Metraipolis › archive › dc_53dcfd2d9736

TRACE: Deployable Tree-Relational Structure Enhancement for Oncology LLMs — arXiv:2609.35810

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.35810
License (as archived)
CC0-1.0
Archived
2026-10-07T16:24:42.918Z
SHA-256 of the archived text
eb3c2655df86a073247da4d454a62d744fade355c740d13dffd0f535af3c2fce
Stable URL of this page
https://metraipolis.com/documento/dc_53dcfd2d9736

How to cite

TRACE: Deployable Tree-Relational Structure Enhancement for Oncology LLMs — arXiv:2609.35810 — arXiv (en), https://arxiv.org/abs/2609.35810 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_53dcfd2d9736 (Metraipolis archive dc_53dcfd2d9736, sha256 eb3c2655df86a073247da4d454a62d744fade355c740d13dffd0f535af3c2fce)
BibTeX
@misc{dc53dcfd2d9736,
  title = {TRACE: Deployable Tree-Relational Structure Enhancement for Oncology LLMs — arXiv:2609.35810},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.35810},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_53dcfd2d9736; sha256 eb3c2655df86a073247da4d454a62d744fade355c740d13dffd0f535af3c2fce},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35810. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models are increasingly used in oncology applications, but their predictions are often weakly grounded in explicit medical structure. We present TRACE, a deployable tree-relational enhancement framework for oncology LLMs. TRACE separates expensive offline structure learning from lightweight online inference: oncology concepts and relations are organized into an updatable tree-relational structure, refined using LM-loss-derived evidence, and retrieved at inference time as compact prompt evidence. This design supports task-adaptive evidence selection without requiring supervised labels in the zero-shot setting. Across ten oncology classification tasks and one MedQuAD CancerGov QA benchmark, TRACE improves both label-free evaluation and supervised fine-tuning. Additional analyses show that TRACE improves over vanilla RAG and generic GraphRAG, remains useful under leakage-controlled METABRIC inputs, and produces interpretable evidence paths aligned with clinical reasoning. These results suggest that explicit, updatable medical structure is a practical path toward more accurate and auditable oncology LLM deployment.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.35810 — 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 2026-10-07T15:54:53.711Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).