Tobias Weinberg
I am a CS PhD Candidate at Matter of Tech Lab and Apple AIML PhD Scholar Fellow, advised by Prof. Thijs Roumen at Cornell Tech and co-advised by Prof. Stephanie Valencia from UMD. My research interests center on reimagining augmentative and alternative communication (AAC) as a vehicle for personal expression rather than a mere substitute for speech. I develop AI-driven tools that transcend basic transactions to foster truly expressive conversational engagement. I focus on creating and researching AAC systems that prioritize user agency and personal identity. By addressing privacy and adaptability, my work seeks to unlock meaningful communication for those whose expressive voices are often ignored.
My research is supported by the Apple Scholar in AIML PhD Fellowship (2026)
Areas of interest: accessibility, disability, AAC, expressive communication, LLMs, human-AI interaction, human-centered machine learning, and HCI.
Publications
Tobias Weinberg, Ricardo E. Gonzalez Penuela, Stephanie Valencia, Thijs Roumen
We study how ultra-personalized AI can support AAC users’ voices. A 10-month auto-ethnography shows how fine-tuned models enhance fluency while raising new questions around agency, identity, and privacy
Project page • Video • Paper
Shuo Feng, Xuening Wang, Yifan (Lavenda) Shan, Krista U Singh, Bo Liu, Amritansh Kwatra, Ritik Batra, Tobias M Weinberg, and Thijs Roumen
We investigate how CAD interfaces can guide exploration and comparison of workflows. Specifically, comparison can advance users' reasoning about design decisions. We developed a prototype interface, CAMeleon, which lets users compare fabrication workflows. Users load 3D models and preview outcomes from different workflows.
Project page • Video • Paper
Tobias Weinberg, Claire O'Connor, Ricardo E. Gonzalez Penuela, Stephanie Valencia, Thijs Roumen
We study how AAC users create and negotiate expressive backchanneling using multi-modal cues, this study reveals design opportunities to support real-time presence and conversational rhythm in AAC technology.
Project page • Video • Paper
Tobias Weinberg, Kowe Kadoma, Ricardo E. Gonzalez Penuela, Stephanie Valencia, Thijs Roumen
Best Paper Honorable Mention Award (best 5%) 🏅
Jury Best Demo Award 🏆
We study how AI-powered AAC interfaces can help users with speech disabilities deliver timely, humorous comments. This study highlights design insights to improve expressivity and timing in AAC technology.
Project page • Video • Paper
Amritansh Kwatra, Tobias Weinberg, Ilan Mandel, Ritik Batra, Peter He, Francois Guimbretiere, and Thijs Roumen
Best Paper Honorable Mention Award (best 5%) 🏅
We present SplatOverflow, a workflow to support end-users, community members and manitainers asynchronously troubleshoot hardware issues.
Project page • Video • PaperNews
- May 2026 Super excited to start my internship as Student Researcher at Google Research in Boulder, CO this summer! I'll be working with Shaun Kane on research to enhance expressivity in AAC technologies with AI! 🎉🤖🗣️
- May 2026 I've been spotlighted by the Cornell gradschool "Student Spotlight: Tobias Weinberg"
- April 2026 Article about our paper "I, Robot? Exploring Ultra-Personalized AI-Powered AAC; an Autoethnographic Account" in the Cornell Chronicle "What does it mean to train an AI to speak you?"
- April 2026 I'm super honored to be named a recipient of the Apple Scholar in AIML PhD Fellowship (2026)!
- April 2026 I'm honored to be invited to speak at Google's Community Perspectives seminar in NYC offices, talk title: Putting the A in AAC: Expressive Participation in Conversations
- Jan 2026 Excited to announce that our paper "Comparing Fabrication Workflows in CAD to Support Design Reasoning" has been accepted to ACM CHI 2026!
- Nov 2025 Super cool article in the Cornell Chronicles about my research "Student draws on experience to transform assistive communication"
- Oct 2025 I received my Master's degree in Computer Science from Cornell University! 🎓
- Oct 2025 I succefully defended my PhD candidacy exam! Excited to continue my research on enhancing expressivity in AAC technologies with AI! 🎉🤖🗣️
- Jun 2025 Excited to announce that our paper "One Does Not Simply ‘Mm-hmm’: Exploring Backchanneling in the AAC Micro-Culture" has been accepted to ACM ASSETS 2025!
- May 2025 Excited to announce that our demo of "Why So Serious? Exploring Timely Humorous Comments in AAC Through AI-Powered Interfaces" won Jury Best Demo Award 🏆 at CHI2025 in Japan! 🇯🇵
- Mar 2025 Article on YAI newslestter about our work together to help people with disabilities "Bringing Humor and Innovation to Assistive Technology"
- Mar 2025 Our paper "Why So Serious? Exploring Timely Humorous Comments in AAC Through AI-Powered Interfaces" won best paper Honorable mention award (best 5%)🏆!!
- Mar 2025 Our paper "SplatOverflow: Asynchronous Hardware Troubleshooting" won best paper Honorable mention award (best 5%) congratulations Amrit🏆!!!
- Jan 2025: My first, first author, paper "Why So Serious? Exploring Timely Humorous Comments in AAC Through AI-Powered Interfaces" has been accepted to CHI 2025! Super excited Japan here we go🇯🇵🇯🇵!
- Jan 2025: Our paper "SplatOverflow: Asynchronous Hardware Troubleshooting" has been accepted to CHI 2025!
- Sep 2024: Article PiTech Fellowship Spotlight "Transforming Care with Smart Home Technology: Data-Driven Solutions for Group Homes"
- Jul 2024: Extremely proud of my team, and extremely honored to receive Amazon cloud computing grant from Cornell Data Science Center to further my research towards leveraging AI to bring expressive communication to people with speech disabilities!🎉💰
- Mar 2023: 🎉 Received Siegel Public Interest Technology Impact Fellowship 🎉
- Mar 2023: Article in Leading Newspaper (Infobae Argentina) "The young man with a brilliant mind and a rare illness who received a scholarship in the United States" (in spanish)
- Oct 2022: Article in the American Technion Society “The sky is not the limit . . . turning ideas into reality to change the world”