Metraipolis › archive › dc_66eada0f89ab

ANI: Adaptive Numerical Injection for Unifying Semantic and Arithmetic Representations in Numerical Reasoning — arXiv:2609.39294

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.39294
License (as archived)
CC0-1.0
Archived
2026-10-07T12:42:37.760Z
SHA-256 of the archived text
355f61fd37246f7d48cfb01a4c916873efd6fedefdf44a0b527567bd6390222f
Stable URL of this page
https://metraipolis.com/documento/dc_66eada0f89ab

How to cite

ANI: Adaptive Numerical Injection for Unifying Semantic and Arithmetic Representations in Numerical Reasoning — arXiv:2609.39294 — arXiv (en), https://arxiv.org/abs/2609.39294 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_66eada0f89ab (Metraipolis archive dc_66eada0f89ab, sha256 355f61fd37246f7d48cfb01a4c916873efd6fedefdf44a0b527567bd6390222f)
BibTeX
@misc{dc66eada0f89ab,
  title = {ANI: Adaptive Numerical Injection for Unifying Semantic and Arithmetic Representations in Numerical Reasoning — arXiv:2609.39294},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.39294},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_66eada0f89ab; sha256 355f61fd37246f7d48cfb01a4c916873efd6fedefdf44a0b527567bd6390222f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39294. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Precise numerical reasoning with Large Language Models (LLMs) is essential for expanding their applicability to complex real-world tasks. However, text-based tokenization often fragments numbers, significantly hindering precise arithmetic reasoning. Meanwhile, numerical embeddings, despite arithmetic precision, rely on context-agnostic substitution that disregards the semantic role of numbers as identifiers. To combine the complementary strengths, we propose \textbf{ANI (Adaptive Numerical Injection)}, a hybrid framework that governs the selective injection of numerical features based on the semantic context. By employing a context-aware gating mechanism, we selectively inject numerical embeddings (specifically FoNE) into the latent space, explicitly preserving nominal identifiers while enhancing quantitative operands. Through extensive evaluations across various LLMs, we demonstrate that ANI enhances MATH performance by 9.5 points over the official reference model, while maintaining robust performance on general linguistic benchmarks.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.39294 — 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 2026-10-07T12:14:13.748Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).