Adding a Bayesian source memory for market-feedback adaptive retrieval to a frozen LLM improves macro-F1 from 0.438 to 0.471 and portfolio Sharpe from 0.52 to 0.84 in point-in-time financial event-impact prediction.
arXiv:2505.19819 [cs.CL] https://arxiv.org/abs/2505.19819
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Point-in-Time Financial RAG with Frozen LLMs and Market-Feedback Adaptive Retrieval
Adding a Bayesian source memory for market-feedback adaptive retrieval to a frozen LLM improves macro-F1 from 0.438 to 0.471 and portfolio Sharpe from 0.52 to 0.84 in point-in-time financial event-impact prediction.