Skip to content
Trending storyDeveloping

Jiaxuan Zou argues agent pretraining needs multi-turn tool-use data

1 article1 sourcesince Nov 11Last article

Overview

AISummary of 1 article

Jiaxuan Zou argues that pretraining and continued pretraining for agents should not rely on static next-token loss alone.

He says training data must contain multi-turn tool-use patterns so the model begins reinforcement learning from a higher starting score. Zou cites reports from MAI, Reflection AI, Mistral and NVIDIA to support the claim. The report contains his argument only, with no experimental results attached.

Written by AI from the articles below · updated Nov 11, 2:43 PM ET

Check the sources:

Latest developmentNov 11, 1:17 PM ET
Jiaxuan Zou argues pretraining must inject multi-turn tool-use patterns for agents

Article timeline

The articles in this story. Times are ET.

Nov 11
  1. Jiaxuan ZouX
    Jiaxuan Zou argues pretraining must inject multi-turn tool-use patterns for agents

    AIJiaxuan Zou argues that pretraining and continued pretraining for agents should not rely on static NLL alone. He says multi-turn tool-use patterns must exist in the training data so the model starts reinforcement learning from a higher score, and cites reports from MAI, Reflection AI, Mistral, and NVIDIA to support the claim.

Heat Trend

Current comparable heat
6
Peak comparable heat
10Nov 11, 3:00 PM ET
Change in past 24 hours
–

The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.