Manuel Tonneau
@manueltonneau.bsky.social
PhD candidate @oiioxford.bsky.social NLP, Computational Social Science @WorldBank
manueltonneau.com
manueltonneau.com
Pinned
Manuel Tonneau
@manueltonneau.bsky.social
· Jul 31
🏆 Thrilled to share that our HateDay paper has received an Outstanding Paper Award at #ACL2025
Big thanks to my wonderful co-authors: @deeliu97.bsky.social, Niyati, @computermacgyver.bsky.social, Sam, Victor, and @paul-rottger.bsky.social!
Thread 👇and data avail at huggingface.co/datasets/man...
Big thanks to my wonderful co-authors: @deeliu97.bsky.social, Niyati, @computermacgyver.bsky.social, Sam, Victor, and @paul-rottger.bsky.social!
Thread 👇and data avail at huggingface.co/datasets/man...
Reposted by Manuel Tonneau
ICYMI: Listen to @manueltonneau.bsky.social @oii.ox.ac.uk's interview with the SOEP podcast talking about his new research into hate speech, online platforms and disparities in content moderation across different European countries. Available here: bit.ly/4ntsiRU
Home - Somewhere On Earth Productions
SOMEWHERE ON EARTH PRODUCTIONS: We are here to connect technology and business to people and new possibilities.
bit.ly
October 1, 2025 at 1:46 PM
ICYMI: Listen to @manueltonneau.bsky.social @oii.ox.ac.uk's interview with the SOEP podcast talking about his new research into hate speech, online platforms and disparities in content moderation across different European countries. Available here: bit.ly/4ntsiRU
Reposted by Manuel Tonneau
🚨Hiring a fully funded (3.5 years) PhD for the @ldnsocmedobs.bsky.social to research social media and politics. Candidates should have quantitative/computational skills and/or be interested in content curation/moderation. UK home candidates only unfortunately. www.royalholloway.ac.uk/media/hquftp...
www.royalholloway.ac.uk
September 29, 2025 at 5:21 PM
🚨Hiring a fully funded (3.5 years) PhD for the @ldnsocmedobs.bsky.social to research social media and politics. Candidates should have quantitative/computational skills and/or be interested in content curation/moderation. UK home candidates only unfortunately. www.royalholloway.ac.uk/media/hquftp...
Reposted by Manuel Tonneau
📣 New Preprint!
Have you ever wondered what the political content in LLM's training data is? What are the political opinions expressed? What is the proportion of left- vs right-leaning documents in the pre- and post-training data? Do they correlate with the political biases reflected in models?
Have you ever wondered what the political content in LLM's training data is? What are the political opinions expressed? What is the proportion of left- vs right-leaning documents in the pre- and post-training data? Do they correlate with the political biases reflected in models?
September 29, 2025 at 2:54 PM
📣 New Preprint!
Have you ever wondered what the political content in LLM's training data is? What are the political opinions expressed? What is the proportion of left- vs right-leaning documents in the pre- and post-training data? Do they correlate with the political biases reflected in models?
Have you ever wondered what the political content in LLM's training data is? What are the political opinions expressed? What is the proportion of left- vs right-leaning documents in the pre- and post-training data? Do they correlate with the political biases reflected in models?
Reposted by Manuel Tonneau
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
Reposted by Manuel Tonneau
🚨 New paper alert 🚨 Using LLMs as data annotators, you can produce any scientific result you want. We call this **LLM Hacking**.
Paper: arxiv.org/pdf/2509.08825
Paper: arxiv.org/pdf/2509.08825
September 12, 2025 at 10:33 AM
🚨 New paper alert 🚨 Using LLMs as data annotators, you can produce any scientific result you want. We call this **LLM Hacking**.
Paper: arxiv.org/pdf/2509.08825
Paper: arxiv.org/pdf/2509.08825
Reposted by Manuel Tonneau
1/ 🚨 Big news 🚨 today we’re launching Tech for Open Minds (TOM) at @DukeU— a global program exploring how technology shapes open-mindedness, humility & polarization 🌍🧠
🔗https://sicss.io/stories/2025-08-18
🔗https://sicss.io/stories/2025-08-18
August 29, 2025 at 4:06 PM
1/ 🚨 Big news 🚨 today we’re launching Tech for Open Minds (TOM) at @DukeU— a global program exploring how technology shapes open-mindedness, humility & polarization 🌍🧠
🔗https://sicss.io/stories/2025-08-18
🔗https://sicss.io/stories/2025-08-18
Reposted by Manuel Tonneau
@themedialeader.bsky.social highlights new insights from @manueltonneau.bsky.social, @deeliu97.bsky.social, Prof. Ralph Schroeder + Prof. @computermacgyver.bsky.social, whose have found that 16mn EU-based X users “do not have moderators for their language.”
uk.themedialeader.com/social-platf...
uk.themedialeader.com/social-platf...
Social platforms' 'language blind spots' in content moderation bring brand safety concerns
A new study of recently mandated transparency data under the EU Digital Services Act found that millions of users of social platforms in the region post in languages without any human moderation.
uk.themedialeader.com
August 29, 2025 at 12:47 PM
@themedialeader.bsky.social highlights new insights from @manueltonneau.bsky.social, @deeliu97.bsky.social, Prof. Ralph Schroeder + Prof. @computermacgyver.bsky.social, whose have found that 16mn EU-based X users “do not have moderators for their language.”
uk.themedialeader.com/social-platf...
uk.themedialeader.com/social-platf...
Reposted by Manuel Tonneau
Millions of users are posting to social media and other platforms in languages with zero moderators, even within the EU.
That's the topline finding from an impressive new working paper leveraging newly mandated transparency data under the DSA led by @manueltonneau.bsky.social osf.io/preprints/so...
That's the topline finding from an impressive new working paper leveraging newly mandated transparency data under the DSA led by @manueltonneau.bsky.social osf.io/preprints/so...
OSF
osf.io
August 28, 2025 at 9:41 AM
Millions of users are posting to social media and other platforms in languages with zero moderators, even within the EU.
That's the topline finding from an impressive new working paper leveraging newly mandated transparency data under the DSA led by @manueltonneau.bsky.social osf.io/preprints/so...
That's the topline finding from an impressive new working paper leveraging newly mandated transparency data under the DSA led by @manueltonneau.bsky.social osf.io/preprints/so...
Social media platforms operate globally, but do they allocate human moderation equitably across languages?
Our new WP shows the answer is no:
-Millions of users post in languages with zero moderators
-Where mods exist, mod count relative to content volume varies widely across langs
osf.io/amfws
Our new WP shows the answer is no:
-Millions of users post in languages with zero moderators
-Where mods exist, mod count relative to content volume varies widely across langs
osf.io/amfws
August 28, 2025 at 8:46 AM
Social media platforms operate globally, but do they allocate human moderation equitably across languages?
Our new WP shows the answer is no:
-Millions of users post in languages with zero moderators
-Where mods exist, mod count relative to content volume varies widely across langs
osf.io/amfws
Our new WP shows the answer is no:
-Millions of users post in languages with zero moderators
-Where mods exist, mod count relative to content volume varies widely across langs
osf.io/amfws
Very cool piece by my colleague @antisomniac.bsky.social on how YouTube is used differently across languages. Worth a read!
I wrote an article about linguistic bias and the internet for the BBC, based on a paper @ze.vin, @ethanz.bsky.social, and I wrote comparing four language-specific samples of YouTube. www.bbc.com/future/artic...
How language is hiding the real internet from you
Most of the internet is out of your reach, but the barrier isn't just algorithms. In another language, the same platforms turn into whole other worlds.
www.bbc.com
August 13, 2025 at 7:31 PM
Very cool piece by my colleague @antisomniac.bsky.social on how YouTube is used differently across languages. Worth a read!
🏆 Thrilled to share that our HateDay paper has received an Outstanding Paper Award at #ACL2025
Big thanks to my wonderful co-authors: @deeliu97.bsky.social, Niyati, @computermacgyver.bsky.social, Sam, Victor, and @paul-rottger.bsky.social!
Thread 👇and data avail at huggingface.co/datasets/man...
Big thanks to my wonderful co-authors: @deeliu97.bsky.social, Niyati, @computermacgyver.bsky.social, Sam, Victor, and @paul-rottger.bsky.social!
Thread 👇and data avail at huggingface.co/datasets/man...
July 31, 2025 at 8:05 AM
🏆 Thrilled to share that our HateDay paper has received an Outstanding Paper Award at #ACL2025
Big thanks to my wonderful co-authors: @deeliu97.bsky.social, Niyati, @computermacgyver.bsky.social, Sam, Victor, and @paul-rottger.bsky.social!
Thread 👇and data avail at huggingface.co/datasets/man...
Big thanks to my wonderful co-authors: @deeliu97.bsky.social, Niyati, @computermacgyver.bsky.social, Sam, Victor, and @paul-rottger.bsky.social!
Thread 👇and data avail at huggingface.co/datasets/man...
Reposted by Manuel Tonneau
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
Reposted by Manuel Tonneau
New! Heading to #ACL2025NLP today? Hear from @oii.ox.ac.uk researchers presenting new research and sharing recent findings which aim to help address inequalities in natural language processing models. 1/4
July 28, 2025 at 9:19 AM
New! Heading to #ACL2025NLP today? Hear from @oii.ox.ac.uk researchers presenting new research and sharing recent findings which aim to help address inequalities in natural language processing models. 1/4
Reposted by Manuel Tonneau
Join @manueltonneau.bsky.social as he presents his co-authored paper ‘HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter’ this afternoon. Mon 28 July, 14.00-15.00. Hall A. 2/4
July 28, 2025 at 9:19 AM
Join @manueltonneau.bsky.social as he presents his co-authored paper ‘HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter’ this afternoon. Mon 28 July, 14.00-15.00. Hall A. 2/4
Excited to give an oral presentation of this work at #ACL2025 next Monday in Vienna! 🇦🇹 Catch me in Hall A from 2–3:30pm — would love to reconnect with familiar faces and get to know new ones!
Can we detect #hatespeech at scale on social media?
To answer this, we introduce 🤬HateDay🗓️, a global hate speech dataset representative of a day on Twitter.
The answer: not really! Detection perf is low and overestimated by traditional eval methods
arxiv.org/abs/2411.15462
🧵
To answer this, we introduce 🤬HateDay🗓️, a global hate speech dataset representative of a day on Twitter.
The answer: not really! Detection perf is low and overestimated by traditional eval methods
arxiv.org/abs/2411.15462
🧵
July 25, 2025 at 3:25 PM
Excited to give an oral presentation of this work at #ACL2025 next Monday in Vienna! 🇦🇹 Catch me in Hall A from 2–3:30pm — would love to reconnect with familiar faces and get to know new ones!
Reposted by Manuel Tonneau
🚨 New study!
We tested whether AI-generated messages – single static messages vs. conversations – can boost intent to screen for colorectal cancer.
Turns out: short, tailored AI messages outperform expert-written materials & match conversations, at a fraction of the time! 🧵👇
We tested whether AI-generated messages – single static messages vs. conversations – can boost intent to screen for colorectal cancer.
Turns out: short, tailored AI messages outperform expert-written materials & match conversations, at a fraction of the time! 🧵👇
July 14, 2025 at 2:06 PM
🚨 New study!
We tested whether AI-generated messages – single static messages vs. conversations – can boost intent to screen for colorectal cancer.
Turns out: short, tailored AI messages outperform expert-written materials & match conversations, at a fraction of the time! 🧵👇
We tested whether AI-generated messages – single static messages vs. conversations – can boost intent to screen for colorectal cancer.
Turns out: short, tailored AI messages outperform expert-written materials & match conversations, at a fraction of the time! 🧵👇
Reposted by Manuel Tonneau
Why did Grok suddenly start talking about “white genocide in South Africa” even if asked about baseball or cute dogs?
Because someone at Musk’s xAi deliberately did this, and we only found out because they were clumsy.
My piece on the real dangers of AI.
Gift link:
www.nytimes.com/2025/05/17/o...
Because someone at Musk’s xAi deliberately did this, and we only found out because they were clumsy.
My piece on the real dangers of AI.
Gift link:
www.nytimes.com/2025/05/17/o...
May 17, 2025 at 11:31 AM
Why did Grok suddenly start talking about “white genocide in South Africa” even if asked about baseball or cute dogs?
Because someone at Musk’s xAi deliberately did this, and we only found out because they were clumsy.
My piece on the real dangers of AI.
Gift link:
www.nytimes.com/2025/05/17/o...
Because someone at Musk’s xAi deliberately did this, and we only found out because they were clumsy.
My piece on the real dangers of AI.
Gift link:
www.nytimes.com/2025/05/17/o...
Reposted by Manuel Tonneau
New preprint with @jbakcoleman.bsky.social @lewan.bsky.social @randomwalker.bsky.social @orbenamy.bsky.social @lfoswaldo.bsky.social where we argue for a complex-system perspective to understand the causal effects of social media on society and for a triangulation of methods
arxiv.org/abs/2505.09254
arxiv.org/abs/2505.09254
Moving towards informative and actionable social media research
Social media is nearly ubiquitous in modern life, and concerns have been raised about its putative societal impacts, ranging from undermining mental health and exacerbating polarization to fomenting v...
arxiv.org
May 15, 2025 at 6:31 AM
New preprint with @jbakcoleman.bsky.social @lewan.bsky.social @randomwalker.bsky.social @orbenamy.bsky.social @lfoswaldo.bsky.social where we argue for a complex-system perspective to understand the causal effects of social media on society and for a triangulation of methods
arxiv.org/abs/2505.09254
arxiv.org/abs/2505.09254
Reposted by Manuel Tonneau
What do experts think about the potential negative impacts of social media on adolescent mental health?
We have a new consensus statement with 120 experts on this topic. Check it out to see where experts agree and where they think more evidence is needed!
We have a new consensus statement with 120 experts on this topic. Check it out to see where experts agree and where they think more evidence is needed!
Are #smartphones and #socialmedia harming a generation?
This is a hotly debated and often polarizing debate. So we surveyed over 120 experts on the topic to see where there was genuine consensus (or not), like experts have previous done for climate change.
See our paper: osf.io/preprints/ps...
This is a hotly debated and often polarizing debate. So we surveyed over 120 experts on the topic to see where there was genuine consensus (or not), like experts have previous done for climate change.
See our paper: osf.io/preprints/ps...
May 16, 2025 at 1:57 AM
What do experts think about the potential negative impacts of social media on adolescent mental health?
We have a new consensus statement with 120 experts on this topic. Check it out to see where experts agree and where they think more evidence is needed!
We have a new consensus statement with 120 experts on this topic. Check it out to see where experts agree and where they think more evidence is needed!
Excited to share that two of our papers got into ACL 2025! 🎉
📌 Main: HateDay: A Global Hate Speech Dataset Representative of Twitter (arxiv.org/abs/2411.15462)
📌 Findings: When Claims Evolve – Robustness to Misinformation Edits (arxiv.org/abs/2503.03417)
See you all in Vienna! 🇦🇹 #ACL2025 #NLProc
📌 Main: HateDay: A Global Hate Speech Dataset Representative of Twitter (arxiv.org/abs/2411.15462)
📌 Findings: When Claims Evolve – Robustness to Misinformation Edits (arxiv.org/abs/2503.03417)
See you all in Vienna! 🇦🇹 #ACL2025 #NLProc
May 16, 2025 at 6:11 AM
Excited to share that two of our papers got into ACL 2025! 🎉
📌 Main: HateDay: A Global Hate Speech Dataset Representative of Twitter (arxiv.org/abs/2411.15462)
📌 Findings: When Claims Evolve – Robustness to Misinformation Edits (arxiv.org/abs/2503.03417)
See you all in Vienna! 🇦🇹 #ACL2025 #NLProc
📌 Main: HateDay: A Global Hate Speech Dataset Representative of Twitter (arxiv.org/abs/2411.15462)
📌 Findings: When Claims Evolve – Robustness to Misinformation Edits (arxiv.org/abs/2503.03417)
See you all in Vienna! 🇦🇹 #ACL2025 #NLProc
Reposted by Manuel Tonneau
Just published at CHI ’25: How Commercial Content Moderation APIs over- and under-moderate hate speech dl.acm.org/doi/10.1145/... w/ @dawiet.bsky.social ky.social, Amin Oueslati @hheuer.bsky.social cial @dimitristaufer.bsky.social al Lena Pohlmann. 🧵
Lost in Moderation: How Commercial Content Moderation APIs Over- and Under-Moderate Group-Targeted Hate Speech and Linguistic Variations | Proceedings of the 2025 CHI Conference on Human Factors in Co...
dl.acm.org
May 12, 2025 at 8:25 PM
Just published at CHI ’25: How Commercial Content Moderation APIs over- and under-moderate hate speech dl.acm.org/doi/10.1145/... w/ @dawiet.bsky.social ky.social, Amin Oueslati @hheuer.bsky.social cial @dimitristaufer.bsky.social al Lena Pohlmann. 🧵
Reposted by Manuel Tonneau
🎓For today's lab seminar, it was a pleasure to have
@manueltonneau.bsky.social presenting "HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter" ✨
#NLProc #hatespeech
@manueltonneau.bsky.social presenting "HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter" ✨
#NLProc #hatespeech
May 9, 2025 at 4:10 PM
🎓For today's lab seminar, it was a pleasure to have
@manueltonneau.bsky.social presenting "HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter" ✨
#NLProc #hatespeech
@manueltonneau.bsky.social presenting "HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter" ✨
#NLProc #hatespeech
Reposted by Manuel Tonneau
⏰ Early‑bird registration for #IC2S2’25 in Norrköping ends May 9—lock in your spot (and the discount) today: www.ic2s2-2025.org/register/
IC2S2'25 Norrköping
www.ic2s2-2025.org
May 6, 2025 at 10:28 AM
⏰ Early‑bird registration for #IC2S2’25 in Norrköping ends May 9—lock in your spot (and the discount) today: www.ic2s2-2025.org/register/
Had a blast presenting my research on hate speech moderation on social media and the potential of human-AI collaboration to improve it, thanks a lot for the invite @hertiedatascience.bsky.social ! Check out this blog post for details on our preliminary results: www.hertie-school.org/en/datascien...
May 8, 2025 at 12:30 PM
Had a blast presenting my research on hate speech moderation on social media and the potential of human-AI collaboration to improve it, thanks a lot for the invite @hertiedatascience.bsky.social ! Check out this blog post for details on our preliminary results: www.hertie-school.org/en/datascien...
Reposted by Manuel Tonneau
🚨 New preprint on AI persuasion and public health 🚨
A 3-min conversation with GPT-4o nudged HPV-vax-hesitant parents (who obv knew it was AI & consented!)—BUT reading standard public-health material still outperformed chatbots in impact and longevity. Details below 👇
A 3-min conversation with GPT-4o nudged HPV-vax-hesitant parents (who obv knew it was AI & consented!)—BUT reading standard public-health material still outperformed chatbots in impact and longevity. Details below 👇
April 30, 2025 at 9:40 AM
🚨 New preprint on AI persuasion and public health 🚨
A 3-min conversation with GPT-4o nudged HPV-vax-hesitant parents (who obv knew it was AI & consented!)—BUT reading standard public-health material still outperformed chatbots in impact and longevity. Details below 👇
A 3-min conversation with GPT-4o nudged HPV-vax-hesitant parents (who obv knew it was AI & consented!)—BUT reading standard public-health material still outperformed chatbots in impact and longevity. Details below 👇