# Apple — From Preferences to Principles: Rubric-Based Alignment for Grounded Knowledge Answers

- Company: Apple (apple.com)
- Announced: 2026-08-27
- Category: research-paper
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://machinelearning.apple.com/research/rubric-based-alignment
- Record: https://forck.live/items/10305-from-preferences-to-principles-rubric-based-alignment-for-grounded-knowledge
- Subject: Machine Learning Research

Apple researchers introduce a rubric-based reward framework for open-domain question answering that generates query-specific rubrics grounded in retrieved evidence and decomposed into multiple quality dimensions. The approach improves over the instruction-tuned baseline by 6.5% and over flat rubric variants by 4% across three evaluation axes (composition, grounding, and instruction-following). Conditioning rubrics on retrieved evidence improves factual support, while decomposing rubrics into quality-specific dimensions further improves coherence, organization, and adherence to query requirements.

## Evidence

Verbatim from https://machinelearning.apple.com/research/rubric-based-alignment:

> Averaged across three evaluation axes (composition, grounding, and instruction-following), our approach improves over the instruction-tuned baseline by 6.5% and over flat rubric variants by 4%, with consistent gains across all evaluation datasets.

---

Record: https://forck.live/items/10305-from-preferences-to-principles-rubric-based-alignment-for-grounded-knowledge
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md