







Why Can't More Than Four People Have a Conversation at Once? - Slashdot
Apparently, there exists something called the 'dinner party problem' which states that it is difficult to sustain a casual conversation that includes more than four speakers. If a fifth person were to join that conversation, so goes the theory, the conversation would quickly fission into smaller gro...

Japan Olympic chief says 'lot more' athletes at Games than expected
Japan's Olympic chief admitted Saturday that "a lot more" athletes and officials came to the Asian Games than the host nation had expected, leading...

2,558 Is Not a Number of People - Ewan’s Blog
Bluesky says I have 2,558 followers. For two years I've treated that number as the verdict on my writing: that many people decided my posts were worth seeing…
People are getting their news from AI – and it’s altering their views
Even when information is factually accurate, how it’s presented can introduce subtle biases. As large language models increasingly bring people the news, this bias is a looming problem.

People are getting their news from AI – and it’s altering their views
Even when information is factually accurate, how it’s presented can introduce subtle biases. As large language models increasingly bring people the news, this bias is a looming problem.


AI is More Persuasive Than People in Online Debates - Slashdot
Chatbots are more persuasive in online debates than people -- especially when they are able to personalize their arguments using information about their opponent. From a report: The finding, published in Nature Human Behaviour on 19 May, highlights how large language models (LLMs) could be used to i...

Do Large Language Models Perform the Way People Expect? Measuring the Human Generalization Function
What makes large language models (LLMs) impressive is also what makes them hard to evaluate: their diversity of uses. To evaluate these models, we must understand the purposes they will be used for. We consider a setting where these deployment decisions are made by people, and in particular, people's beliefs about where an LLM will perform well. We model such beliefs as the consequence of a human generalization function: having seen what an LLM gets right or wrong, people generalize to where else it might succeed. We collect a dataset of 19K examples of how humans make generalizations across 79 tasks from the MMLU and BIG-Bench benchmarks. We show that the human generalization function can be predicted using NLP methods: people have consistent structured ways to generalize. We then evaluate LLM alignment with the human generalization function. Our results show that -- especially for cases where the cost of mistakes is high -- more capable models (e.g. GPT-4) can do worse on the instances people choose to use them for, exactly because they are not aligned with the human generalization function.

Another Symptom of Growing Isolation
A recent study found that people spoke fewer words a day in 2019 than in 2005.
Researchers say we’re talking less than ever
It’s not just small talk, it’s all talk that’s dying.

Quantifying the Persona Effect in LLM Simulations
Large language models (LLMs) have shown remarkable promise in simulating human language and behavior. This study investigates how integrating persona variables—demographic, social, and behavioral factors—impacts LLMs' ability to simulate diverse perspectives. We find that persona variables account for \ensuremath<10% variance in annotations in existing subjective NLP datasets. Nonetheless, incorporating persona variables via prompting in LLMs provides modest but statistically significant improvements. Persona prompting is most effective in samples where many annotators disagree, but their disagreements are relatively minor. Notably, we find a linear relationship in our setting: the stronger the correlation between persona variables and human annotations, the more accurate the LLM predictions are using persona prompting. In a zero-shot setting, a powerful 70b model with persona prompting captures 81% of the annotation variance achievable by linear regression trained on ground truth annotations. However, for most subjective NLP datasets, where persona variables have limited explanatory power, the benefits of persona prompting are limited.
True in my case
3.9K votes, 105 comments. 36M subscribers in the memes community. Memes! A way of describing cultural information being shared. An element of a…
Claude Usage Limits Discussion Megathread Ongoing (sort this by New!)
628 votes, 2.3K comments. Why a Usage Limits Discussion Megathread? This Megathread makes it easier for everyone to see what others are experiencing…
As many as 48 million Twitter accounts aren't people, says study
A big chunk of those "likes," "retweets," and "followers" lighting up your Twitter account may not be coming from human hands.

We asked 15,000 people who they are, and how they’re learning to code
More than 15,000 people responded to the 2016 New Coder Survey, granting researchers an unprecedented glimpse into how adults are learning to code. We’ve released the entire dataset of participants’ individual responses to all 48 questions — under th...
