A two-stage fine-tuning and reinforcement-learning method makes LLMs generate token-efficient natural-language search plans, reporting strong accuracy gains on financial and news search benchmarks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Enhancing LLMs' Reasoning-Intensive Multimedia Search Capabilities through Fine-Tuning and Reinforcement Learning
A two-stage fine-tuning and reinforcement-learning method makes LLMs generate token-efficient natural-language search plans, reporting strong accuracy gains on financial and news search benchmarks.