| CARVIEW |
Select Language
HTTP/2 200
server: GitHub.com
content-type: text/html; charset=utf-8
last-modified: Thu, 18 Dec 2025 03:34:16 GMT
access-control-allow-origin: *
strict-transport-security: max-age=31556952
etag: W/"69437638-4395"
expires: Mon, 29 Dec 2025 04:09:30 GMT
cache-control: max-age=600
content-encoding: gzip
x-proxy-cache: MISS
x-github-request-id: AF76:234FE9:84ABD3:95082C:6951FCA1
accept-ranges: bytes
age: 0
date: Mon, 29 Dec 2025 03:59:30 GMT
via: 1.1 varnish
x-served-by: cache-bom-vanm7210041-BOM
x-cache: MISS
x-cache-hits: 0
x-timer: S1766980770.172411,VS0,VE204
vary: Accept-Encoding
x-fastly-request-id: d7d48d88ef39e8676c3a9199d6691dca2794a360
content-length: 5593
Zora Wang
Apr 2025: Released two new works on agent skill induction via online adaptation [ASI]π and offline exploration [SkillWeaver].
Apr 2025: Gave an tutorial at the 2nd CMU Agent Workshop about "Language Model-Based Agents" [slides]
Mar 2025: Gave an invited talk at the Advancing LLM-Based Multi-Agent Collaboration Workshop [slides]
Mar 2025: Gave two talks at AAAI Web Agent Revolution Workshop, about Agent Workflow Memory and The Agent Company.
Jan 2025: Check out the podcast episode about interesting discussions⨠around "Agent Workflow Memory" [link]
Dec 2024: Gave a talk at Tsinghua University about "From Workflows to Tools: Building Adaptive Agents On the Fly" (slides)
Nov 2024: Gave a talk at UIUC iSE Seminar about "Solving Real-World Tasks via Program Generation". Feel free to check out my slides here.
Oct 2024: Joined the panelist of Our CS (Workshop for Undergraduates in CS) and shared some of my thoughts about research
Sep 2024: Gave an invited talk at Camel-AI πͺ about Agent Workflow Memory, check out the video, paper, and tweet
Jul 2024: Gave a tutorial at SIGIR about Large Language Models for Tabular Data, check out the slides & recordings if you're interested!
May 2024: Organized our CMU Agent Workshop π€ with plenty of events -- insightful tutorials, talks, and posters! I also gave two (short) tutorials about tool-augmented LMs and codegen testbeds.
Mar 2024: Gave a guest lecture about "Language Agents and Tool Use" at the Advanced NLP course (11-711) course, check out the recordings!
Mar 2024: Gave a talk at the [FLAME_ (Foundation and LAnguage Model) π₯] seminar, about [our recent survey] and [TroVE]
Feb 2024: Gave a talk about Language Models with Tools at the LLM as Agent Seminar, about [TroVE] and works in progress π€«
Feb 2024: Gave a lecture about Evaluation (metrics and benchmarks) for the Neural Code Generation (11-891) course π»
Jan 2024: TAing for the new course [11-891 Neural Code Generation], reach out if you want to discuss more project ideas πͺ
Nov 2023: Gave a guest lecture about Evaluation and Benchmarks for Code Generation for the Advanced NLP course (11-711) π©βπ« more details [here]
Aug 2023: A talk about π οΈ Tool using, learning, and making with LLMs at Code Generation Reading Group, check out the [video]
Apr 2023: Gave a talk about [ODEX] at the Machine Learning Methods in Software Engineering (video) hosted by JetBrains Research Team π©βπ»
PhD Student
Carnegie Mellon University
Language Technologies Institute
I am currently a PhD student at the Language Technologies Institute at Carnegie Mellon University working with Daniel Fried and Graham Neubig. My primary research interest is to use programmatic approaches to solve real-world tasks. Particularly, I am working on:
Computer-Use Agents for Work
- Study "How AI agents do human work?"
- Induce and analyze ai and human workflows across jobs [workflow induction toolkit]
- Develop agent workers for various tasks (e.g., presentation making [AutoPresent]) and occupations (software engineering [ODEX], jobs across company [TAC])
Memory- and Skill-Adaptive Agents
- Digital agents that learns navigation workflows in memory [AWM] and specialized skills as actuators [ASI] [SkillWeaver]
- Build verifiable and efficient toolboxes for math, data analysis, and visual reasoning tasks [TroVE]
- These agents often facilitate human verification [TroVE] and improves task success [CowPilot]
Augmented Language Models
- Tools [Survey]: designed by human experts [AutoPresent] or LMs [TroVE]; their behaviors to under-specified queries and execution failures [Fail-TaLMs]
- Programs: Challenging benchmarks for open domain [ODEX], multilingual [MCoNaLa], and data science [HiTab] problems; Performant code LMs [StarCoder] [ECCO] with [Code RAG], assistive [APIs], and libraries in [DocPrompting]
- Knowledge: retrieve texts [FilCo] [ReAtt] [RAGGED] and structured data [WikiTable] [TUTA]
News
- Dec 2025: Give an invited talk "How Do AI Agents Do Human Work?" at AI-TIME
- Dec 2025: Organized the DL4C Workshop at NeurIPS 2025, featuring "Deep Learning for Code in the Agentic Era"
- Nov 2025: Give an invited talk "How Do AI Agents Do Human Work?" at Ploutos Fireside Chat
- Oct 2025: Excited to be awarded the Google PhD Fellowship!
- Sep 2025: Give a guest lecture for Human-Centered NLP about language model-based agents [slides]
- Jul 2025: Give a talk about "Building Memory- and Skill-Adaptive Agents" at Renmin University of China (RUC) [slides] [poster]
- Jun 2025: Give two talks about "Building Memory- and Skill-Adaptive Agents" [slides] and "Rethinking Human's Role in Building Better Agents" [slides] in SF π
- May 2025: Started as a Visiting Student Researcher at Stanford NLP Group working with Diyi Yang. Happy to meet new/old friends! π
πΎ Older News
Upcoming Events
- Jul 2026: Will give a tutorial on the "Future of Work in the Age of LLMs" at ACL 2026, see people in San Diego (again)!
My Recent Favorite Publications
![[poster]](carview.php?tsp=images/ruc-talk-poster.jpg){kind=link}
Get Connected
If you want to get connected, discuss potential project ideas, ask about CMU application (check out GASP), or chat about any other relevant topics, my office hours are available every Friday 4-5 pm EST. Email me to secure a time to chat!
I often mentor a small number of students every semester, please fill out this application form if you're interested in working with me! However, as I got plenty of emails, I might not be able to reply to all of them.
If you are from underrepresented groups, or do not have much research experience, you are encouraged to reach out!