Notes on Reading, Writing, and Thinking with AI
Joseph Chee Chang
I am a research scientist at the Allen Institute for AI (Ai2) working on the Asta project. I received my PhD from the Human-Computer Interaction Institute and the Language Technologies Institute and at Carnegie Mellon University.
My research sits at the intersection of HCI and NLP, with the goal of building systems that help people navigate, understand, and synthesize information. I primarily publish in HCI venues (e.g., CHI, UIST) and occationally in NLP conferences (e.g., ACL, COLM).
I have several active research threads: developing literature review tools and deep research systems, building AI-augmented reading interfaces that make scholarly papers interactive and intelligent, and studying broad human-AI collaboration to design effective partnerships between people and AI agents. Recently, I have also focused on interactive interfaces for exploring causal knowledge graphs, literature-based ideation, and theory development. I have also focused on crowdsourcing and human computation and supporting online research with novel web browsers in the past.
Active Research Topics
- Deep Research and AI Lit Review Tools
- AI-Augmented Reading Interfaces
- Understanding Human-AI Interaction
- Social Computing for the Research Community
- New: Interactive Causal Knowledge Graphs
Come Work With Me
I am actively looking for research interns to join my group at the Allen Institute for AI. Interns work closely with me and a small group of co-mentors on a focused research project, typically resulting in a publication at a top venue. I typically mentor HCI interns, but also co-mentor NLP/ML interns.
Past interns have published at top venues such as CHI, UIST, ACL, and COLM, and have come from institutions across the world including UW, KAIST, U of Toronto, and MIT.
Interested? Learn more and apply.
Find Me
Recent Activities
CHI 2026 — Barcelona, Spain. Kevin will present our paper Cocoa: Co-Planning and Co-Execution with AI Agents (best paper) and I will host a workshop on Augmenting Reading
COLM 2025 — Montreal, Canada. I presented our paper LLMs as Research Tools: A Large Scale Survey of Researchers' Usage and Perceptions
ACL 2025 — Vienna, Austria. I gave a tutorial on Human-AI Collaboration: How AIs Augment Human Teammates and presented our demo Ai2 Scholar QA
CHI 2025 — Yokohama, Japan. Kevin presented our paper IdeaSynth: Iterative Research Idea Development Through Evolving and Composing Idea Facets with Literature-Grounded Feedback and Ruotong presented Social-RAG: Retrieving from Group Interactions to Socially Ground AI Generation
The Birder's Column
When not building AI tools, I can often be found in the forests and wetlands of the Pacific Northwest with a camera and a long lens. I recently became an avid birder, and am slowly building a photography portfolio of the region's incredible avian diversity.
Photo: Short-eared Owl, Skagit Valley