PhD student @MIT working on behavioral machine learning
Cambridge, MA
Joined March 2016
- Tweets117
- Following451
- Followers250
- Likes349
Pinned Tweet
New paper in @PNASNews: we find that AI agents are sensitive to nudges, often more than humans.
Defaults, suggestions, and highlighted information can alter model behavior substantially—for both better and worse.
We believe this has important implications for agent design and delegation.
Work w/ @nikhilsinghmus @PattieMaes
pnas.org/doi/full/10.1073/pn…
Manuel Cherep retweeted
We won best paper at the COLM Workshop on Agent Behavior for our paper: AI Assistants Overassist!
Congratulations @verona_teo, @jainraghav10! Check out the preprint arxiv.org/abs/2607.21306
The Workshop on Agent Behavior is tomorrow starting at 8:30 AM in Continental Ballroom 4! If you’re @COLM_conf come check it out. We have some amazing speakers giving very different perspectives on how to study agents.
✨Announcing the first Workshop on Agent Behavior @COLM_conf 2026 (Oct 9, San Francisco 🌅)
aiagentbehavior.com/
We invite two types of contributions: (i) papers, and (ii) benchmark proposals. We are also seeking reviewers. More details below!
I'm at COLM this week! If you're interested in chatting about AI agent vs human behavior, feel free to reach out :) I will also be on campus at Stanford (Oct 6-7) and Berkeley (Oct 8)
Manuel Cherep retweeted
Students sometimes ask me if it still makes sense, in this accelerating age, to pursue a PhD in AI.
Perhaps counterintuitively, I think it's a great time to do so.
I wrote up some thoughts on this here: web.mit.edu/phillipi/www/wri…
Manuel Cherep retweeted
📢 Just wrapped up some cool agent evaluation work for ICLR? Or are you simply interested in learning more about the evals landscape? Come to our Agent Evaluation Science Fall Symposium in New York City on November 20, 2026 where we bring together researchers & practitioners!
The day will focus on how agents should be scientifically evaluated, with topics spanning measurement and benchmark design, reliability, safety, and real-world performance. If you have some exciting work in this space, please consider submitting it for poster presentation!
Abstract registration deadline: October 20
Submission deadline: October 25
OpenReview: openreview.net/group?id=eval…
We are also excited to welcome a strong lineup of speakers, with more to be announced. Confirmed so far: @yoavartzi (Cornell Tech), @jaseweston (Meta AI), @nouhadziri (Cohere Labs), @PeterHndrsn (Princeton University), @walesalaudeen96 (Microsoft Research), @criticalneuro (Microsoft Research), and @saurabhjha2010 (IBM Research).
Important: You do not need to submit work in order to join! Just register through the Luma link below. If space becomes limited however, we will prioritize authors with accepted submissions.
Register: luma.com/cxmx1g4n
Details: evalscience.org/2026fall-nyc…
Huge thanks to the entire organizing team, including @yininghua, @HongbinNLP, @manuelcherep, @dubrovnik07, @HuaYilun, @ishapuri101, @liuzhuang1234, Lori Chibnik, @vbingliu, @CsabaSzepesvari, @PeterStone_TX, @sirrice, @ang3linawang, and @nikhilsinghmus for putting this together!
Feel free to reach out with any questions and we hope to see many of you there!
Manuel Cherep retweeted
Happy to share that our paper "Mecha-nudges for Machines" has been accepted as a spotlight at NeurIPS 2026!
Last week, I was invited to present our @icmlconf 2026 paper “Visual Persuasion: What Influences Decisions of Vision-Language Models?” at the ECCV VisPer Workshop.
We created a visual prompt optimization method (CVPO), in which we iteratively edit images (e.g., w/ @NanoBanana 🍌) to maximize the feedback given by VLM judges. While this method can be used for lots of applications, we use it to discover factors that persuade agents (and, as a proxy, people) while making decisions. More details and examples below!
Work w/ Pranav M R*, @PattieMaes, and @nikhilsinghmus
This is the broader point of our @icmlconf 2026 position paper “Position: Behavioral Systems Require Behavioral Tests.” There, we argue agent evals should look beyond the outcome, at the agents’ chains of actions, how they interact with the environment, and how they interact with each other.
openreview.net/pdf?id=jftMWd…
If you’re interested in hearing more about how agents behave, come to our Workshop on Agent Behavior at COLM 2026, in SF (Oct 9).
xitter.cf/manuelcherep/status/20…
✨Announcing the first Workshop on Agent Behavior @COLM_conf 2026 (Oct 9, San Francisco 🌅)
aiagentbehavior.com/
We invite two types of contributions: (i) papers, and (ii) benchmark proposals. We are also seeking reviewers. More details below!
Manuel Cherep retweeted
I'm hiring a predoc to work with me on AI x econ at UChicago!
This role is ideal for someone with an econ background looking to transition into AI research, or someone with an AI background looking to acquire more depth in economics.
This is not a standard predoc position---you'll be given the research freedom of a PhD student, and be a (co-)first author on any papers we write together.
Ideally, we'll work on something in the space of: post-training, AI alignment, AI governance, and the economics of AI. You aren't limited to working on these topics either---a top priority of mine is to find a topic that we're both excited out, even if it's further afield. To get a sense of the work I do, see the paper I wrote with my current predoc on mecha-nudges arxiv.org/abs/2603.23433 (Outstanding Paper at ICML's TAIGR workshop and Spotlight and EC's Game Theory with LLMs workshop).
Manuel Cherep retweeted
Congratulations to my awesome student @yujintang99 for leading this work on isolating and measuring visual memory in AI agents, now accepted at #EMNLP 2026.
Manuel Cherep retweeted
Thrilled to share I'll be joining UC Berkeley next year as an assistant professor in @UCBStatistics and affiliated with @Berkeley_EECS.
My research will build methods to test/improve the implicit world models of AI systems, so that they reflect reality and human understanding.
