Files
myWiki/reviews/rubrics-survey-review-2026-06-27.md

49 lines
3.4 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

---
title: "Rubrics Survey Review"
created: 2026-06-27
updated: 2026-06-27
type: review
paper: "rubrics-survey-2026"
---
# Review: The Rules of the Game — A Survey of Rubrics for LLMs
## 📌 基本信息
- **标题**: The Rules of the Game: A Survey of Rubrics for Large Language Models
- **作者**: Wenhan Liu, Jiajie Jin, Zhaoheng Huang, Tongyu Wen, Guanting Dong, Ziliang Zhao, Yutao Zhu, Zhicheng Dou, Ji-Rong Wen (Renmin University of China)
- **日期**: 2026-05-22
- **领域**: LLM Evaluation, Reward Modeling, Rubric
- **Wiki 添加时间**: 2026-06-27
## 🎯 核心概念
1. **[[rubrics-for-llms|Rubrics for LLMs]]** — 结构化、显式的多维度评估标准,填补了传统标量奖励和 LLM judge 在开放任务中的不足
2. **[[rubric-construction|Rubric Construction]]** — 四大构建范式:直接生成 → 对比生成 → 迭代精炼 → 在线协同演化,从静态到动态递进
3. **[[rubric-aggregation|Rubric Aggregation]]** — 逐项评分后如何合并为总分(直接平均 / 加权求和 / 隐式聚合)
4. **[[rubric-based-reward-modeling|Rubric-Based Reward Modeling]]** — rubric 从评估工具变为训练信号policy model RL × reward model 训练的三种角色
5. **[[rubric-driven-evaluation|Rubric-Driven Evaluation]]** — 覆盖通用任务(推理/研究/Agent/对齐)和领域特定任务的全景评估体系
6. **[[rubric-personalization|Rubric Personalization]]** — 从通用标准到个体偏好的过渡PREFINE 的伪用户 agent 方法
7. **[[rubric-safety|Rubric Safety]]** — RIPD 攻击面rubric 本身可能成为隐秘的偏好偏移向量
## 🔗 概念网络
- **核心连接**: [[rubrics-for-llms]] ↔ [[rubric-construction]] ↔ [[rubric-based-reward-modeling]] ↔ [[rubric-driven-evaluation]]
- **跨概念连接**: [[rubrics-for-llms]] ↔ [[llm-as-a-judge]] / [[rlvr-unified-framework|RLVR]] / [[reward-hacking]]
- **开放问题链**: [[rubric-safety]] → RIPD → [[reward-hacking]] → [[rubric-personalization]](偏好推断)
- **复用已有概念**: [[reward-hacking]], [[rlvr-unified-framework]]
## 📚 Wiki 集成
- **新增页面**: 11 个1 论文 + 1 raw + 9 概念)
- **9 概念**: rubrics-for-llms, rubric-construction, rubric-aggregation, rubric-based-reward-modeling, rubric-driven-evaluation, rubric-personalization, rubric-safety, llm-as-a-judge
- **链接密度**: 论文主页引用 10+ 概念,概念间高密度交叉引用
## 💡 关键洞察
1. **Rubric 是评估基础设施的"中间层"** — 它不是替代 reward model 或 LLM judge而是为两者提供显式、可编辑的评估标准。这与 sz 对 Agent Harness 多维度约束的思考高度对齐rubric 正是将"质量维度"可操作化的接口。
2. **从静态到协同演化的范式演进** — rubric construction 从一次性生成Direct/Contrastive→ 迭代精炼Refinement→ 在线协同演化Online/Co-evolving与 Agent 训练中的 curriculum learning、self-play 趋势共振。最激进的 Online 方法DR-Tulu, SibylSense本质上是让 rubric 和 policy 互相塑造。
3. **Rubric 安全是一个被低估的系统性风险** — RIPD 揭示的 pipeline 级联效应rubric → judge → preference data → policy → 不可检测的漂移)与 reward hacking 一样危险但更隐蔽,因为 rubric 修改"看起来合理"。这直接关联到 Agent 记忆系统评估的可靠性问题。