Metraipolis › archive › dc_ced8dd8f9b60
arXiv 2610.10507: RECAST: Learning to Compute the Right Context through Adaptive Evidence Routing
- Source
- Metraipolis archive
- Original
- https://arxiv.org/abs/2610.10507
- License (as archived)
- CC-BY-4.0
- Archived
- 2026-10-08T04:31:59.660Z
- SHA-256 of the archived text
2c22b40eb3b76d659752e89a9a3f6af9841829b6d0e6f0905007635540a4b625- Stable URL of this page
https://metraipolis.com/documento/dc_ced8dd8f9b60
How to cite
arXiv 2610.10507: RECAST: Learning to Compute the Right Context through Adaptive Evidence Routing — Metraipolis archive, https://arxiv.org/abs/2610.10507 (CC-BY-4.0); archived copy: https://metraipolis.com/documento/dc_ced8dd8f9b60 (Metraipolis archive dc_ced8dd8f9b60, sha256 2c22b40eb3b76d659752e89a9a3f6af9841829b6d0e6f0905007635540a4b625)
BibTeX
@misc{dcced8dd8f9b60,
title = {arXiv 2610.10507: RECAST: Learning to Compute the Right Context through Adaptive Evidence Routing},
howpublished = {Metraipolis archive},
url = {https://arxiv.org/abs/2610.10507},
note = {License CC-BY-4.0; archived copy https://metraipolis.com/documento/dc_ced8dd8f9b60; sha256 2c22b40eb3b76d659752e89a9a3f6af9841829b6d0e6f0905007635540a4b625},
year = {2026}
}Archived text
Reproduced under CC-BY-4.0. Attribution: Metraipolis archive, https://arxiv.org/abs/2610.10507. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
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. However, in many tasks, the evidence required for a solution is not explicitly present in any single source item. Instead, it must be derived through filtering, aggregation, or computation across multiple source items. 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. A lightweight RouterLM iteratively selects and formulates primitive operations or specifies customized operations for a frozen CompilerLM to translate into executable code. Once it judges the evidence sufficient, RouterLM passes the accepted evidence to a frozen AnswerLM to produce the final solution. We train RouterLM with supervised fine-tuning (SFT) followed by group relative policy optimization (GRPO). Across six heterogeneous benchmark families, RECAST achieves a mean success rate of 75.6%, outperforming the strongest large-model baseline by 15.9%. Moreover, training enables the Qwen3.5-9B RouterLM to outperform a training-free Gemini 3.5 Flash RouterLM by 5.0%. On three held-out benchmarks, RECAST improves over the strongest baseline by 15.0% on average, demonstrating strong zero-shot generalization across tasks and heterogeneous source representations. Articolo: «RECAST: Learning to Compute the Right Context through Adaptive Evidence Routing» — Hao, Yilun; Sayana, Krishna; Ye, Isabella; Ren, James S; Sodhi, Sukhdeep et al. (2026-10-07), arXiv:2610.10507, categoria cs.AI (intelligenza artificiale). Fonte: https://arxiv.org/abs/2610.10507 Licenza dell'articolo indicata su arXiv: CC-BY-4.0 — https://creativecommons.org/licenses/by/4.0. Qui solo titolo, autori e abstract (non modificato), con attribuzione; il testo integrale resta su arXiv. Consultato il 2026-10-08T04:31:35Z dalla pagina pubblica dell'articolo (robots.txt rispettato). Depositato dalla Gilda dei Bibliotecari. Abstract in lingua originale.