can we auto-update memory docs like Cursor rules as….
0
22
0
RT @bingjian_huang: #CHI2025 [Free Dev Kit!✨].Are you an HCI researcher curious about haptics, but don’t know where to start?🤔.Or a haptics….
0
10
0
RT @ToviGrossman: Congrats to @karthikm0 (and amazing co-authors!) on the Best Paper Award at #HRI2025 for "ImageInThat: Manipulating Image….
0
7
0
RT @marissa_rad: We're seeking CS researchers to participate in a study on working with AI tools to come up with research project ideas!….
0
8
0
RT @HCI_Prof_YC: #CHI2025 Coding Partner or Intruder? We designed, tested Codellaborator, a proactive AI assistant that initiates coding he….
0
6
0
🎉 Shoutout to my collaborators @dnlzro @IanArawjo @HaijunXia @ZiangXiao @ToviGrossman @HCI_Prof_YC! . The paper is conditionally accepted to #CHI2025, link to preprint: Share your thoughts and let’s discuss how to design better AI tools for developers!.
0
0
11
👩🏻🎨Implications:.We share 5 design implications to enhance human-AI collaboration in programming, including facilitating code understanding, adapting AI salience, and adjusting proactivity for each sub-process. We hope these insights can help guide future AI programming tools!.
1
0
1
📋 Takeaways:.Users desired more proactivity for specific processes like code scaffolding or debugging.Balancing proactivity with flexible user control and agency is key.Proactivity reduces low-level efforts but also leads to over-reliance, citing concerns over code understanding.
1
0
0
📊 Quantitative findings:.Efficiency: Users spent less time interpreting proactive AI responses than prompted ones.Awareness: Visual context cues improved users' understanding of AI actions.Trade-offs: Proactivity boosted the sense of collaboration but also disrupted workflows.
1
1
1
👩🏻🔬We explored three probe systems:.- PromptOnly: User prompts AI for help via chat or code comments.- CodeGhost: AI agent proactively intervenes based on user activities.- Codellaborator: Proactive agent with presence cues and both global and local interaction scopes
1
0
1
🛠️ Design Probe Dimensions:.- Proactive timing: AI intervenes during task boundaries or signals of user need.- Visual presence: Represent AI actions and context to increase awareness.- Flexible scope: Supports global and local interactions via chat and breakout threads
1
0
1
🤔Why Proactive?.Most tools require prompting, which can be tedious and interrupt flow. Proactive tools aim to intervene at the right time, reducing user effort. But:.- How can we design proactive support?.- Can they improve coding efficiency?.- Do they disrupt users' workflows?.
1
0
0
🤖👩🏻💻Proactive AI tools like @GitHubCopilot @cursor_ai @allhands_ai promise to assist developers by anticipating their needs and automating engineering processes—but do they truly help?.We evaluated three design probes to explore the trade-offs of proactive AI programming support
1
14
59
A great line of research works! Feel free to check out my two #CHI2025 papers:.IdeaSynth: Iterative Research Idea Development. Assistance or Disruption? Evaluating Designs of Proactive AI Programming.More details on this one soon!😉.
That’s the lineup! 🚀 Full details here:. 🌟 Shoutout to the incredible grad students and post-docs involved this work! Give them a follow to stay updated on their research:. @LepingQ, @joaleong, @kevpjk, @VioletHanyinuo, @ItsHaiNotHi, @MajeedKazemi,.
5
3
37