Siyan Sylvia Li ✨
@siyan-li-sparkle.bsky.social
120 followers 130 following 20 posts
2nd year PhD @columbianlp • Prev @stanfordnlp @GeorgiaTech • Weird Little Guy Academic • NLP, Dialogue Systems, Education • Caffeine Gremlin 🩷💜💙
Posts Media Videos Starter Packs
siyan-li-sparkle.bsky.social
Should I point out in my meta review that 2/3 reviewers on the paper used AI and therefore I don't take those reviews seriously
siyan-li-sparkle.bsky.social
📅 Deadlines:
Regular submissions: Aug 8
Fast-track submissions: Sept 1
Submit & learn more 👉https://sites.google.com/view/hciandnlp
Let's shape the future of human-centered NLP together. 💬❤️ #HCINLP2025 #EMNLP2025
8/8
siyan-li-sparkle.bsky.social
📢 Call for Papers
We invite:
🔹 Research papers (up to 8p, archival or not)
🔹 Extended abstracts (4p, non-archival)
🔹 Fast-track resubmissions with previous reviews
🔹 Provocations, position papers, surveys, demos & more

7/
siyan-li-sparkle.bsky.social
In addition to the regular workshop topics, we welcome work on:
• Human feedback & preferences
• Interfaces for collecting feedback
• New paradigms for human input
• Tasks where human involvement really matters

6/
siyan-li-sparkle.bsky.social
🧠 This year’s special theme: Human involvement in post-training
As labs pivot toward AI-generated feedback and synthetic data, we ask: What should the role of real human input be in shaping models?

5/
siyan-li-sparkle.bsky.social
That’s where HCI comes in. HCI+NLP is a space for researchers who care about people, context, and design to shape the future of language technologies. ✨

4/
siyan-li-sparkle.bsky.social
Fairness, evaluation in the wild, explainability—these aren’t just technical issues. They're human ones. 🤝 Now more than ever, NLP must integrate human-centered methods into its core research.

3/
siyan-li-sparkle.bsky.social
With the rise of LLMs, NLP is more present than ever in our daily lives—chatbots, writing tools, conversational search, and more. But with this expansion come critical challenges.

2/
siyan-li-sparkle.bsky.social
🎉 Excited to announce that the 4th HCI+NLP workshop will be co-located with @emnlpmeeting.bsky.social in Suzhou, China! 🌍📍 Join us to explore the intersection of human-computer interaction and NLP. 🧵

1/
siyan-li-sparkle.bsky.social
This is true yes, okay I feel better about it now lol ❤️
siyan-li-sparkle.bsky.social
Just learned that my paper will be the first in the session at NAACL (9:00 am to 9:15 am); will people even show up? 😭😭😭
siyan-li-sparkle.bsky.social
The process for collecting this dataset was documented in our prior work: arxiv.org/abs/2404.13764
This dataset has powered the development of the empathetic response module in EDEN! We manually labeled a subset of this data with markers of negative emotions and excessive pauses.
Using Adaptive Empathetic Responses for Teaching English
Existing English-teaching chatbots rarely incorporate empathy explicitly in their feedback, but empathetic feedback could help keep students engaged and reduce learner anxiety. Toward this end, we pro...
arxiv.org
siyan-li-sparkle.bsky.social
📢 We are releasing an Accented Speech ASR dataset!! This dataset is collected from native Mandarin speakers practicing English from my advisor Zhou Yu's platform. There are 3081 audio clips here, plus high-quality, human-verified transcripts.
huggingface.co/datasets/syl...
sylviali/EDEN_ASR_Data · Datasets at Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
siyan-li-sparkle.bsky.social
Zotero!! I sync it with my computer so I can add papers on my computer and read it on my iPad!
Reposted by Siyan Sylvia Li ✨
balloch.bsky.social
I am also looking for new positions, so if you are looking for someone with deep expertise in reinforcement learning, robotics, and deep learning in general, hit me me up!
siyan-li-sparkle.bsky.social
Congratulations!!!!!!! So happy for you!!!
Reposted by Siyan Sylvia Li ✨
srushnlp.bsky.social
Rare personal tweet:
Subletting our furnished apartment in Brooklyn for the spring at a significant discount. It's quite nice and in a fun location. under price. Email me know if you are interested, I will send pictures.
Reposted by Siyan Sylvia Li ✨
joestacey.bsky.social
Okay genius idea to improve quality of #nlp #arr reviews. Literally give gold stars to the best reviewers, visible on open review next to your anonymously ID during review process.

Here’s why it would work, and why would you should RT this fab idea:
siyan-li-sparkle.bsky.social
I am currently in a class where homeworks are down-weighted significantly and exams are up-weighted significantly due to ChatGPT, but that causes unnecessary stress as well... I feel like a more ideal setup is some form of time-constrained pop quiz so that students won't have time to consult online?
Reposted by Siyan Sylvia Li ✨
davidberenstein.bsky.social
I created a collection with good models for dataset curation

- NSFW classifiers
- PII classifiers
- blazing fast embeddings by model2vec
- quality classifier
- educational value classifier
- domain classifier

Collection: huggingface.co/collections/...
Models for dataset curation - a Dataset-Tools Collection
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
siyan-li-sparkle.bsky.social
But coming up with fun acronyms for systems is the best part!!!
Reposted by Siyan Sylvia Li ✨
michael-j-black.bsky.social
For those who missed this post on the-network-that-is-not-to-be-named, I made public my "secrets" for writing a good CVPR paper (or any scientific paper). I've compiled these tips of many years. It's long but hopefully it helps people write better papers. perceiving-systems.blog/en/post/writ...
Writing a good scientific paper
perceiving-systems.blog
Reposted by Siyan Sylvia Li ✨
saxon.me
I propose that instead of "posts" or "skeets" (ew) we refer to bluesky posts as BS

"I BS'd that"
"Oh I saw some BS earlier about that"
"You should make that a BS"
siyan-li-sparkle.bsky.social
🙏🙏🙏✨ Thank you!!!