AGENTIC R AG-R1: Agentic Reinforcement Learning with Stack Memory for Multi-Step Reasoning, Retrieval and Memorizing
Not assessedPlan blockedFindingclaim-long-horizon
The paper reports increasing AGENTIC R AG-R1 F1 as the maximum reasoning-step budget grows on selected 2Wiki, Bamboogle, and TriviaQA evaluations.
Source: paper-fixed:page 9, Table 3
Reported and observed measurements
F1
m-horizon-2wiki-10
Reported 23.91 percentage_points
Observed — percentage_points
F1
m-horizon-2wiki-30
Reported 47.45 percentage_points
Observed — percentage_points
Assessments (0)
No immutable Assessment has been published for this Claim yet.
Runs (0)
No execution has been linked to this Claim yet.