RECAST: Learning to Compute the Right Context through Adaptive Evidence Routing
In this work, we introduce RECAST (Routing Evidence through Computation, Access, and Synthesized Tools), a learned framework that formulates evidence construction as a sequential decision process over heterogeneous retrieval and computation operations, allowing evidence to be actively derived rather than merely retrieved.
Key points
- Large language models are increasingly applied to tasks grounded in long, heterogeneous information sources.
- Conventional Retrieval-Augmented Generation (RAG) relies on fixed similarity-based retrieval, while agentic variants adapt queries and tool use but remain largely retrieval-centric.
- A lightweight RouterLM iteratively selects and formulates primitive operations or specifies customized operations for a frozen CompilerLM to translate into executable code.
- We train RouterLM with supervised fine-tuning (SFT) followed by group relative policy optimization (GRPO).
Sources (1)
- [1]RECAST: Learning to Compute the Right Context through Adaptive Evidence RoutingarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 05:51 PM
In this work, we introduce RECAST (Routing Evidence through Computation, Access, and Synthesized Tools), a learned framework that formulates evidence construction as a sequential decision process over heterogeneous retrieval and computation operations, allowing evidence to be actively derived rather than merely retrieved.
Large language models are increasingly applied to tasks grounded in long, heterogeneous information sources.
Extractive summary: sentences quoted from the sources.