Metraipolis › archive › dc_37b09f95fb98

The Attribution-Compression Frontier in Retrieval-Augmented Generation — arXiv:2609.14245

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.14245
License (as archived)
CC0-1.0
Archived
2026-10-15T22:51:01.058Z
SHA-256 of the archived text
048c6040eade20d6195f54362d7bf12d691e25d5bf39dab1ac797f221ab7fb1e
Stable URL of this page
https://metraipolis.com/documento/dc_37b09f95fb98

How to cite

The Attribution-Compression Frontier in Retrieval-Augmented Generation — arXiv:2609.14245 — arXiv (en), https://arxiv.org/abs/2609.14245 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_37b09f95fb98 (Metraipolis archive dc_37b09f95fb98, sha256 048c6040eade20d6195f54362d7bf12d691e25d5bf39dab1ac797f221ab7fb1e)
BibTeX
@misc{dc37b09f95fb98,
  title = {The Attribution-Compression Frontier in Retrieval-Augmented Generation — arXiv:2609.14245},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.14245},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_37b09f95fb98; sha256 048c6040eade20d6195f54362d7bf12d691e25d5bf39dab1ac797f221ab7fb1e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.14245. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Context compression reduces generator input in retrieval-augmented generation, but answer quality alone does not characterize citation attribution. We measure citation attribution across compression methods and budgets, comparing reranking, extractive selection, abstractive summarization, token pruning, and an extract-cluster-rewrite construction on ASQA and QASPER under a fixed generator and primary entailment evaluator. On ASQA at a nominal 0.25 budget (achieved compression 0.08), a RECOMP-style compressor's citations score 0.86 precision against its summaries but 0.12 against source spans under our re-attributability protocol. These estimates depend on a shared NLI model for span recovery and citation scoring and lack independent human calibration. Extractive selection's observed grounded precision ranges from 0.43 to 0.49 across nominal budgets from one-half to one-tenth of the ASQA context, while answer quality declines. For the same RECOMP setting, claim verification after source recovery yields an unsupported rate of 0.88 versus 0.17 when checking summaries. This gap persists beyond structural rejection of missing provenance, but remains evaluator-dependent. A 200-question TRUE T5-XXL audit also finds emitted--grounded gaps under both fixed and recomputed source mappings, without establishing human-calibrated support rates.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.14245 — 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 2026-10-15T22:29:03.647Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).