# Apple — RISED: Rubrics for Agentic Multi-Environment Selection and Self-Distillation

- Company: Apple (apple.com)
- Announced: 2026-10-06
- Category: research-paper
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://machinelearning.apple.com/research/rised-multi-environment-selection
- Record: https://forck.live/items/16309-rised-rubrics-for-agentic-multi-environment-selection-and-self-distillation
- Subject: Machine Learning Research

Apple researchers propose RISED, a method for training a single LLM agent across diverse interactive environments. An LLM judge tags rollouts using a predefined rubric vocabulary, guiding data selection and providing token-level supervision via self-distillation. RISED achieves the highest mean pass rate across environments and ranks first or second in every individual environment.

## Evidence

Verbatim from https://machinelearning.apple.com/research/rised-multi-environment-selection:

> An LLM judge tags each rollout using a predefined rubric vocabulary shared across environments. The resulting profiles guide the selection of data that aligns with the overall behavioural composition of the mixed-environment batch while limiting overlap with already-selected data.

---

Record: https://forck.live/items/16309-rised-rubrics-for-agentic-multi-environment-selection-and-self-distillation
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md