Metraipolis › archive › dc_2af1c97bbc4c
R-DEIM Net: An Efficient Rationale-Augmented Dual-Expert Interaction Model for Paraphrase Detection — arXiv:2609.30100
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.30100
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T04:28:02.235Z
- SHA-256 of the archived text
36f8d0561800366787d41bf6f5940314d5e7b10f52af7ad322b7f0a9cac7e7fe- Stable URL of this page
https://metraipolis.com/documento/dc_2af1c97bbc4c
How to cite
R-DEIM Net: An Efficient Rationale-Augmented Dual-Expert Interaction Model for Paraphrase Detection — arXiv:2609.30100 — arXiv (en), https://arxiv.org/abs/2609.30100 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2af1c97bbc4c (Metraipolis archive dc_2af1c97bbc4c, sha256 36f8d0561800366787d41bf6f5940314d5e7b10f52af7ad322b7f0a9cac7e7fe)
BibTeX
@misc{dc2af1c97bbc4c,
title = {R-DEIM Net: An Efficient Rationale-Augmented Dual-Expert Interaction Model for Paraphrase Detection — arXiv:2609.30100},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.30100},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2af1c97bbc4c; sha256 36f8d0561800366787d41bf6f5940314d5e7b10f52af7ad322b7f0a9cac7e7fe},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30100. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Recent advances in paraphrase detection reveal a fundamental trade-off: large language models achieve high accuracy but require high computation, while efficient Siamese-BERT variants offer practical scalability with reduced transparency in rationale generation. We present R-DEIM Net, a 76M-parameter dual-expert architecture exploring whether moderate-scale models can achieve competitive accuracy on paraphrase detection while enabling human-readable rationale generation. The architecture combines two specialized components: an Interaction Expert that captures token-level similarity patterns through multi-scale 2D convolutions and attention head allowing variable input length, and a Reasoning Expert that uses a Flan-T5-small decoder to generate rationales as auxiliary supervision. Rather than re-encoding generated text, we extract and pool decoder hidden states as complementary features for classification. On the Quora Question Pairs dataset, R-DEIM Net achieves 90.07\% accuracy and 90.16\% F1-score via 10-fold cross-validation. This represents competitive performance with strong transformer-based baselines (e.g., MFAE BERT: 90.54\% accuracy) and recent large language model based approaches (LLaMA-70B) while using a substantially smaller parameter budget. The model generates rationales alongside predictions, providing potential for auxiliary human-readable descriptions. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.30100 — 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 2026-10-08T03:50:58.239Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).