Skip to content
Trending storyDeveloping

Harvey reports context graphs raise legal agent accuracy on LAB tasks

1 article1 sourcesince Oct 15Last article 2h ago ·

Overview

AISummary of 1 article

Harvey reports that context graphs raised its legal agents' accuracy on LAB tasks, according to a post by Elvis Saravia that cites Harvey's report.

Firm Knowledge accuracy rose from 65% to 80%, precedent lists from 67% to 80%, and matter counting from 53% to 83%. Harvey also reports a 48% cost reduction for agents given a context graph.

Saravia says structured data works well for LLM-based agents, especially in complex codebases, and that context graphs may have broader uses. The figures are Harvey's own reported results, and the post does not include independent verification.

Written by AI from the articles below · updated Oct 15, 5:16 PM ET

Check the sources:

Latest developmentOct 15, 5:06 PM ET
Context graphs lift Harvey's legal agent accuracy on LAB tasks

Article timeline

The articles in this story. Times are ET.

Oct 15
  1. elvisX
    Context graphs lift Harvey's legal agent accuracy on LAB tasks

    AIElvis Saravia says structured data works well for LLM-based agents, especially in complex codebases, and that context graphs may have broader uses. He points to Harvey's report, where context graphs raised LAB task accuracy: Firm Knowledge rose from 65% to 80%, precedent lists from 67% to 80%, and matter counting from 53% to 83%. Harvey also reports a 48% cost reduction for agents given a context graph.

Heat trend

Not enough continuous observations to show a trend yet.