Metraipolis › archive › dc_533ca4e2c63f

Sieve and Sage: Efficient Distraction Filtering for Reliable RALM Abstention — arXiv:2609.35794

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.35794
License (as archived): CC0-1.0
Archived: 2026-10-07T16:56:16.314Z
SHA-256 of the archived text: 293312482b6183b8d635a7e590350d54140591b7039f07e307999e18934cd386
Stable URL of this page: https://metraipolis.com/documento/dc_533ca4e2c63f

How to cite

Sieve and Sage: Efficient Distraction Filtering for Reliable RALM Abstention — arXiv:2609.35794 — arXiv (en), https://arxiv.org/abs/2609.35794 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_533ca4e2c63f (Metraipolis archive dc_533ca4e2c63f, sha256 293312482b6183b8d635a7e590350d54140591b7039f07e307999e18934cd386)

BibTeX

@misc{dc533ca4e2c63f,
  title = {Sieve and Sage: Efficient Distraction Filtering for Reliable RALM Abstention — arXiv:2609.35794},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.35794},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_533ca4e2c63f; sha256 293312482b6183b8d635a7e590350d54140591b7039f07e307999e18934cd386},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35794. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Just as Socrates recognized the limits of his own knowledge, Retrieval-Augmented Language Models (RALMs) should learn to abstain when the retrieved evidence cannot support a reliable response. Existing approaches largely rely on monolithic LLMs to handle heterogeneous retrieval failures in a single step, resulting in limited abstention performance and high computational costs. We instead decompose retrieval failures into two distinct states: (i) the unanswerable state, where the required evidence is absent, and (ii) the distracted state, where relevant evidence is mixed with conflicting, negated, or adversarial information. Based on this decomposition, we introduce a lightweight module (Sieve) that screens retrieved document sets for distracting evidence before invoking a costly LLM (Sage) for grounded generation and abstention. Evaluated across both general and high-stakes expert domains, our Sieve and Sage framework preemptively detects distracting noise, improving system accuracy by up to 69.4 percentage points and Macro-F1 by 55.2 percentage points compared to one-stage baselines. Furthermore, it achieves up to a 1.99x speedup, establishing a highly efficient and reliable abstention pipeline for RALM with abstention.

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) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.35794 — 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 2026-10-07T16:27:03.145Z 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).