Not Worth Another Token: Marginal Value Estimation for Efficient Deep Research Agents
The results show that pruning effectiveness depends more on where pruning is applied than on the specific scoring rule: early pruning yields the largest end-to-end savings, while later pruning mainly refines the final synthesis context.
Harshitha Kolukuluru, Reshma Ashok, Kirat Arora et al.
· 0 citations