Manoel Horta Ribeiro
@manoelhortaribeiro.bsky.social
Assistant Professor @ Princeton
Previously: EPFL 🇨🇭, UFMG 🇧🇷
Interests: Computational Social Science, Platforms, GenAI, Moderation
Previously: EPFL 🇨🇭, UFMG 🇧🇷
Interests: Computational Social Science, Platforms, GenAI, Moderation
Reposted by Manoel Horta Ribeiro
Have been waiting for this paper to come out ever since Hongyao told me about it -- more academics should be taking advantage of such open data to answer important questions
Coverage of the paper by @ellebeyoud.bsky.social of Bloomberg here (gift link):
www.bloomberg.com/news/article...
www.bloomberg.com/news/article...
November 7, 2025 at 1:45 PM
Have been waiting for this paper to come out ever since Hongyao told me about it -- more academics should be taking advantage of such open data to answer important questions
Reposted by Manoel Horta Ribeiro
There’s plenty of evidence for political bias in LLMs, but very few evals reflect realistic LLM use cases — which is where bias actually matters.
IssueBench, our attempt to fix this, is accepted at TACL, and I will be at #EMNLP2025 next week to talk about it!
New results 🧵
IssueBench, our attempt to fix this, is accepted at TACL, and I will be at #EMNLP2025 next week to talk about it!
New results 🧵
Are LLMs biased when they write about political issues?
We just released IssueBench – the largest, most realistic benchmark of its kind – to answer this question more robustly than ever before.
Long 🧵with spicy results 👇
We just released IssueBench – the largest, most realistic benchmark of its kind – to answer this question more robustly than ever before.
Long 🧵with spicy results 👇
October 29, 2025 at 4:12 PM
There’s plenty of evidence for political bias in LLMs, but very few evals reflect realistic LLM use cases — which is where bias actually matters.
IssueBench, our attempt to fix this, is accepted at TACL, and I will be at #EMNLP2025 next week to talk about it!
New results 🧵
IssueBench, our attempt to fix this, is accepted at TACL, and I will be at #EMNLP2025 next week to talk about it!
New results 🧵
The debate over “LLMs as annotators” feels familiar: excitement, backlash, and anxiety about bad science. My take in a new blogpost is that LLMs don’t break measurement; they expose how fragile it already was.
doomscrollingbabel.manoel.xyz/p/labeling-d...
doomscrollingbabel.manoel.xyz/p/labeling-d...
Labeling Data with Language Models: Trick or Treat?
Large language models are now labeling data for us.
doomscrollingbabel.manoel.xyz
October 25, 2025 at 6:29 PM
The debate over “LLMs as annotators” feels familiar: excitement, backlash, and anxiety about bad science. My take in a new blogpost is that LLMs don’t break measurement; they expose how fragile it already was.
doomscrollingbabel.manoel.xyz/p/labeling-d...
doomscrollingbabel.manoel.xyz/p/labeling-d...
Reposted by Manoel Horta Ribeiro
Graze is doing great work in supporting the entire ATProto Ecosystem, and their latest newsletter is a great illustration. Also featuring an interview with @sjgreenwood.bsky.social about Paper Skygest (bsky.app/profile/pape...)
We’ve started describing ATProto as the new standard gauge — a shared track that finally lets social apps move together instead of apart. Our latest Graze newsletter looks at what that means in practice (plus a great chat with @sjgreenwood.bsky.social ).
graze-newsletter.beehiiv.com/p/the-new-st...
graze-newsletter.beehiiv.com/p/the-new-st...
The new standard gauge
Graze is Bluesky. By You.
graze-newsletter.beehiiv.com
October 21, 2025 at 12:39 PM
Graze is doing great work in supporting the entire ATProto Ecosystem, and their latest newsletter is a great illustration. Also featuring an interview with @sjgreenwood.bsky.social about Paper Skygest (bsky.app/profile/pape...)
Reposted by Manoel Horta Ribeiro
Join us as postdoc at the Inequality Discourse Observatory at the University of Konstanz: stellen.uni-konstanz.de/jobposting/7...
We will do epic research between Linguistics and Computational Social Science at the Cluster of Politics of Inequality. Feel free to DM if you have any questions.
We will do epic research between Linguistics and Computational Social Science at the Cluster of Politics of Inequality. Feel free to DM if you have any questions.
One postdoctoral Research Position
Deadline: November 15th, 2025
stellen.uni-konstanz.de
October 13, 2025 at 3:06 PM
Join us as postdoc at the Inequality Discourse Observatory at the University of Konstanz: stellen.uni-konstanz.de/jobposting/7...
We will do epic research between Linguistics and Computational Social Science at the Cluster of Politics of Inequality. Feel free to DM if you have any questions.
We will do epic research between Linguistics and Computational Social Science at the Cluster of Politics of Inequality. Feel free to DM if you have any questions.
Computer Science is no longer just about building systems or proving theorems--it's about observation and experiments.
In my latest blog post, I argue it’s time we had our own "Econometrics," a discipline devoted to empirical rigor.
doomscrollingbabel.manoel.xyz/p/the-missin...
In my latest blog post, I argue it’s time we had our own "Econometrics," a discipline devoted to empirical rigor.
doomscrollingbabel.manoel.xyz/p/the-missin...
October 5, 2025 at 4:07 PM
Computer Science is no longer just about building systems or proving theorems--it's about observation and experiments.
In my latest blog post, I argue it’s time we had our own "Econometrics," a discipline devoted to empirical rigor.
doomscrollingbabel.manoel.xyz/p/the-missin...
In my latest blog post, I argue it’s time we had our own "Econometrics," a discipline devoted to empirical rigor.
doomscrollingbabel.manoel.xyz/p/the-missin...
Reposted by Manoel Horta Ribeiro
🌱✨ Life update: I just started my PhD at Princeton University!
I will be supervised by @manoelhortaribeiro.bsky.social and affiliated with Princeton CITP.
It's only been a month, but the energy feels amazing —very grateful for such a welcoming community. Excited for what’s ahead! 🚀
I will be supervised by @manoelhortaribeiro.bsky.social and affiliated with Princeton CITP.
It's only been a month, but the energy feels amazing —very grateful for such a welcoming community. Excited for what’s ahead! 🚀
October 3, 2025 at 5:56 PM
🌱✨ Life update: I just started my PhD at Princeton University!
I will be supervised by @manoelhortaribeiro.bsky.social and affiliated with Princeton CITP.
It's only been a month, but the energy feels amazing —very grateful for such a welcoming community. Excited for what’s ahead! 🚀
I will be supervised by @manoelhortaribeiro.bsky.social and affiliated with Princeton CITP.
It's only been a month, but the energy feels amazing —very grateful for such a welcoming community. Excited for what’s ahead! 🚀
Social media feeds today are optimized for engagement, often leading to misalignment between users' intentions and technology use.
In a new paper, we introduce Bonsai, a tool to create feeds based on stated preferences, rather than predicted engagement.
arxiv.org/abs/2509.10776
In a new paper, we introduce Bonsai, a tool to create feeds based on stated preferences, rather than predicted engagement.
arxiv.org/abs/2509.10776
September 16, 2025 at 1:24 PM
Social media feeds today are optimized for engagement, often leading to misalignment between users' intentions and technology use.
In a new paper, we introduce Bonsai, a tool to create feeds based on stated preferences, rather than predicted engagement.
arxiv.org/abs/2509.10776
In a new paper, we introduce Bonsai, a tool to create feeds based on stated preferences, rather than predicted engagement.
arxiv.org/abs/2509.10776
I'm thinking a lot about the debate around whether LMs can "understand." Enter a new thought experiment I call "the lean Chinese room." IMO, it shows how the pressure of compression and speed blurs the line between symbol-shuffling and understanding.
doomscrollingbabel.manoel.xyz/p/the-lean-c...
doomscrollingbabel.manoel.xyz/p/the-lean-c...
August 25, 2025 at 4:30 PM
I'm thinking a lot about the debate around whether LMs can "understand." Enter a new thought experiment I call "the lean Chinese room." IMO, it shows how the pressure of compression and speed blurs the line between symbol-shuffling and understanding.
doomscrollingbabel.manoel.xyz/p/the-lean-c...
doomscrollingbabel.manoel.xyz/p/the-lean-c...
Whether or not AIs “truly” understand is less important than how they reshape jobs, politics, and culture.
My new post: why the “Human of the Gaps” fallacy distracts us from governing real-world AI impacts.
doomscrollingbabel.manoel.xyz/p/the-retrea...
My new post: why the “Human of the Gaps” fallacy distracts us from governing real-world AI impacts.
doomscrollingbabel.manoel.xyz/p/the-retrea...
August 17, 2025 at 2:58 PM
Whether or not AIs “truly” understand is less important than how they reshape jobs, politics, and culture.
My new post: why the “Human of the Gaps” fallacy distracts us from governing real-world AI impacts.
doomscrollingbabel.manoel.xyz/p/the-retrea...
My new post: why the “Human of the Gaps” fallacy distracts us from governing real-world AI impacts.
doomscrollingbabel.manoel.xyz/p/the-retrea...
Creators pour years into building a following, but in a growing underground market, you can simply buy accounts and inherit their audience.
In our new pre-print, we find this practice of repurposing accounts to be prevalent and consequential on YouTube!
arxiv.org/abs/2507.16045
In our new pre-print, we find this practice of repurposing accounts to be prevalent and consequential on YouTube!
arxiv.org/abs/2507.16045
July 30, 2025 at 8:29 PM
Creators pour years into building a following, but in a growing underground market, you can simply buy accounts and inherit their audience.
In our new pre-print, we find this practice of repurposing accounts to be prevalent and consequential on YouTube!
arxiv.org/abs/2507.16045
In our new pre-print, we find this practice of repurposing accounts to be prevalent and consequential on YouTube!
arxiv.org/abs/2507.16045
🚨 Study alert!
We've developed a prototype that enables you to create personalized social media feeds using natural language rules and import them into Bluesky.
Join our study to try out the tool, give us your feedback, and earn a $50 gift card!
Sign up: forms.gle/MkSGKzxDfBEc...
Please share :)
We've developed a prototype that enables you to create personalized social media feeds using natural language rules and import them into Bluesky.
Join our study to try out the tool, give us your feedback, and earn a $50 gift card!
Sign up: forms.gle/MkSGKzxDfBEc...
Please share :)
July 22, 2025 at 1:23 PM
🚨 Study alert!
We've developed a prototype that enables you to create personalized social media feeds using natural language rules and import them into Bluesky.
Join our study to try out the tool, give us your feedback, and earn a $50 gift card!
Sign up: forms.gle/MkSGKzxDfBEc...
Please share :)
We've developed a prototype that enables you to create personalized social media feeds using natural language rules and import them into Bluesky.
Join our study to try out the tool, give us your feedback, and earn a $50 gift card!
Sign up: forms.gle/MkSGKzxDfBEc...
Please share :)
In a new blogpost, I argue GenAI as is won't "eat social media," because of:
- authenticity
- user preferences
- platforms being able to handle content at scale
doomscrollingbabel.manoel.xyz/publish/post...
- authenticity
- user preferences
- platforms being able to handle content at scale
doomscrollingbabel.manoel.xyz/publish/post...
July 5, 2025 at 11:36 AM
In a new blogpost, I argue GenAI as is won't "eat social media," because of:
- authenticity
- user preferences
- platforms being able to handle content at scale
doomscrollingbabel.manoel.xyz/publish/post...
- authenticity
- user preferences
- platforms being able to handle content at scale
doomscrollingbabel.manoel.xyz/publish/post...
In a new blog post, I...
- show that AI-generated videos are quite expensive to create
- argue that even if they weren't, we still would not drown in it.
doomscrollingbabel.manoel.xyz/p/ai-generat...
- show that AI-generated videos are quite expensive to create
- argue that even if they weren't, we still would not drown in it.
doomscrollingbabel.manoel.xyz/p/ai-generat...
July 1, 2025 at 7:56 PM
In a new blog post, I...
- show that AI-generated videos are quite expensive to create
- argue that even if they weren't, we still would not drown in it.
doomscrollingbabel.manoel.xyz/p/ai-generat...
- show that AI-generated videos are quite expensive to create
- argue that even if they weren't, we still would not drown in it.
doomscrollingbabel.manoel.xyz/p/ai-generat...
A good chunk of the AI-generated content we see right now is paid for by VC money. E.g., Leonardo.ai, a popular image/video creator, offers Veo3 **cheaper than Google**! It is like early-day Uber.
July 1, 2025 at 6:23 PM
A good chunk of the AI-generated content we see right now is paid for by VC money. E.g., Leonardo.ai, a popular image/video creator, offers Veo3 **cheaper than Google**! It is like early-day Uber.
I’m stoked to have received two honorable mention awards at @icwsm.bsky.social ! With work led by Peppe, thorsten and Maciej! Sad I couldn’t attend this year but looking very much forward to catching up with this community next year!!!
June 26, 2025 at 1:54 PM
I’m stoked to have received two honorable mention awards at @icwsm.bsky.social ! With work led by Peppe, thorsten and Maciej! Sad I couldn’t attend this year but looking very much forward to catching up with this community next year!!!
Reposted by Manoel Horta Ribeiro
Excellent talk by Hayoung in the Political Information and Opinion Dynamics panel at #ICWSM
How does YouTube’s search algorithm handle COVID🦠 misinfo in the United States🇺🇸(US) and South Africa🇿🇦(SA)?
In our #icwsm '25 paper w/ @prerna6.bsky.social @tanumitra.bsky.social, we found bots in SA received significantly more misinfo in top-10 search results, which accounts for 95% of user traffic
In our #icwsm '25 paper w/ @prerna6.bsky.social @tanumitra.bsky.social, we found bots in SA received significantly more misinfo in top-10 search results, which accounts for 95% of user traffic
June 24, 2025 at 12:35 PM
Excellent talk by Hayoung in the Political Information and Opinion Dynamics panel at #ICWSM
This @icwsm.bsky.social paper fine-tunes Temporal Adapters for Llama3 8B on timelines from a panel of British Twitter users, finding that the fine-tuned LLMs retrieve collective emotions! Cool extension of simulation work to the longitudinal scenario:
ojs.aaai.org/index.php/IC...
cc @dgarcia.eu
ojs.aaai.org/index.php/IC...
cc @dgarcia.eu
June 15, 2025 at 6:47 PM
This @icwsm.bsky.social paper fine-tunes Temporal Adapters for Llama3 8B on timelines from a panel of British Twitter users, finding that the fine-tuned LLMs retrieve collective emotions! Cool extension of simulation work to the longitudinal scenario:
ojs.aaai.org/index.php/IC...
cc @dgarcia.eu
ojs.aaai.org/index.php/IC...
cc @dgarcia.eu
In a new (very short) post, I wonder about whether the inability to forget harms large reasoning models
doomscrollingbabel.manoel.xyz/p/chatgpt-th...
doomscrollingbabel.manoel.xyz/p/chatgpt-th...
ChatGPT the Memorious
In “Funes the Memorious,” Argentine writer Jorge Luis Borges depicts a series of fictional encounters with Ireneo Funes, an Uruguayan teenager.
doomscrollingbabel.manoel.xyz
June 15, 2025 at 2:22 PM
In a new (very short) post, I wonder about whether the inability to forget harms large reasoning models
doomscrollingbabel.manoel.xyz/p/chatgpt-th...
doomscrollingbabel.manoel.xyz/p/chatgpt-th...
Reposted by Manoel Horta Ribeiro
Laypeople often learn about science from expert explanations & those explanations often contain JARGON. Does jargon make explanations better or worse? In a paper out today in Nature Human Behaviour, @cruzf.bsky.social and I find that jargon can support illusions of understanding...
June 12, 2025 at 12:19 PM
Laypeople often learn about science from expert explanations & those explanations often contain JARGON. Does jargon make explanations better or worse? In a paper out today in Nature Human Behaviour, @cruzf.bsky.social and I find that jargon can support illusions of understanding...
This paper characterizes (self-described) TikTok addiction with surveys and (n=107) data donations. One interesting finding is that watching positive videos predicts addiction, even controlling for session number and session time.
ojs.aaai.org/index.php/IC...
cc: @ingmarweber.de
ojs.aaai.org/index.php/IC...
cc: @ingmarweber.de
June 12, 2025 at 12:51 PM
This paper characterizes (self-described) TikTok addiction with surveys and (n=107) data donations. One interesting finding is that watching positive videos predicts addiction, even controlling for session number and session time.
ojs.aaai.org/index.php/IC...
cc: @ingmarweber.de
ojs.aaai.org/index.php/IC...
cc: @ingmarweber.de
This @chi.acm.org paper compares 1-hour, 1-day, and 3-day suspensions with two massive experiments on Roblox (N=700k+). TL;DR Longer suspensions work better.
I hope this gets the recognition it deserves. We need more 'basic' studies on content moderation.
dl.acm.org/doi/10.1145/...
I hope this gets the recognition it deserves. We need more 'basic' studies on content moderation.
dl.acm.org/doi/10.1145/...
June 9, 2025 at 5:12 PM
This @chi.acm.org paper compares 1-hour, 1-day, and 3-day suspensions with two massive experiments on Roblox (N=700k+). TL;DR Longer suspensions work better.
I hope this gets the recognition it deserves. We need more 'basic' studies on content moderation.
dl.acm.org/doi/10.1145/...
I hope this gets the recognition it deserves. We need more 'basic' studies on content moderation.
dl.acm.org/doi/10.1145/...
This @chi.acm.org paper surveys 319 knowledge workers. The authors find that GenAI reduces cog load for some tasks (info search) but creates new tasks (verification). This made me wonder to what extent automating search will create "rich-get-richer" effects.
dl.acm.org/doi/10.1145/...
dl.acm.org/doi/10.1145/...
June 6, 2025 at 1:38 PM
This @chi.acm.org paper surveys 319 knowledge workers. The authors find that GenAI reduces cog load for some tasks (info search) but creates new tasks (verification). This made me wonder to what extent automating search will create "rich-get-richer" effects.
dl.acm.org/doi/10.1145/...
dl.acm.org/doi/10.1145/...
Reposted by Manoel Horta Ribeiro
Before that, I'll spend 2025-2026 as a postdoc at Princeton's CITP @princetoncitp.bsky.social. I'm excited to conduct research about informedness and mental health on social media with @manoelhortaribeiro.bsky.social and @andyguess.com.
June 4, 2025 at 5:27 PM
Before that, I'll spend 2025-2026 as a postdoc at Princeton's CITP @princetoncitp.bsky.social. I'm excited to conduct research about informedness and mental health on social media with @manoelhortaribeiro.bsky.social and @andyguess.com.