---
title: "Rubric-Driven Evaluation"
created: 2026-06-27
updated: 2026-06-27
type: concept
tags:
  - rubric
  - evaluation
  - benchmark
  - llm
sources:
  - "rubrics-survey-2026"
---

# Rubric-Driven Evaluation

## 定义

使用结构化的 rubric 作为评估框架，对 LLM 在各类任务上的表现进行**多维度、可解释**的定量评估。

## 通用任务评估 (General-Task)

### Reasoning Capability
- MathCheck, SedarEval, RefGrader, RubricCode, TRACE, ProfBench, MoReBench
- 重点：数学正确性、推理步骤有效性、逻辑一致性

### Deep Research & Open-ended Generation
- HelloBench, WritingBench, DeepResearchBench, DeepResearchBenchII, DEER, ResearchRubrics, PencilsDown
- 重点：证据支撑、主题覆盖度、事实一致性、结构清晰度

### General Agent Capability
- AgentBoard, AdaRubric, TRAJECT-Bench, MCP-Universe, MultiChallenge, Scribe, AstraBench, PaperBench
- 重点：子目标完成、工具选择正确性、轨迹级评估

### Alignment Evaluation
- FLASK, G-Eval, InFoBench, AdvancedIF, WildBench, JudgeBench, StrongREJECT
- 重点：指令遵循、安全合规、偏好对齐

## 领域特定任务评估 (Domain-Specific)

### Intermediate Trajectories（中间轨迹）
- PaperBench, MoReBench, ProfBench, RubricCode, Scribe, MedMT-Bench
- 评估模型的推理/规划**过程**而非仅最终输出

### Final Outputs（最终输出）
- HealthBench, DeepResearchBenchII, PRBench, TechImage-Bench, LongShotBench, PresentBench, SpeechL2, LLM-RUBRIC
- 涵盖医疗 QA、多模态生成、代码生成、视频理解等

## 参考

- [[rubrics-for-llms|Rubrics for LLMs]]
- [[rubrics-survey-2026|Rubrics Survey (2026)]]
- [[rubric-aggregation]]