pith. the verified trust layer for science. sign in

I need to take an egg from countertop 2 to proceed with the task

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.LG 1

years

2025 1

verdicts

UNVERDICTED 1

representative citing papers

Group-in-Group Policy Optimization for LLM Agent Training

cs.LG · 2025-05-16 · unverdicted · novelty 7.0

GiGPO adds a hierarchical grouping mechanism to group-based RL so that LLM agents receive both global trajectory and local step-level credit signals, yielding >12% gains on ALFWorld and >9% on WebShop over GRPO while keeping the same rollout and memory footprint.

citing papers explorer

Showing 1 of 1 citing paper.

  • Group-in-Group Policy Optimization for LLM Agent Training cs.LG · 2025-05-16 · unverdicted · none · ref 100

    GiGPO adds a hierarchical grouping mechanism to group-based RL so that LLM agents receive both global trajectory and local step-level credit signals, yielding >12% gains on ALFWorld and >9% on WebShop over GRPO while keeping the same rollout and memory footprint.