Industry

Perplexity 计划构建多模型多智能体框架,打造自动化研究员

Why we picked it

Perplexity 的 CEO 亲自说了路线图:多模型加多智能体并行跑假设、做消融实验,目标是自动化研究员。想了解 agent 框架怎么用于科研流程的开发者,这条推文信息量不大但方向明确,适合配合 deep research 类产品一起对比着看。

Perplexity CEO Aravind Srinivas 在 X 上表示,公司将持续构建一个 multi model 与 multi agent 框架,用于处理开放性问题回答、实验设计、假设构建等复杂任务。

Body

该框架的思路是让多个智能体并行地对假设进行测试与 ablation(消融实验),再逐步迭代优化预测与答案。这条推文同时列出了多条推文约 1597 次浏览、5 次点赞的数据,说明其引发一定讨论。

最终目标是做出一个 automated researcher(自动化研究员),把科研中假设、验证、修正的循环交给智能体完成。这延续了行业里 multi agent 协作研究系统的方向,与 OpenAI 的 deep research 类产品思路相近,区别在于 Perplexity 强调多模型混用和并行消融的 harness 设计。

Original · Aravind Srinivas

We intend to keep building on a multi model and multi agent harness for more complex tasks like answering open ended questions, designing experiments, building hypotheses, testing and ablating them in parallel, and incrementally refining the predictions and answers; all with the goal of building an automated researcher.📊 2 ⚡

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Industry" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Perplexity, multi-agent, deep research.

Source attribution: curated from Aravind Srinivas (@AravSrinivas).