In our experiment, a person is given an opinionated claim on a topic and asked to rate their level of support. They’re then presented with an argument in support of that claim, written by LMs or another person, and asked to re-rate their support of the original claim.
@anthropicai
-
Measuring Language Model Persuasiveness Compared to Humans
By
–
To assess persuasiveness, we measure the shift in people’s support between their initial view on a claim and their view after reading arguments written by either a human or an LM. We define the persuasiveness metric as the difference between the support scores.
-
Studying Malleable Opinions on Technology and Education
By
–
We focus on arguments regarding less polarized issues, such as views on new technologies, space exploration, and education. We did this because we thought people’s opinions on these topics might be more malleable than their opinions on polarizing issues.
-
Claude 3 Opus Arguments Match Human Persuasiveness Quality
By
–
We find that Claude 3 Opus generates arguments that don't statistically differ in persuasiveness compared to arguments written by humans. We also find a scaling trend across model generations: newer models tended to be rated as more persuasive than previous ones.
-
Anthropic Research: Measuring Language Model Persuasiveness
By
–
New Anthropic research: Measuring Model Persuasiveness We developed a way to test how persuasive language models (LMs) are, and analyzed how persuasiveness scales across different versions of Claude. Read our blog post here: http://
anthropic.com/news/measuring
-model-persuasiveness
… -
Claude Tool Use Now Available Beta Anthropic Messages API
By
–
Tool use is now available in beta to all customers in the Anthropic Messages API, enabling Claude to interact with external tools using structured outputs. pic.twitter.com/kJHhSvyLdK
— Anthropic (@AnthropicAI) 4 avril 2024Tool use is now available in beta to all customers in the Anthropic Messages API, enabling Claude to interact with external tools using structured outputs.
-
Claude Agentic Capabilities: Document Retrieval and Task Orchestration
By
–
If instructed, Claude can enable agentic retrieval of documents from your internal knowledge base and APIs, complete tasks requiring real-time data or complex computations, and orchestrate Claude subagents for granular requests.
-
Anthropic Releases Tool Use Documentation for Claude API
By
–
We look forward to your feedback. Read more in our developer documentation: https://
docs.anthropic.com/claude/docs/to
ol-use
… -
Anthropic Alignment Science Team Hiring Research Engineer
By
–
If you’re interested in working with us on this and related problems, our Alignment Science team is hiring. Take a look at our Research Engineer job listing: https://
jobs.lever.co/Anthropic/444e
7d1e-ec6a-4d18-b712-0cb79c27f00f
… -
Accenture and AWS Partner to Scale Customized AI Solutions
By
–
We’re excited to team up with @Accenture and @awscloud to help organizations adopt and scale customized AI solutions.