Portrait of Ian Arawjo

Ian Arawjo

Associate Academic Member
Assistant Professor, Université de Montréal, Department of Computer Science and Operations Research
Website
X

Biography

Ian Arawjo is an assistant professor in the Department of Computer Science and Operations Research (DIRO) at Université de Montréal. He holds a PhD in information science from Cornell University, where he was advised by Tapan Parikh. His dissertation work spanned the intersection of computer programming and culture, investigating programming as a social and cultural practice. Arawjo has experience applying a range of human-computer interaction (HCI) methods, from ethnographic fieldwork, to archival research, to developing novel systems (used by thousands of people) and running usability studies.

Currently, he works on projects at the intersection of programming, AI and HCI, including how new AI capabilities can help us reimagine the practice of programming. He also works on large language model (LLM) evaluation, through high-visibility open-source projects such as ChainForge. His first-authored papers have won awards at top HCI conferences, including the Conference on Human Factors in Computing Systems (CHI), the Computer-Supported Cooperative Work and Social Computing Conference (CSCW) and the User Interface Software and Technology Symposium (UIST).

Current Students

Munyeong Kim
Master's Research - Université de Montréal
jingyue zhang
PhD - Université de Montréal
Principal supervisor :
Damien Masson

Publications

Who Validates the Validators? Aligning LLM-Assisted Evaluation of LLM Outputs with Human Preferences
Shreya Shankar
J.D. Zamfirescu-Pereira
Bjorn Hartmann
Aditya G Parameswaran
Ian Arawjo
Due to the cumbersome nature of human evaluation and limitations of code-based evaluation, Large Language Models (LLMs) are increasingly bei… (see more)ng used to assist humans in evaluating LLM outputs. Yet LLM-generated evaluators simply inherit all the problems of the LLMs they evaluate, requiring further human validation. We present a mixed-initiative approach to ``validate the validators'' -- aligning LLM-generated evaluation functions (be it prompts or code) with human requirements. Our interface, EvalGen, provides automated assistance to users in generating evaluation criteria and implementing assertions. While generating candidate implementations (Python functions, LLM grader prompts), EvalGen asks humans to grade a subset of LLM outputs; this feedback is used to select implementations that better align with user grades. A qualitative study finds overall support for EvalGen but underscores the subjectivity and iterative process of alignment. In particular, we identify a phenomenon we dub \emph{criteria drift}: users need criteria to grade outputs, but grading outputs helps users define criteria. What is more, some criteria appears \emph{dependent} on the specific LLM outputs observed (rather than independent criteria that can be defined \emph{a priori}), raising serious questions for approaches that assume the independence of evaluation from observation of model outputs. We present our interface and implementation details, a comparison of our algorithm with a baseline approach, and implications for the design of future LLM evaluation assistants.
2024-04-18
ArXiv (preprint)
doi.org
arxiv.org
Who Validates the Validators? Aligning LLM-Assisted Evaluation of LLM Outputs with Human Preferences
Shreya Shankar
J.D. Zamfirescu-Pereira
Bjorn Hartmann
Aditya G Parameswaran
Ian Arawjo
Due to the cumbersome nature of human evaluation and limitations of code-based evaluation, Large Language Models (LLMs) are increasingly bei… (see more)ng used to assist humans in evaluating LLM outputs. Yet LLM-generated evaluators simply inherit all the problems of the LLMs they evaluate, requiring further human validation. We present a mixed-initiative approach to ``validate the validators'' -- aligning LLM-generated evaluation functions (be it prompts or code) with human requirements. Our interface, EvalGen, provides automated assistance to users in generating evaluation criteria and implementing assertions. While generating candidate implementations (Python functions, LLM grader prompts), EvalGen asks humans to grade a subset of LLM outputs; this feedback is used to select implementations that better align with user grades. A qualitative study finds overall support for EvalGen but underscores the subjectivity and iterative process of alignment. In particular, we identify a phenomenon we dub \emph{criteria drift}: users need criteria to grade outputs, but grading outputs helps users define criteria. What is more, some criteria appears \emph{dependent} on the specific LLM outputs observed (rather than independent criteria that can be defined \emph{a priori}), raising serious questions for approaches that assume the independence of evaluation from observation of model outputs. We present our interface and implementation details, a comparison of our algorithm with a baseline approach, and implications for the design of future LLM evaluation assistants.
2024-04-18
ArXiv (preprint)
doi.org
arxiv.org
Who Validates the Validators? Aligning LLM-Assisted Evaluation of LLM Outputs with Human Preferences
Shreya Shankar
J.D. Zamfirescu-Pereira
Bjorn Hartmann
Aditya G Parameswaran
Ian Arawjo
Due to the cumbersome nature of human evaluation and limitations of code-based evaluation, Large Language Models (LLMs) are increasingly bei… (see more)ng used to assist humans in evaluating LLM outputs. Yet LLM-generated evaluators simply inherit all the problems of the LLMs they evaluate, requiring further human validation. We present a mixed-initiative approach to ``validate the validators'' -- aligning LLM-generated evaluation functions (be it prompts or code) with human requirements. Our interface, EvalGen, provides automated assistance to users in generating evaluation criteria and implementing assertions. While generating candidate implementations (Python functions, LLM grader prompts), EvalGen asks humans to grade a subset of LLM outputs; this feedback is used to select implementations that better align with user grades. A qualitative study finds overall support for EvalGen but underscores the subjectivity and iterative process of alignment. In particular, we identify a phenomenon we dub \emph{criteria drift}: users need criteria to grade outputs, but grading outputs helps users define criteria. What is more, some criteria appears \emph{dependent} on the specific LLM outputs observed (rather than independent criteria that can be defined \emph{a priori}), raising serious questions for approaches that assume the independence of evaluation from observation of model outputs. We present our interface and implementation details, a comparison of our algorithm with a baseline approach, and implications for the design of future LLM evaluation assistants.
2024-04-18
ArXiv (preprint)
doi.org
arxiv.org
Antagonistic AI
Alice Cai
Ian Arawjo
Elena L. Glassman
2024-02-12
ArXiv (preprint)
doi.org
arxiv.org
Antagonistic AI
Alice Cai
Ian Arawjo
Elena L. Glassman
The vast majority of discourse around AI development assumes that subservient,"moral"models aligned with"human values"are universally benefi… (see more)cial -- in short, that good AI is sycophantic AI. We explore the shadow of the sycophantic paradigm, a design space we term antagonistic AI: AI systems that are disagreeable, rude, interrupting, confrontational, challenging, etc. -- embedding opposite behaviors or values. Far from being"bad"or"immoral,"we consider whether antagonistic AI systems may sometimes have benefits to users, such as forcing users to confront their assumptions, build resilience, or develop healthier relational boundaries. Drawing from formative explorations and a speculative design workshop where participants designed fictional AI technologies that employ antagonism, we lay out a design space for antagonistic AI, articulating potential benefits, design techniques, and methods of embedding antagonistic elements into user experience. Finally, we discuss the many ethical challenges of this space and identify three dimensions for the responsible design of antagonistic AI -- consent, context, and framing.
2024-02-12
ArXiv (preprint)
doi.org
arxiv.org
Antagonistic AI
Alice Cai
Ian Arawjo
Elena L. Glassman
The vast majority of discourse around AI development assumes that subservient,"moral"models aligned with"human values"are universally benefi… (see more)cial -- in short, that good AI is sycophantic AI. We explore the shadow of the sycophantic paradigm, a design space we term antagonistic AI: AI systems that are disagreeable, rude, interrupting, confrontational, challenging, etc. -- embedding opposite behaviors or values. Far from being"bad"or"immoral,"we consider whether antagonistic AI systems may sometimes have benefits to users, such as forcing users to confront their assumptions, build resilience, or develop healthier relational boundaries. Drawing from formative explorations and a speculative design workshop where participants designed fictional AI technologies that employ antagonism, we lay out a design space for antagonistic AI, articulating potential benefits, design techniques, and methods of embedding antagonistic elements into user experience. Finally, we discuss the many ethical challenges of this space and identify three dimensions for the responsible design of antagonistic AI -- consent, context, and framing.
2024-02-12
ArXiv (preprint)
doi.org
arxiv.org
Antagonistic AI
Alice Cai
Ian Arawjo
Elena L. Glassman
2024-02-12
ArXiv (preprint)
doi.org
arxiv.org
Antagonistic AI
Alice Cai
Ian Arawjo
Elena L. Glassman
The vast majority of discourse around AI development assumes that subservient,"moral"models aligned with"human values"are universally benefi… (see more)cial -- in short, that good AI is sycophantic AI. We explore the shadow of the sycophantic paradigm, a design space we term antagonistic AI: AI systems that are disagreeable, rude, interrupting, confrontational, challenging, etc. -- embedding opposite behaviors or values. Far from being"bad"or"immoral,"we consider whether antagonistic AI systems may sometimes have benefits to users, such as forcing users to confront their assumptions, build resilience, or develop healthier relational boundaries. Drawing from formative explorations and a speculative design workshop where participants designed fictional AI technologies that employ antagonism, we lay out a design space for antagonistic AI, articulating potential benefits, design techniques, and methods of embedding antagonistic elements into user experience. Finally, we discuss the many ethical challenges of this space and identify three dimensions for the responsible design of antagonistic AI -- consent, context, and framing.
2024-02-12
ArXiv (preprint)
doi.org
arxiv.org
ChainBuddy: An AI Agent System for Generating LLM Pipelines
jingyue zhang
Ian Arawjo
2024-01-01
arXiv.org (preprint)
doi.org
arxiv.org
ChainBuddy: An AI-assisted Agent System for Helping Users Set up LLM Pipelines
jingyue zhang
Ian Arawjo
2024-01-01
UIST (Adjunct Volume) (published)
doi.org
ChainForge: An open-source visual programming environment for prompt engineering
Ian Arawjo
Priyan Vaithilingam
Martin Wattenberg
Elena Glassman
2023-10-29
Adjunct Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology (published)
doi.org
Notational Programming for Notebook Environments: A Case Study with Quantum Circuits
Ian Arawjo
Anthony DeArmas
Michael Roberts
Shrutarshi Basu
Tapan Parikh
We articulate a vision for computer programming that includes pen-based computing, a paradigm we term notational programming. Notational pro… (see more)gramming blurs contexts: certain typewritten variables can be referenced in handwritten notation and vice-versa. To illustrate this paradigm, we developed an extension, Notate, to computational notebooks which allows users to open drawing canvases within lines of code. As a case study, we explore quantum programming and designed a notation, Qaw, that extends quantum circuit notation with abstraction features, such as variable-sized wire bundles and recursion. Results from a usability study with novices suggest that users find our core interaction of implicit cross-context references intuitive, but suggests further improvements to debugging infrastructure, interface design, and recognition rates. Throughout, we discuss questions raised by the notational paradigm, including a shift from ‘recognition’ of notations to ‘reconfiguration’ of practices and values around programming, and from ‘sketching’ to writing and drawing, or what we call ‘notating.’
2022-10-28
ACM Symposium on User Interface Software and Technology (published)
doi.org