Metraipolis › archive › dc_0d4fe7c2ab8a
IREA: Intermediate Representation-based Embedding Alignment for Normative RAG — arXiv:2610.04974
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.04974
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:28:12.189Z
- SHA-256 of the archived text
96b1a6e249fcdb8af74a09797e5962875236923e98bc802a27d183cf4016d6c9- Stable URL of this page
https://metraipolis.com/documento/dc_0d4fe7c2ab8a
How to cite
IREA: Intermediate Representation-based Embedding Alignment for Normative RAG — arXiv:2610.04974 — arXiv, https://arxiv.org/abs/2610.04974 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0d4fe7c2ab8a (Metraipolis archive dc_0d4fe7c2ab8a, sha256 96b1a6e249fcdb8af74a09797e5962875236923e98bc802a27d183cf4016d6c9)
BibTeX
@misc{dc0d4fe7c2ab8a,
title = {IREA: Intermediate Representation-based Embedding Alignment for Normative RAG — arXiv:2610.04974},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.04974},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0d4fe7c2ab8a; sha256 96b1a6e249fcdb8af74a09797e5962875236923e98bc802a27d183cf4016d6c9},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04974. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models (LLMs) have shown strong performance across various tasks, but they still struggle with questions involving ethical judgment. Previous studies have attempted to train LLMs on ethical standards, but the diversity and relativity of ethical norms make them difficult to fully internalize in model parameters. As an alternative, we introduce normative RAG, a retrieval-augmented approach that supports ethical judgment using external normative knowledge. Normative retrieval involves a distinct asymmetry between context rich narrative queries and generalized normative statements. Existing factual retrieval methods rely on query-only expansion into a document-like form, making them insufficient for resolving this asymmetry. Therefore, we propose Intermediate Representation-based Embedding Alignment (IREA), a bidirectional alignment method that maps both text types into a shared situation-behavior representation. This representation captures ethically salient contextual and behavioral information in a normalized form, reducing surface-level discrepancies and improving alignment in the embedding space. Experimental results show that IREA improves normative retrieval and downstream ethical judgment across multiple settings, demonstrating the effectiveness of bidirectional alignment for normative RAG. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2610.04974 — 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 2026-10-06T05:24:20.292Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).