Metraipolis › archive › dc_3d4e60d64a4a

A Survey on Rubric-Guided Reinforcement Learning for Language Models — arXiv:2608.27505

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.27505
License (as archived): CC0-1.0
Archived: 2026-10-17T10:47:46.393Z
SHA-256 of the archived text: e5ac476961b89478345a385332f07b3424c87c1ded2bd54d4bd59b36cf712104
Stable URL of this page: https://metraipolis.com/documento/dc_3d4e60d64a4a

How to cite

A Survey on Rubric-Guided Reinforcement Learning for Language Models — arXiv:2608.27505 — arXiv (en), https://arxiv.org/abs/2608.27505 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3d4e60d64a4a (Metraipolis archive dc_3d4e60d64a4a, sha256 e5ac476961b89478345a385332f07b3424c87c1ded2bd54d4bd59b36cf712104)

BibTeX

@misc{dc3d4e60d64a4a,
  title = {A Survey on Rubric-Guided Reinforcement Learning for Language Models — arXiv:2608.27505},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.27505},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3d4e60d64a4a; sha256 e5ac476961b89478345a385332f07b3424c87c1ded2bd54d4bd59b36cf712104},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.27505. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reinforcement learning from human feedback (RLHF) has become the dominant paradigm for aligning large language models (LLMs) with human preferences. However, traditional RLHF relies on scalar reward signals that lack interpretability and fail to capture the multifaceted nature of response quality. Rubric-guided reinforcement learning addresses these limitations by introducing structured, interpretable evaluation criteria, or rubrics, as the backbone of reward design, feedback generation, and policy optimization. In this survey, we introduce a Bayesian framework that defines constitutions as prior distributions $P(R)$ over evaluation criteria and rubrics as conditional instantiations $R_x \sim P(R|x)$. Under this unified view, we present a taxonomy of rubric-guided RL along the prior-posterior axis, covering constitutional AI, instance-specific rubrics, process-level supervision, self-evolving rubrics, and their agentic and multimodal extensions. Furthermore, as rubrics are natural-language artifacts, we present a linguistic analysis of how granularity trade-offs, semantic drift, and linguistic reward hacking impact alignment reliability, identifying key open problems for future research.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-31 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.27505 — 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 2026-10-17T10:33:42.445Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).