Danica Dillion
@danicajdillion.bsky.social
Postdoc @csh.ac.at studying how social and technological change reshapes our beliefs. morality, AI, religion, politics, social networks. danicadillion.com
We hope WorldValuesBench supports building LLMs that are safer, fairer, and more culturally aware.
Led by Wenlong Zhao with Debanjan Mondal, Niket Tandon,
@kurtjgray.bsky.social, and Yuling Gu
Code: github.com/Demon702/Wor...
Paper: aclanthology.org/2024.lrec-ma...
Led by Wenlong Zhao with Debanjan Mondal, Niket Tandon,
@kurtjgray.bsky.social, and Yuling Gu
Code: github.com/Demon702/Wor...
Paper: aclanthology.org/2024.lrec-ma...
WorldValuesBench: A Large-Scale Benchmark Dataset for Multi-Cultural Value Awareness of Language Models
Wenlong Zhao, Debanjan Mondal, Niket Tandon, Danica Dillion, Kurt Gray, Yuling Gu. Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluatio...
aclanthology.org
September 19, 2025 at 1:11 PM
We hope WorldValuesBench supports building LLMs that are safer, fairer, and more culturally aware.
Led by Wenlong Zhao with Debanjan Mondal, Niket Tandon,
@kurtjgray.bsky.social, and Yuling Gu
Code: github.com/Demon702/Wor...
Paper: aclanthology.org/2024.lrec-ma...
Led by Wenlong Zhao with Debanjan Mondal, Niket Tandon,
@kurtjgray.bsky.social, and Yuling Gu
Code: github.com/Demon702/Wor...
Paper: aclanthology.org/2024.lrec-ma...
The World Values Survey is one of the most widely used datasets on global values:
🌏 94,728 participants across 64 countries in the latest wave
📊 Representative national samples
🗣️ Primarily face-to-face interviews
❓ 200+ value questions spanning social, political, economic, and moral domains
🌏 94,728 participants across 64 countries in the latest wave
📊 Representative national samples
🗣️ Primarily face-to-face interviews
❓ 200+ value questions spanning social, political, economic, and moral domains
September 19, 2025 at 1:11 PM
The World Values Survey is one of the most widely used datasets on global values:
🌏 94,728 participants across 64 countries in the latest wave
📊 Representative national samples
🗣️ Primarily face-to-face interviews
❓ 200+ value questions spanning social, political, economic, and moral domains
🌏 94,728 participants across 64 countries in the latest wave
📊 Representative national samples
🗣️ Primarily face-to-face interviews
❓ 200+ value questions spanning social, political, economic, and moral domains
What makes WorldValuesBench different?
🎯 Fine-grained demographics: 42 attributes (age, gender, country, urban/rural, etc.)
🌐 Breadth of values: social norms, trust, economics, religion, politics, and more
📊 Scale: 20M examples for evaluation
📏 Distributions rather than averages of human values
🎯 Fine-grained demographics: 42 attributes (age, gender, country, urban/rural, etc.)
🌐 Breadth of values: social norms, trust, economics, religion, politics, and more
📊 Scale: 20M examples for evaluation
📏 Distributions rather than averages of human values
September 19, 2025 at 1:11 PM
What makes WorldValuesBench different?
🎯 Fine-grained demographics: 42 attributes (age, gender, country, urban/rural, etc.)
🌐 Breadth of values: social norms, trust, economics, religion, politics, and more
📊 Scale: 20M examples for evaluation
📏 Distributions rather than averages of human values
🎯 Fine-grained demographics: 42 attributes (age, gender, country, urban/rural, etc.)
🌐 Breadth of values: social norms, trust, economics, religion, politics, and more
📊 Scale: 20M examples for evaluation
📏 Distributions rather than averages of human values
LLMs are already shaping advice, education, and policy around the world. To be safe and serve global users, it's important they reflect cultural variation in values rather than a single “global average” or disproportionately Western views.
September 19, 2025 at 1:11 PM
LLMs are already shaping advice, education, and policy around the world. To be safe and serve global users, it's important they reflect cultural variation in values rather than a single “global average” or disproportionately Western views.
Huge thanks to brilliant coauthors @helenldevine.bsky.social and @kurtjgray.bsky.social!
Preprint available here: doi.org/10.31234/osf...
Data, code, and preregs here: osf.io/d94b8/
Preprint available here: doi.org/10.31234/osf...
Data, code, and preregs here: osf.io/d94b8/
OSF
doi.org
August 26, 2025 at 5:01 PM
Huge thanks to brilliant coauthors @helenldevine.bsky.social and @kurtjgray.bsky.social!
Preprint available here: doi.org/10.31234/osf...
Data, code, and preregs here: osf.io/d94b8/
Preprint available here: doi.org/10.31234/osf...
Data, code, and preregs here: osf.io/d94b8/
Dogs are amazing companions.
But when love for dogs surpasses love for people, it can come at a cost: less concern for others and deeper social disconnection.
But when love for dogs surpasses love for people, it can come at a cost: less concern for others and deeper social disconnection.
August 26, 2025 at 5:01 PM
Dogs are amazing companions.
But when love for dogs surpasses love for people, it can come at a cost: less concern for others and deeper social disconnection.
But when love for dogs surpasses love for people, it can come at a cost: less concern for others and deeper social disconnection.
For many, dogs aren’t just valued as much as people—they’re valued more.
This can have real impacts: Owners chose to give more to animal charities than those helping people, including a children's hospital and food bank. 🐶>🏥🥫
Even children in need can come second to dogs.
This can have real impacts: Owners chose to give more to animal charities than those helping people, including a children's hospital and food bank. 🐶>🏥🥫
Even children in need can come second to dogs.
August 26, 2025 at 5:01 PM
For many, dogs aren’t just valued as much as people—they’re valued more.
This can have real impacts: Owners chose to give more to animal charities than those helping people, including a children's hospital and food bank. 🐶>🏥🥫
Even children in need can come second to dogs.
This can have real impacts: Owners chose to give more to animal charities than those helping people, including a children's hospital and food bank. 🐶>🏥🥫
Even children in need can come second to dogs.
Dog owners were more likely than non-owners to prioritize dogs, but even non-owners scored high on these measures.
This suggests that seeing dogs as soulmates is widespread. 🐶💞
This suggests that seeing dogs as soulmates is widespread. 🐶💞
August 26, 2025 at 5:01 PM
Dog owners were more likely than non-owners to prioritize dogs, but even non-owners scored high on these measures.
This suggests that seeing dogs as soulmates is widespread. 🐶💞
This suggests that seeing dogs as soulmates is widespread. 🐶💞
The same trend appears locally:
US counties with lower birth rates have more pet stores and higher pet industry earnings, even after controlling for population and GDP.
US counties with lower birth rates have more pet stores and higher pet industry earnings, even after controlling for population and GDP.
August 26, 2025 at 5:01 PM
The same trend appears locally:
US counties with lower birth rates have more pet stores and higher pet industry earnings, even after controlling for population and GDP.
US counties with lower birth rates have more pet stores and higher pet industry earnings, even after controlling for population and GDP.
As the US birth rate falls, pet spending rises—with a strong negative correlation (r = -.93)
This link holds even after accounting for inflation, GDP, population, the poverty rate, and median age.
This link holds even after accounting for inflation, GDP, population, the poverty rate, and median age.
August 26, 2025 at 5:01 PM
As the US birth rate falls, pet spending rises—with a strong negative correlation (r = -.93)
This link holds even after accounting for inflation, GDP, population, the poverty rate, and median age.
This link holds even after accounting for inflation, GDP, population, the poverty rate, and median age.
Dog-over-human choices were strongest among owners without kids. Eg:
🐶 73% of childless owners would save their dog over a stranger
🍼 Only 44% of owners with kids would do the same
Childless owners were also more likely to say their dog takes the place of a baby.
And looking @ national trends...
🐶 73% of childless owners would save their dog over a stranger
🍼 Only 44% of owners with kids would do the same
Childless owners were also more likely to say their dog takes the place of a baby.
And looking @ national trends...
August 26, 2025 at 5:01 PM
Dog-over-human choices were strongest among owners without kids. Eg:
🐶 73% of childless owners would save their dog over a stranger
🍼 Only 44% of owners with kids would do the same
Childless owners were also more likely to say their dog takes the place of a baby.
And looking @ national trends...
🐶 73% of childless owners would save their dog over a stranger
🍼 Only 44% of owners with kids would do the same
Childless owners were also more likely to say their dog takes the place of a baby.
And looking @ national trends...
56% of owners would save their own dog over a person. And it doesn't stop there:
🐶 25% would give $50 to a puppy over a child in need
🐶 43% would feed a hungry puppy over a hungry stranger
🐶 20% would save an unfamiliar puppy over a human life
🐶 25% would give $50 to a puppy over a child in need
🐶 43% would feed a hungry puppy over a hungry stranger
🐶 20% would save an unfamiliar puppy over a human life
August 26, 2025 at 5:01 PM
56% of owners would save their own dog over a person. And it doesn't stop there:
🐶 25% would give $50 to a puppy over a child in need
🐶 43% would feed a hungry puppy over a hungry stranger
🐶 20% would save an unfamiliar puppy over a human life
🐶 25% would give $50 to a puppy over a child in need
🐶 43% would feed a hungry puppy over a hungry stranger
🐶 20% would save an unfamiliar puppy over a human life
We wanted to know whether this soulmate bond shifts moral concern from people to dogs.
Would owners pick a dog’s life over a human’s—even a stranger’s dog?
So we posed moral dilemmas pitting dogs against people 💁⚖️🐶
Would owners pick a dog’s life over a human’s—even a stranger’s dog?
So we posed moral dilemmas pitting dogs against people 💁⚖️🐶
August 26, 2025 at 5:01 PM
We wanted to know whether this soulmate bond shifts moral concern from people to dogs.
Would owners pick a dog’s life over a human’s—even a stranger’s dog?
So we posed moral dilemmas pitting dogs against people 💁⚖️🐶
Would owners pick a dog’s life over a human’s—even a stranger’s dog?
So we posed moral dilemmas pitting dogs against people 💁⚖️🐶
It’s no surprise that people who see their dog as a soulmate pamper them—sharing plates, beds, and fancy treats.
But many go further: preferring dogs over friends, and choosing dog time over people time.
But many go further: preferring dogs over friends, and choosing dog time over people time.
August 26, 2025 at 5:01 PM
It’s no surprise that people who see their dog as a soulmate pamper them—sharing plates, beds, and fancy treats.
But many go further: preferring dogs over friends, and choosing dog time over people time.
But many go further: preferring dogs over friends, and choosing dog time over people time.
73% of dog owners see their dog as a soulmate ❤️🐶
They endorse statements like:
"My dog is my main companion"
"My dog is my main emotional support"
"My dog’s love is purer than most people’s love"
They endorse statements like:
"My dog is my main companion"
"My dog is my main emotional support"
"My dog’s love is purer than most people’s love"
August 26, 2025 at 5:01 PM
73% of dog owners see their dog as a soulmate ❤️🐶
They endorse statements like:
"My dog is my main companion"
"My dog is my main emotional support"
"My dog’s love is purer than most people’s love"
They endorse statements like:
"My dog is my main companion"
"My dog is my main emotional support"
"My dog’s love is purer than most people’s love"
Many thanks to my amazing co-authors Debanjan Mondal, Wenlong Zhao, Niket Tandon, and @kurtjgray.bsky.social!
Preprint link: doi.org/10.31234/osf...
Preprint link: doi.org/10.31234/osf...
OSF
doi.org
August 25, 2025 at 4:54 PM
Many thanks to my amazing co-authors Debanjan Mondal, Wenlong Zhao, Niket Tandon, and @kurtjgray.bsky.social!
Preprint link: doi.org/10.31234/osf...
Preprint link: doi.org/10.31234/osf...
AI is already shaping moral decisions in high-stakes domains like the military, healthcare, and education.
We need AI that reflects people’s values and is more transparent in its decisions.
Our work shows how moral psychology can help guide the design of more interpretable, morally aligned AI.
We need AI that reflects people’s values and is more transparent in its decisions.
Our work shows how moral psychology can help guide the design of more interpretable, morally aligned AI.
August 25, 2025 at 4:54 PM
AI is already shaping moral decisions in high-stakes domains like the military, healthcare, and education.
We need AI that reflects people’s values and is more transparent in its decisions.
Our work shows how moral psychology can help guide the design of more interpretable, morally aligned AI.
We need AI that reflects people’s values and is more transparent in its decisions.
Our work shows how moral psychology can help guide the design of more interpretable, morally aligned AI.
It’s also promising that bottlenecks beat the baseline across several moral frameworks.
For practical uses, bottlenecks should be tailored to the people they serve and the models they’re used with.
We’re hopeful similar approaches can be adapted to different use cases.
For practical uses, bottlenecks should be tailored to the people they serve and the models they’re used with.
We’re hopeful similar approaches can be adapted to different use cases.
August 25, 2025 at 4:54 PM
It’s also promising that bottlenecks beat the baseline across several moral frameworks.
For practical uses, bottlenecks should be tailored to the people they serve and the models they’re used with.
We’re hopeful similar approaches can be adapted to different use cases.
For practical uses, bottlenecks should be tailored to the people they serve and the models they’re used with.
We’re hopeful similar approaches can be adapted to different use cases.
One great thing about the bottleneck method is that it’s simple. It doesn’t require additional training or data.
August 25, 2025 at 4:54 PM
One great thing about the bottleneck method is that it’s simple. It doesn’t require additional training or data.
How do people feel about the bottleneck method?
In a preregistered study (n=239), an AI described as using a cognitive bottleneck was rated as more transparent, trustworthy, and moral than a baseline end-to-end model.
In a preregistered study (n=239), an AI described as using a cognitive bottleneck was rated as more transparent, trustworthy, and moral than a baseline end-to-end model.
August 25, 2025 at 4:54 PM
How do people feel about the bottleneck method?
In a preregistered study (n=239), an AI described as using a cognitive bottleneck was rated as more transparent, trustworthy, and moral than a baseline end-to-end model.
In a preregistered study (n=239), an AI described as using a cognitive bottleneck was rated as more transparent, trustworthy, and moral than a baseline end-to-end model.
Of course, bottlenecks can’t make LLMs fully transparent, as their inner workings are too complex to trace.
But bottleneck feature ratings explained over 90% of the variance in model moral scores, adding a layer of interpretability.
But bottleneck feature ratings explained over 90% of the variance in model moral scores, adding a layer of interpretability.
August 25, 2025 at 4:54 PM
Of course, bottlenecks can’t make LLMs fully transparent, as their inner workings are too complex to trace.
But bottleneck feature ratings explained over 90% of the variance in model moral scores, adding a layer of interpretability.
But bottleneck feature ratings explained over 90% of the variance in model moral scores, adding a layer of interpretability.