Metraipolis › archive › dc_7c2b0dba2cfe
AdaTutoRank: Learning to Rerank Document Sets via Adaptive Tutoring Optimization for RAG and Deep Research — arXiv:2609.32472
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.32472
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T21:50:09.541Z
- SHA-256 of the archived text
13a90a5f044431b6913e9b45d58f3f8b250eeb9772c65bca28d017716b36790f- Stable URL of this page
https://metraipolis.com/documento/dc_7c2b0dba2cfe
How to cite
AdaTutoRank: Learning to Rerank Document Sets via Adaptive Tutoring Optimization for RAG and Deep Research — arXiv:2609.32472 — arXiv (en), https://arxiv.org/abs/2609.32472 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7c2b0dba2cfe (Metraipolis archive dc_7c2b0dba2cfe, sha256 13a90a5f044431b6913e9b45d58f3f8b250eeb9772c65bca28d017716b36790f)
BibTeX
@misc{dc7c2b0dba2cfe,
title = {AdaTutoRank: Learning to Rerank Document Sets via Adaptive Tutoring Optimization for RAG and Deep Research — arXiv:2609.32472},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.32472},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7c2b0dba2cfe; sha256 13a90a5f044431b6913e9b45d58f3f8b250eeb9772c65bca28d017716b36790f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32472. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Document rerankers determine what evidence reaches the downstream model in RAG and deep research, yet mainstream rerankers select by relevance matching, and individually relevant documents rarely constitute the complete, complementary, non-redundant set a complex information need demands. Prior work rewards a set by its aggregate rubric score, shifting the objective from ranking documents to composing sets. Yet that score is one scalar shared by every document in the set, so the supervision is sparse: a redundant document is rewarded with the rest whenever the set scores well, and a decisive one penalized with the rest whenever it does not; credit assignment leaves contributors indistinguishable from free riders. On-policy distillation could densify this supervision, but existing methods give every rollout the same fixed guidance, too prescriptive for strong rollouts and too abstract for weak ones. We therefore propose AdaTutoRank, a setwise reranker trained with Adaptive Tutoring Optimization (ATO) under a three-level hierarchy of nine rubric dimensions, which supplies silver labels for the cold start, rewards for reinforcement learning, and hints for distillation. ATO draws three hint forms of increasing specificity from the policy's own frozen snapshot: the rubrics alone, a self-selector's sibling-set chosen under rubrics, and a self-reflector's reflection contrasting the rollout with that sibling-set; each rollout receives the form matched to its quality. Re-scoring that rollout under the hint-conditioned frozen teacher and the hint-free snapshot distills the hint's effect into a token-level advantage that complements the group-relative outcome advantage. Across ten benchmarks spanning RAG, deep research, and setwise evaluation, AdaTutoRank attains the best overall performance while issuing fewer retrieval calls. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.32472 — 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 2026-10-07T21:27:08.433Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).