Perplexity.AI improves agent tool-call reliability with new training method
Perplexity.AI has announced a new training methodology for its AI agent, which it claims has reduced tool-call failures by 21%. This approach, termed Hint-Guided Self-Distillation, involves training the agent on real-world mistakes to improve its reliability. The company is the primary source for these reported improvements.
2 articles from 2 outlets covered this story. The underlying claim is sourced from a benchmark.
What do all outlets agree on?
2 outlets covered “Perplexity.AI improves agent tool-call reliability with new training method”. All of them report the following:
- Perplexity.AI developed a new training approach for its AI agent
- The approach aims to reduce tool-call failures
- The method involves training on real mistakes
- The method is called Hint-Guided Self-Distillation
- The training resulted in a 21% reduction in tool-call failures
Which outlets covered this?
All 2 articles found on this story, grouped by the stance of the piece. Every link goes to the original publisher.
pragmatist2