Metraipolis › archive › dc_524819b5bdcd
Pairwise Ranking Outperforms Single-Action RL for Offline Explanation Selection: A Practical Lesson — arXiv:2608.18531
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.18531
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T13:44:31.009Z
- SHA-256 of the archived text
62b6cc1b4208d8fe49f9c2022d4f36ef38ec1084d7a4213c4b0592eb17548d76- Stable URL of this page
https://metraipolis.com/documento/dc_524819b5bdcd
How to cite
Pairwise Ranking Outperforms Single-Action RL for Offline Explanation Selection: A Practical Lesson — arXiv:2608.18531 — arXiv (en), https://arxiv.org/abs/2608.18531 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_524819b5bdcd (Metraipolis archive dc_524819b5bdcd, sha256 62b6cc1b4208d8fe49f9c2022d4f36ef38ec1084d7a4213c4b0592eb17548d76)
BibTeX
@misc{dc524819b5bdcd,
title = {Pairwise Ranking Outperforms Single-Action RL for Offline Explanation Selection: A Practical Lesson — arXiv:2608.18531},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.18531},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_524819b5bdcd; sha256 62b6cc1b4208d8fe49f9c2022d4f36ef38ec1084d7a4213c4b0592eb17548d76},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.18531. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Industrial explainable-recommendation systems built on LLMs incur a substantial serving cost: each request triggers an LLM generation, with latency in the hundreds of milliseconds and cost that scales linearly with traffic. We separate generation from selection: explanations are produced ahead of time as a frozen candidate pool (six prompt styles, two commodity LLMs), and a small CPU-resident selector picks one at request time. The stack needs no GPU and returns in under 100 ms. Our primary benchmark is a 2,958-pair XRec Google Local subset, evaluating six offline-pool selectors (LambdaRank, PPO, GRPO, DPO, teacher-student distillation) and three KG-path selectors (random walks, edge-disjoint enumeration, MMR-reranked paths). A 300-pair MovieLens-1M split with Claude-Sonnet-4.5 references serves as an internal cross-dataset check, since no public benchmark exists for this setting. All variants use the same BERTScore-F1 protocol as XRec and G-Refer, averaged across five seeds. LambdaRank reaches F1 = 0.500 on Google Local, exceeding both G-Refer and XRec, and F1 = 0.329 on the MovieLens-1M check. With seed variance below 0.003 F1, the ordering is reliable: pairwise learning-to-rank outperforms single-action RL (PPO, GRPO, DPO), which use only one labelled candidate per rollout, leaving K-1 labels unused. The KG-path family targets a different objective: all three variants reach USR = 1.000 on Google Local and 0.997-1.000 on MovieLens-1M, since per-request path grounding yields a unique output per query, avoiding template-collapse failures affecting cached-LLM outputs. A generator-pool study comparing Claude 3 Haiku and Claude Haiku 4.5 shows small F1 shifts (0.001-0.006) while preserving selector ranking: selector and generator can be evaluated independently, though absolute F1 depends on the generator. End-to-end build cost is near $15 on commodity hardware. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-20 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.18531 — 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 2026-10-18T13:41:17.865Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-20?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).