ParallelSearch applies reinforcement learning with parallel-decomposition rewards to teach LLM search agents to issue independent sub-queries concurrently, lifting QA accuracy by 2.9% on average and 12.7% on parallelizable questions while using about 70% of the baseline's LLM calls.
A review of factors influencing user satisfaction in information retrieval
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ParallelSearch: Train your LLMs to Decompose Query and Search Sub-queries in Parallel with Reinforcement Learning
ParallelSearch applies reinforcement learning with parallel-decomposition rewards to teach LLM search agents to issue independent sub-queries concurrently, lifting QA accuracy by 2.9% on average and 12.7% on parallelizable questions while using about 70% of the baseline's LLM calls.