Elinor🎗️
@elinorpd.bsky.social
MIT // researching fairness, equity, & pluralistic alignment in LLMs
previously @ MIT media lab, mila / mcgill
i like language and dogs and plants and ultimate frisbee and baking and sunsets
https://elinorp-d.github.io
previously @ MIT media lab, mila / mcgill
i like language and dogs and plants and ultimate frisbee and baking and sunsets
https://elinorp-d.github.io
Pinned
Elinor🎗️
@elinorpd.bsky.social
· Nov 3
What makes dialogue 💬 constructive 🫂?
We address this question in our #EMNLP2025 paper investigating how **responsivity** can characterize conversation quality ✨
Brandon Roy will be presenting our work (Oral) on Nov 7, Room A109 at 10:30.
🧵👇
https://aclanthology.org/2025.emnlp-main.1798/
We address this question in our #EMNLP2025 paper investigating how **responsivity** can characterize conversation quality ✨
Brandon Roy will be presenting our work (Oral) on Nov 7, Room A109 at 10:30.
🧵👇
https://aclanthology.org/2025.emnlp-main.1798/
Reposted by Elinor🎗️
Most LLM evals use API calls or offline inference, testing models in a memory-less silo. Our new Patterns paper shows this misses how LLMs actually behave in real user interfaces, where personalization and interaction history shape responses: arxiv.org/abs/2509.19364
December 12, 2025 at 8:42 PM
Most LLM evals use API calls or offline inference, testing models in a memory-less silo. Our new Patterns paper shows this misses how LLMs actually behave in real user interfaces, where personalization and interaction history shape responses: arxiv.org/abs/2509.19364
Reposted by Elinor🎗️
Elinor Poole-Dayan, Jiayi Wu, Taylor Sorensen, Jiaxin Pei, Michiel A. Bakker: Benchmarking Overton Pluralism in LLMs https://arxiv.org/abs/2512.01351 https://arxiv.org/pdf/2512.01351 https://arxiv.org/html/2512.01351
December 2, 2025 at 6:29 AM
Elinor Poole-Dayan, Jiayi Wu, Taylor Sorensen, Jiaxin Pei, Michiel A. Bakker: Benchmarking Overton Pluralism in LLMs https://arxiv.org/abs/2512.01351 https://arxiv.org/pdf/2512.01351 https://arxiv.org/html/2512.01351
Reposted by Elinor🎗️
Reposted by Elinor🎗️
Thoughtful (as always) blog post from Nicholas Carlini. "Are large language models worth it?" A nice read giving his perspective on risks of ML models.
Post: nicholas.carlini.com/writing/2025...
For people who prefer, this is the video of the talk from @colmweb.org www.youtube.com/watch?v=PngH...
Post: nicholas.carlini.com/writing/2025...
For people who prefer, this is the video of the talk from @colmweb.org www.youtube.com/watch?v=PngH...
November 19, 2025 at 4:56 PM
Thoughtful (as always) blog post from Nicholas Carlini. "Are large language models worth it?" A nice read giving his perspective on risks of ML models.
Post: nicholas.carlini.com/writing/2025...
For people who prefer, this is the video of the talk from @colmweb.org www.youtube.com/watch?v=PngH...
Post: nicholas.carlini.com/writing/2025...
For people who prefer, this is the video of the talk from @colmweb.org www.youtube.com/watch?v=PngH...
Reposted by Elinor🎗️
Extremely thrilled to talk about our new paper: "Who Evaluates AI’s Social Impacts? Mapping Coverage And Gaps In First And Third Party Evaluations".
This is the first big project output from the
@eval-eval.bsky.social coalition! Thread below:
This is the first big project output from the
@eval-eval.bsky.social coalition! Thread below:
November 13, 2025 at 2:35 PM
Extremely thrilled to talk about our new paper: "Who Evaluates AI’s Social Impacts? Mapping Coverage And Gaps In First And Third Party Evaluations".
This is the first big project output from the
@eval-eval.bsky.social coalition! Thread below:
This is the first big project output from the
@eval-eval.bsky.social coalition! Thread below:
Congratulations @sivareddyg.bsky.social ! 🥳 Incredibly well deserved!!
Congratulations to @sivareddyg.bsky.social, Core Academic Member at Mila, who has received the prestigious Outstanding Early Career Computer Science Researcher Award from CS-Can|Info-Can. mila.quebec/en/news/siva...
November 14, 2025 at 5:11 PM
Congratulations @sivareddyg.bsky.social ! 🥳 Incredibly well deserved!!
Reposted by Elinor🎗️
We're excited to announce that the website and registration for IC2S2 2026 (July 28-31) will launch in early December! The Vermont Complex Systems Institute @vcsi.bsky.social at the University of Vermont will be hosting IC2S2 in 2026: youtube.com/watch?v=p412S4GnPkc&feature=youtu.be
IC2S2 2026 | Burlington, Vermont
YouTube video by UVM Office of Research
youtube.com
November 13, 2025 at 3:36 PM
We're excited to announce that the website and registration for IC2S2 2026 (July 28-31) will launch in early December! The Vermont Complex Systems Institute @vcsi.bsky.social at the University of Vermont will be hosting IC2S2 in 2026: youtube.com/watch?v=p412S4GnPkc&feature=youtu.be
Reposted by Elinor🎗️
It's the season for PhD apps!! 🥧 🦃 ☃️ ❄️
Apply to Wisconsin CS to research
- Societal impact of AI
- NLP ←→ CSS and cultural analytics
- Computational sociolinguistics
- Human-AI interaction
- Culturally competent and inclusive NLP
with me!
lucy3.github.io/prospective-...
Apply to Wisconsin CS to research
- Societal impact of AI
- NLP ←→ CSS and cultural analytics
- Computational sociolinguistics
- Human-AI interaction
- Culturally competent and inclusive NLP
with me!
lucy3.github.io/prospective-...
November 11, 2025 at 10:32 PM
It's the season for PhD apps!! 🥧 🦃 ☃️ ❄️
Apply to Wisconsin CS to research
- Societal impact of AI
- NLP ←→ CSS and cultural analytics
- Computational sociolinguistics
- Human-AI interaction
- Culturally competent and inclusive NLP
with me!
lucy3.github.io/prospective-...
Apply to Wisconsin CS to research
- Societal impact of AI
- NLP ←→ CSS and cultural analytics
- Computational sociolinguistics
- Human-AI interaction
- Culturally competent and inclusive NLP
with me!
lucy3.github.io/prospective-...
@bennokrojer.bsky.social didn't your lab have something like this happen
November 7, 2025 at 9:02 PM
@bennokrojer.bsky.social didn't your lab have something like this happen
such a valuable resource! thanks for sharing
November 7, 2025 at 1:51 PM
such a valuable resource! thanks for sharing
Reposted by Elinor🎗️
It’s grad school application season, and I wanted to give some public advice.
Caveats:
-*-*-*-*
> These are my opinions, based on my experiences, they are not secret tricks or guarantees
> They are general guidelines, not meant to cover a host of idiosyncrasies and special cases
Caveats:
-*-*-*-*
> These are my opinions, based on my experiences, they are not secret tricks or guarantees
> They are general guidelines, not meant to cover a host of idiosyncrasies and special cases
November 6, 2025 at 2:55 PM
It’s grad school application season, and I wanted to give some public advice.
Caveats:
-*-*-*-*
> These are my opinions, based on my experiences, they are not secret tricks or guarantees
> They are general guidelines, not meant to cover a host of idiosyncrasies and special cases
Caveats:
-*-*-*-*
> These are my opinions, based on my experiences, they are not secret tricks or guarantees
> They are general guidelines, not meant to cover a host of idiosyncrasies and special cases
Reposted by Elinor🎗️
🧵Excited to present our work at #EMNLP2025 “Analyzing Dialectal Biases in LLMs for Knowledge and Reasoning Benchmarks”!
Paper 📄 arxiv.org/abs/2510.00962
w/ Eileen Pan, Skyler Seto, @allisonkoe.bsky.social @maartjeterhoeve.bsky.social
Paper 📄 arxiv.org/abs/2510.00962
w/ Eileen Pan, Skyler Seto, @allisonkoe.bsky.social @maartjeterhoeve.bsky.social
November 6, 2025 at 12:08 AM
🧵Excited to present our work at #EMNLP2025 “Analyzing Dialectal Biases in LLMs for Knowledge and Reasoning Benchmarks”!
Paper 📄 arxiv.org/abs/2510.00962
w/ Eileen Pan, Skyler Seto, @allisonkoe.bsky.social @maartjeterhoeve.bsky.social
Paper 📄 arxiv.org/abs/2510.00962
w/ Eileen Pan, Skyler Seto, @allisonkoe.bsky.social @maartjeterhoeve.bsky.social
wow. you always hear these horror stories (eg spelling the university/prof name wrong, or worse, writing the wrong name altogether) and think itll never happen to you but this is a great reminder that nobody is immune. hopefully your typos weren't so bad!
November 6, 2025 at 3:19 AM
wow. you always hear these horror stories (eg spelling the university/prof name wrong, or worse, writing the wrong name altogether) and think itll never happen to you but this is a great reminder that nobody is immune. hopefully your typos weren't so bad!
oh my 😅 hope you're not speaking from your own experience
November 6, 2025 at 3:13 AM
oh my 😅 hope you're not speaking from your own experience
inspired by having to re-upload my arxiv submission 3!!! times tonight
November 6, 2025 at 1:23 AM
inspired by having to re-upload my arxiv submission 3!!! times tonight
and once you correct that typo, proofread again and find nothing else, and re-submit, only then does a SECOND typo suddenly emerge 😭
November 6, 2025 at 1:23 AM
and once you correct that typo, proofread again and find nothing else, and re-submit, only then does a SECOND typo suddenly emerge 😭
it's crazy how typos are impossible to catch until *after* you submit a paper, after which they become glaringly noticeable
November 6, 2025 at 1:23 AM
it's crazy how typos are impossible to catch until *after* you submit a paper, after which they become glaringly noticeable
Reposted by Elinor🎗️
Which, whose, and how much knowledge do LLMs represent?
I'm excited to share our preprint answering these questions:
"Epistemic Diversity and Knowledge Collapse in Large Language Models"
📄Paper: arxiv.org/pdf/2510.04226
💻Code: github.com/dwright37/ll...
1/10
I'm excited to share our preprint answering these questions:
"Epistemic Diversity and Knowledge Collapse in Large Language Models"
📄Paper: arxiv.org/pdf/2510.04226
💻Code: github.com/dwright37/ll...
1/10
October 13, 2025 at 11:25 AM
Which, whose, and how much knowledge do LLMs represent?
I'm excited to share our preprint answering these questions:
"Epistemic Diversity and Knowledge Collapse in Large Language Models"
📄Paper: arxiv.org/pdf/2510.04226
💻Code: github.com/dwright37/ll...
1/10
I'm excited to share our preprint answering these questions:
"Epistemic Diversity and Knowledge Collapse in Large Language Models"
📄Paper: arxiv.org/pdf/2510.04226
💻Code: github.com/dwright37/ll...
1/10
Forgot to attach this screenshot from our discussion in the reply
November 4, 2025 at 11:13 PM
Forgot to attach this screenshot from our discussion in the reply
We definitely think the platform / setting would strongly influence this (especially online vs in person & if online, how it’s moderated)!
November 4, 2025 at 11:10 PM
We definitely think the platform / setting would strongly influence this (especially online vs in person & if online, how it’s moderated)!
Yes! While this work focused on in-person small-group facilitated conversations, they did range from more discussions on community topics as well as conversation games and policy deliberations. The biggest factors were purpose (eg game vs story sharing vs deliberation) as well as facilitator style
November 4, 2025 at 11:10 PM
Yes! While this work focused on in-person small-group facilitated conversations, they did range from more discussions on community topics as well as conversation games and policy deliberations. The biggest factors were purpose (eg game vs story sharing vs deliberation) as well as facilitator style
This work was done with awesome collaborators Maggie Hughes, Brandon Roy, Deb Roy, and @jad-kabbara.bsky.social from @mit.edu @medialab.bsky.social
Make sure to check out Brandon’s presentation on Friday Nov 7, Room A109 at 10:30!
Paper link: https://aclanthology.org/2025.emnlp-main.1798/
8/8
Make sure to check out Brandon’s presentation on Friday Nov 7, Room A109 at 10:30!
Paper link: https://aclanthology.org/2025.emnlp-main.1798/
8/8
November 3, 2025 at 10:41 PM
This work was done with awesome collaborators Maggie Hughes, Brandon Roy, Deb Roy, and @jad-kabbara.bsky.social from @mit.edu @medialab.bsky.social
Make sure to check out Brandon’s presentation on Friday Nov 7, Room A109 at 10:30!
Paper link: https://aclanthology.org/2025.emnlp-main.1798/
8/8
Make sure to check out Brandon’s presentation on Friday Nov 7, Room A109 at 10:30!
Paper link: https://aclanthology.org/2025.emnlp-main.1798/
8/8
Overall, responsivity offers a structural lens on how dialogue unfolds 🧱,
laying groundwork for understanding what makes conversations constructive, collaborative, and engaging across contexts 🌱🫂
Future work: richer metrics + applications in civic 🗳️ + online spaces 🌍
7/
laying groundwork for understanding what makes conversations constructive, collaborative, and engaging across contexts 🌱🫂
Future work: richer metrics + applications in civic 🗳️ + online spaces 🌍
7/
November 3, 2025 at 10:41 PM
Overall, responsivity offers a structural lens on how dialogue unfolds 🧱,
laying groundwork for understanding what makes conversations constructive, collaborative, and engaging across contexts 🌱🫂
Future work: richer metrics + applications in civic 🗳️ + online spaces 🌍
7/
laying groundwork for understanding what makes conversations constructive, collaborative, and engaging across contexts 🌱🫂
Future work: richer metrics + applications in civic 🗳️ + online spaces 🌍
7/
When applied across diverse facilitated discussions, our metrics reveal 🔎
5 distinct, interpretable 🧠 conversation types aligned with their intended purposes 🎯 + facilitation styles 📣
6/
5 distinct, interpretable 🧠 conversation types aligned with their intended purposes 🎯 + facilitation styles 📣
6/
November 3, 2025 at 10:41 PM
When applied across diverse facilitated discussions, our metrics reveal 🔎
5 distinct, interpretable 🧠 conversation types aligned with their intended purposes 🎯 + facilitation styles 📣
6/
5 distinct, interpretable 🧠 conversation types aligned with their intended purposes 🎯 + facilitation styles 📣
6/
Using these annotations, we derive conversation-level metrics capturing:
• Balance b/t speakers ⚖️
• Turn sequence variability 🔁
• Who responds to whom 🙋♀️↔️👂
• Distribution of substantive responses 📊
• Facilitator-participant dynamics 🧩
5/
• Balance b/t speakers ⚖️
• Turn sequence variability 🔁
• Who responds to whom 🙋♀️↔️👂
• Distribution of substantive responses 📊
• Facilitator-participant dynamics 🧩
5/
November 3, 2025 at 10:41 PM
Using these annotations, we derive conversation-level metrics capturing:
• Balance b/t speakers ⚖️
• Turn sequence variability 🔁
• Who responds to whom 🙋♀️↔️👂
• Distribution of substantive responses 📊
• Facilitator-participant dynamics 🧩
5/
• Balance b/t speakers ⚖️
• Turn sequence variability 🔁
• Who responds to whom 🙋♀️↔️👂
• Distribution of substantive responses 📊
• Facilitator-participant dynamics 🧩
5/