Ada Tur

I'm an incoming PhD student in Computer Science at the University of Illinois at Urbana-Champaign, where I am very fortunate to be advised by Professor Heng Ji and supported by the NSF Graduate Research Fellowship and UIUC Surge Fellowship. I earned my undergraduate degrees in Computer Science and Linguistics at McGill University in Montreal, QC, where I also conducted undergraduate research with McGillNLP and Mila under Professor Siva Reddy.

I also used to host shows for CKUT 90.3FM!

My academic research is primarily concerned with understanding if computational intelligence can actually understand the human world, particularly through the lens of language acquisition and grounding. I'm also quite concerned about the safety and reliability of such models, especially for agentic usages.

Previously, I was a research intern at Mila under Professor Mirco Ravanelli, where I worked on the SpeechBrain project implementing neural rescoring for ASR systems. I was also the co-president of the McGill AI Society, and on the Podcast Team prior to that. I was also a AAAI UC Scholar in 2024! I also worked under Professor David Traum in the USC Institute for Creative Technologies researching human-computer interaction, as well as conducting research at CyVision on computer vision for driving assistance systems.

CV / Scholar / BlueSky / Github

News

<April 13, 2026> I was awarded the National Science Foundation's Graduate Research Fellowship!

<March 28, 2026> I'll be continuing my academic career in the PhD program at the University of Illinois at Urbana Champaign, working with Professor Heng Ji!

<May 1, 2025> "SafeArena: Evaluating the Safety of Autonomous Web Agents" was accepted at ICML 2025! Come check out our poster in Vancouver!!

<April 20, 2025> "Language Models Largely Exhibit Human-like Constituent Ordering Preferences" received the SAC Award for Linguistic Theories Track! 🏆

<March 9, 2025> "Language Models Largely Exhibit Human-like Constituent Ordering Preferences" was accepted for oral presentation at NAACL main conference, 2025! See you in ABQ!!

Research

	SafeArena: Evaluating the Safety of Autonomous Web Agents Ada D. Tur, Nicholas Meade, Xing Han Lù, Alejandra Zambrano, Arkil Patel, Esin Durmus, Spandana Gella, Karolina Stańczak, Siva Reddy Accepted at ICML 2025!* code / paper / website / leaderboard Agents like OpenAI Operator can solve complex computer tasks, but what happens when users use them to cause harm, e.g. automate hate speech and spread misinformation?
	Language Models Largely Exhibit Human-like Constituent Ordering Preferences Ada D. Tur, Gaurav Kamath, Siva Reddy, NAACL 2025 Main Conference Area Chair Award! code / paper / data Investigating constituent ordering behaviors and motivations in large language models and humans.
	ProGRes: Prompted Generative Rescoring on ASR n-Best Ada D. Tur, Adel Moumen, Mirco Ravanelli, IEEE Spoken Language Technology Workshop 2024 code / preprint Enhancing the performance of automatic speech recognition with large instruction-tuned language models.
	Deep Learning for Style Transfer and Experimentation with Audio Effects and Music Creation Ada D. Tur AAAI Undergraduate Consortium 2024 paper / poster A proposal for a set of Music+AI methods that serves to assist with the writing of and melodies, modelling and transferring of timbres, applying a wide variety of audio effects, including research into experimental audio effects, and production of audio samples using style transfers
	President Botrick: An Analysis of Deep Learning-Based Conversational AI Models to Identify and Create Influential Political Speeches Ada D. Tur, Julia Hirschberg AAAI Workshop for AI and Diplomacy 2023 github / paper Exploring the defining qualities of natural language that are considered influential and charismatic in the context of political speech using LLMs.
	Comparing Approaches to Language Understanding for Human-Robot Dialogue: An Error Taxonomy and Analysis Ada D. Tur, David R. Traum Language Resources and Evaluation Conference 2022 github / paper A comparison between relevance-based classification and generative transformers for natural language understanding in a human-robot interaction domain.
	ML‐Based Eye Tracking for Augmented Reality Heads‐Up Displays (AR HUDs) Ada D. Tur, Deniz Yaralioglu, Cemalettin Yilmaz SID International Symposium 2021 paper 3D Augmented Reality (AR) Heads‐up Displays (HUDs) have the potential of overlaying virtual objects at the correct locations with accurate motion parallax. Accurate overlays require tracking the pupils of the driver's eyes. We developed an ML‐based pupil tracking system based on a convolutional neural network (CNN) to find the precise location of the pupils.

Ada Tur, 2025 | Credits to Jon Barron