Metraipolis › archive › dc_273f83e5fc7d
Beyond the Context Window: An Adaptive Entropy-Based Routing Framework for Hybrid Retrieval and Long-Context Language Models — arXiv:2609.35831
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.35831
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T16:56:20.874Z
- SHA-256 of the archived text
3cf6f685c47da65ac43771c069f0a0d5d31e833d7d3faac650271a8da085cc42- Stable URL of this page
https://metraipolis.com/documento/dc_273f83e5fc7d
How to cite
Beyond the Context Window: An Adaptive Entropy-Based Routing Framework for Hybrid Retrieval and Long-Context Language Models — arXiv:2609.35831 — arXiv (en), https://arxiv.org/abs/2609.35831 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_273f83e5fc7d (Metraipolis archive dc_273f83e5fc7d, sha256 3cf6f685c47da65ac43771c069f0a0d5d31e833d7d3faac650271a8da085cc42)
BibTeX
@misc{dc273f83e5fc7d,
title = {Beyond the Context Window: An Adaptive Entropy-Based Routing Framework for Hybrid Retrieval and Long-Context Language Models — arXiv:2609.35831},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.35831},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_273f83e5fc7d; sha256 3cf6f685c47da65ac43771c069f0a0d5d31e833d7d3faac650271a8da085cc42},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35831. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Modern large language models now support context windows of more than one million tokens, which has raised the question of whether retrieval-augmented generation (RAG) is still necessary. Pure long-context (LC) processing is expensive and is known to under-attend to information placed in the middle of long inputs, while pure RAG is fast but bounded by retrieval quality and prone to errors when retrieved chunks are partially relevant or contradictory. We propose the Entropy-Driven Adaptive Router (EDAR), a framework that decides at inference time whether to answer a query from retrieved chunks or to escalate it to full long-context processing. The decision uses the predictive entropy of the token-level probability distribution computed over the first few generated tokens of the RAG response. The entropy threshold is selected on a held-out validation set by sweeping cost against accuracy. Experiments compare EDAR against pure-RAG and pure-LC baselines on LongBench v2 and Infinity-Bench. Predictive entropy correlates strongly with hallucination rate on a held-out set of 2,000 generations (Pearson r = 0.85, 95% CI [0.83, 0.87]). On the long-context benchmarks, EDAR retains 97.4% of the accuracy of the pure long-context baseline while reducing total token expenditure by 70.7%, escalating only 18.2% of incoming queries. The accuracy gap between EDAR and the pure long-context system is not statistically distinguishable from zero at standard sample sizes. Predictive entropy is a useful model-internal signal for routing between RAG and long-context inference, and a threshold-based hybrid system can recover most of the accuracy of long-context models at a small fraction of the cost. The framework does not depend on a specific retriever or LC backbone, and it does not require additional supervision beyond what is normally produced during decoding. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.35831 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T16:27:03.162Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).