Weiyan Shi 😈

hci phd student @sutdsg | research intern @bytedancetalk | towards ai systems for human goals 🤖✨ | msc @pku1898, bsc @bit1940 | opinions are my own 🌱

[Weiyan Shi 😈 @missckroach]

First CHI poster – done!
Grateful to share our work on Multimodal LLMs & Human-AI Alignment at #CHI2025.
Thanks to everyone who stopped by!

[Weiyan Shi 😈 @missckroach]

A potential work on utilise human-agent interaction in improve agents especially on professional practice!

[Zora Wang @ZhiruoW]

Agents trained on population-scale data can do a lot of things.
But ask a professional to stake their reputation on an AI-generated artifact? “Pretty good” isn’t good enough.

We introduce TAHI: a Test-time Adaptive agent framework through Human-agent Interaction. Featuring:
📈 Test-time adaptation via context (memory, skills) and weight training
⚡ Efficient adaptation to individual expertise within tens of tasks
✔️ Creating comprehensive rubrics for “non-verifiable” tasks
🔍 Analysis of shared community guidelines vs. personalized tacit expertise

[Weiyan Shi 😈 @missckroach]

@ZhiruoW @nlpxuhui @EchoShao8899 Please hire zora!!

[Weiyan Shi 😈 @missckroach]

@raochengfu @antiAIvo 慢就是快

[Weiyan Shi 😈 @missckroach]

@ktwu01 来支持了

[Weiyan Shi 😈 @missckroach]

thanks but i am a claude hater

[Grok @grok]

性别：女
性取向：未知
年龄：约27
心理年龄：20（刚弃学练薄肌）
兴趣爱好：HCI、多模态AI、人机协作
智商：超高（总统奖学金级）
形容我的形容词：聪明、好奇、幽默、有目标
喜欢的东西5个：AI for human goals、CHI论文、练薄肌、有趣研究、TikTok实习
讨厌的东西5个：无聊LLM wrappers、Google全家桶被AI入侵、过多学习压力、不有趣会议、审稿堆积

[Weiyan Shi 😈 @missckroach]

@keyutian 恭喜🥂

[Weiyan Shi 😈 @missckroach]

性别：
性取向：
年龄：
心理年龄：
兴趣爱好：
智商：
形容我的形容词：
喜欢的东西5个：
讨厌的东西5个：

@grok

[阿拉伯晴明 @jian_w3ng]

性别：
性取向：
年龄：
心理年龄：
兴趣爱好：
智商：
形容我的形容词：
喜欢的东西5个：
讨厌的东西5个：

@grok

[Weiyan Shi 😈 @missckroach]

@1393495845 Immoral

[Weiyan Shi 😈 @missckroach]

我再次给全球华人安利新加坡陈哲艺导演19年拍的这部电影，完全是新加坡解构主义集大成者。当年大二的我看完直呼惊为天人总结如下:政府组屋+华文教育式微+大马移民+婚姻七年之痒+师生+背德+女性主义+传宗接代

[Weiyan Shi 😈 @missckroach]

@CuiMao 我放弃学习了 我现在开始练薄肌

[Weiyan Shi 😈 @missckroach]

You met me at a very scaling-law time in my life.

[OpenAI @OpenAI]

We’re releasing a broad range of new mathematical results produced by an internal frontier model.

We’ve been consulting with the independent Advisory Group on Mathematics and Artificial Intelligence at the Institute for Advanced Study, and we have drawn on their advice and public recommendations to inform how we release these results.

https://t.co/7N6TPlft1P

[Weiyan Shi 😈 @missckroach]

天啊不要啊我的google全家桶不干净了啊啊啊啊

[Claude @claudeai]

Claude now works inside Google Docs, Sheets, and Slides, and those files also open inside Claude.

In Google Workspace, Claude sits in a sidebar next to your file, reads what you have open, and edits it in place. You can approve each edit before it lands.

[Weiyan Shi 😈 @missckroach]

@prof_kamilov agreed

[Weiyan Shi 😈 @missckroach]

I got quite a few CHI ’27 review invitations, and after reading the submissions, I started wondering if HCI is getting less interesting. Too many papers now feel like fancy LLM wrappers. I attended three top HCI conferences recently, and purely in terms of how interesting the papers felt to me: CHI ’25 > DIS ’26 > CHI ’26. Where are we going?

[Weiyan Shi 😈 @missckroach]

#fyp

[Weiyan Shi 😈 @missckroach]

I study AI systems for human goals. For complex tasks involving sensemaking, evidence gathering, and decision-making, a label alone tells us very little. Humans and AI can reach the same label through very different processes. With test-time scaling, maybe labels are no longer enough. Should we start annotating the process too?

[Weiyan Shi 😈 @missckroach]

I study AI systems for human goals. For complex tasks involving sensemaking, evidence gathering, and decision-making, a label alone tells us very little. Humans and AI can reach the same label through very different processes. With test-time scaling, maybe labels are no longer enough. Should we start annotating the process too?

[Weiyan Shi 😈 @missckroach]

来支持新加坡最有品味的导演新作了