发生了什么
Aravind Srinivas 分享了关于 Perplexity Computer 的 Brain 的研究,该功能利用过往会话轨迹,在发现具有足够证据的重复模式时提出个性化的智能体技能建议。在一项针对生产流量的八周随机测试中,72.9% 的用户获得了至少一项新的个人技能,而手动添加的用户比例为 6.5%。工具调用失败率下降了 11%,不满意的对话串下降了 7.6%,且成本和延迟没有可测量的变化。
推荐理由
一项针对生产流量的八周随机测试发现,工具调用失败和不满意的对话串有所减少,而成本和延迟没有显示出可测量的变化;共享技能还会经过回放测试和工程师审查。
Sharing new research on self-improving skills in Perplexity Computer. Brain, which already maintains a knowledge wiki about your work, now also writes and updates agent skills for you based on your past sessions. An improvement agent reads the full session trajectories (tool calls, reasoning, results), finds recurring patterns, and only proposes a new skill or an update when there’s enough evidence. Built-in skills that every user gets go through paired replay on failure scenarios and engineer review before any change goes live. In an 8-week randomized test on production traffic, 72.9% of users got at least one new personal skill vs 6.5% who added one manually. Tool call failures dropped 11% and dissatisfied threads dropped 7.6%, with no measurable change in cost or latency.
在 X 查看被引用的帖子Computer's personalized self-improvement agent, Brain, now automatically creates self-improving agent skills for you.
Brain learns how you work from past sessions and memory, then writes and updates personalized skills for future tasks.
这件事到目前为止
Perplexity 表示,其 Perplexity Computer 中的个性化自我改进智能体 Brain 现可为未来任务自动创建和更新个性化智能体技能。据 Perplexity 称,Brain 会从过往会话和记忆中学习用户的工作方式,并在过往会话轨迹中发现有足够证据支持的重复模式时提出新技能。 Aravind Srinivas 分享了一项在生产流量上对该功能进行评估的为期八周的随机测试结果。Srinivas 表示,在测试中,72.9% 的用户至少获得了一项新的个人技能,相比之下手动添加技能的用户比例为 6.5%。工具调用失败率下降了 11%,不满意的对话串减少了 7.6%,而成本和延迟没有明显变化。
