How do VLMs perform on answering questions about process visualizations? @kzintas presents findings at the #VizWiz workshop @ #CVPR2025 . hdi.cs.umd.edu/papers/VLM_Pro…
0
3
4
285
Human-Data Interaction Group @ UMD retweeted
A week ago, I defended my PhD dissertation. It still feels surreal, and I’m still taking it all in, but I’m beyond grateful to be done! 😅

A huge thank you to my committee members, @zcliu, @leibatt, and members of @hdi_umd for your invaluable support and encouragement 🙏🏾
1
1
7
271
Human-Data Interaction Group @ UMD retweeted
😃Just defended my PhD on "Bridging the Gulf of Evaluation in Human-AI interaction for Knowledge Workers", huge thanks to my advisor @zcliu, my committee Prof. David Weintrop, @haldaume3 @fumeng_yang @RuohanGao1 , my labmates @hdi_umd, and collaborators at @Adobe @BoschGlobal
7
6
91
5,781
Check out the full paper published at CHI 2025 here: hdi.cs.umd.edu/papers/COALA_C…
Poster = small multiples of slides! “Comparing native and non-native English speakers’ behaviors in collaborative writing through visual analytics” #in2writing #naacl #NAACL2025
0
0
1
69
We just released a new open-source repository by @kzintas, which includes implementations of three visual summarization techniques for event sequence data: Sequence Synoposis, SentenTree, and CoreFlow. Check it out here: github.com/hdi-umd/Sequen…
0
3
5
537
Human-Data Interaction Group @ UMD retweeted
We created COALA, a visual analytics tool that helps understand how native & non-native English speakers write differently when writing together. Our findings can help build better AI writing assistants for teams. Related #CSCW paper: dl.acm.org/doi/10.1145/36…
Excited to share our #CHI2025 paper "Comparing Native and Non-native English Speakers' Behaviors in Collaborative Writing through Visual Analytics" with @zcliu @kzintas @umdcs Yimin Xiao, Ge Gao @INFOCollegeUMD @hcil_umd Naomi Yamashita @NTTPR #NAACL #in2writing #DataAnalytics
0
5
9
4,798
Human-Data Interaction Group @ UMD retweeted
Can you feel the Chen-ergy?! Huge Congratulations to Dr @ChenChe91591871 from all of us at @hdi_umd for successfully defending his dissertation! 🥳🥳🎓
0
2
2
476
Human-Data Interaction Group @ UMD retweeted
I’m excited to share our work on evaluating the semantic profiling abilities of Large Language Models (#LLMs) for NL utterances at #IEEEVIS! 💻✨
Checkout our paper and Repo:
📰arxiv.org/abs/2407.06129
💾 Github: github.com/hdi-umd/Semant…
1
3
11
828
We will be presenting three full papers (one of which won an Honorable Mention award), two short papers, and one workshop article at #IEEEVIS 2024. Here is a preview of the upcoming presentations. 🧵 1/7
1
6
7
771
Replying to @hdi_umd
On Thursday (10/17) morning, @hannah_bako will present results from an evaluation of the semantic profiling abilities of LLMs on natural language utterances in data visualization. Joint work with Arshnoor Bhutani, Xinyi Liu, @kwesicobbina1, and @zcliu. 6/7
1
0
2
176
Replying to @hdi_umd
On Thursday (10/17) afternoon, @hannah_bako will unveil the influence of examples on data visualization design outcomes. Spoiler alert: timing and data schema matter in example-based design. Joint work with Xinyi Liu, Grace Ko, Hyemi Song, @leibatt , @zcliu. 7/7
0
0
2
156
At #CHI2024, @tracyyxchen is presenting our work on turning instructional videos into interactive mixed media tutorials for physical tasks. #TutoAI #humanAI_interaction #co_creation
0
4
8
734
Human-Data Interaction Group @ UMD retweeted
Excited to share our #CHI2024 paper "TutoAI: A Cross-domain Framework for AI-assisted Mixed-media Tutorial Creation on Physical Tasks" with @zcliu from @umdcs @hcil_umd and Vlad Morariu and Anh Truong from @AdobeResearch hdi.cs.umd.edu/papers/TutoAI_…
1
4
7
857
.@ChenChe91591871 will present our work on deconstructing SVG charts for layout reuse at @ieeevis. His talk is on Wednesday, in the meantime, you can check out the paper arxiv.org/abs/2307.13567 and the demo website cchen-vis.github.io/Mystique/
0
1
3
1,238
Human-Data Interaction Group @ UMD retweeted
Enjoyed presenting @ChenChe91591871's great work on the state of the art in creating visualization corpora for automated chart analysis. Thanks for the interesting discussions! Lots of work to be done to enhance the rigor and generalizability of AI4VIS works.
1
3
20
1,773
Many visual summarization techniques have been invented for event sequence data, which ones work better and why? @kzintas presents the design and results of the first study to compare the effectiveness of three algorithms: Coreflow, SentenTree and Sequence Synopsis. #eurovis
0
4
9
1,011