Rashi Shrivastava is a reporter covering technology with a focus in artificial intelligence. She writes a weekly Forbes newsletter on all things AI called The Prompt. She joined Forbes in January 2022 and is based in Upstate New York.
AI evaluation company Patronus AI claims that its new model, Lynx can not only catch hallucinations produced by large language models but also explain why they’re wrong.Rebecca Qian and Anand Kannappan , former AI researchers at Meta founded Patronus AI to develop automation that detects factual inaccuracies and harmful content produced by AI models.
Before starting the company in September 2023, the duo spoke with about 60 company executives and found that their worst fear was launching an AI product and making headlines for the wrong reasons. CEO Kannappan hopes that Lynx can help assuage those concerns. He thinks of it as a “coach” for other AI models that can guide them to be more accurate.
That’s a contrast to the way AI products are stress tested now before shipment, he continued, which involves a range of techniques. One of those is “,” which involves manually hacking AI models to expose vulnerabilities that might lead to mistakes. Other development teams use AI models like GPT-4 to catch hallucinations, Kannappan said, who criticized this approach as “literally GPT-4 evaluating GPT-4 itself.
The company has also released a new benchmark called HaluBench, which rates how well different AI models can catch hallucinations in model output, especially across legal, financial and medical domains. That benchmark shows that even Lynx isn’t perfect–it scores about 88% accuracy–but it does outperform most others, Kannappan said.
Lynx AI Model AI Hallucination Red Teaming Meta GPT-4 Benchmark Evaluation
United States Latest News, United States Headlines
Similar News:You can also read news stories similar to this one that we have collected from other news sources.
Karen Read accused of killing her Boston police boyfriend she says she admiredThe woman accused of ramming her luxury SUV into her boyfriend and callously leaving him on the ground before a snowstorm had dated the Boston police officer briefly when they were in their 20s, and they reconnected after the pandemic.
Read more »
Cyndi Lauper Says She’s ‘Proud’ of Taylor Swift: ‘She Writes Some Wonderful Songs’Cyndi Lauper praised Taylor Swift's songwriting, saying 'I'm proud of her' when asked about the pop star in a new interview. Get the details.
Read more »
Judge Judy knew she made it big when she went from secondhand clothes to designer shoesJudge Judy told The Post’s Cindy Adams that she realized she was successful when she could afford Robert Clergerie shoes.
Read more »
ONLY ON 13: Woman says she caught man peeping into Baytown apartment window as she undressedA woman getting undressed says an alert of motion detected outside her boyfriend's Baytown apartment led her to find a man bending over and staring through the blinds.
Read more »
Pam Anderson's no-botox skincare routine revealed — why she always looks like she's glowingPamela Anderson goes for Old Hollywood glamour at the 2024 Met Gala
Read more »
Harley Quinn Names the 1 Justice League Hero She Actually Respects (& 1 She Definitely Doesn't)harley quinn justice league roller derby team
Read more »