Context Recall

Available

rag.context_recall

Measures how fully the retrieved material covers the evidence needed to support the reference answer.

Contract

Field	Value
version	1.0.0
metric_type	pointwise
scorer_contract	per_row
direction	higher_is_better
entrypoint	aip_metrics_rag.answered.context_recall
target_kind	None

Required columns

• input_id
• prompt
• expected_output
• sut_response
• retrieved_context

Accepted schemas

[
  {
    "name": "gdi_text_v1",
    "task_types": [
      "multi_turn_rag",
      "single_turn_rag"
    ]
  }
]

Methodology

1. An LLM judge (Ragas context_recall) receives the question, the reference answer, and the retrieved contexts.
2. Ragas breaks the reference answer into individual claims.
3. The judge decides, per claim, whether it is attributable to the retrieved contexts.
4. The score is the fraction of attributable claims, normalised to 0-1.

Score semantics

Scores range 0-1, the fraction of reference-answer claims that are attributable to the retrieved context. A high score means the retrieval brought back the evidence the answer relies on; a low score means ground-truth information is missing. Higher is better.

Worked example

A reference answer with 4 claims where the retrieved context supports 3 of them scores 3/4 = 0.75.

Configuration schema

{
  "properties": {
    "concurrency_limit": {
      "default": 10,
      "title": "Concurrency Limit",
      "type": "integer",
      "x-aip-param-role": "operational"
    },
    "inverted": {
      "default": false,
      "title": "Inverted",
      "type": "boolean",
      "x-aip-param-role": "scoring_metadata"
    },
    "lang": {
      "default": "en",
      "enum": [
        "en",
        "de",
        "fr"
      ],
      "title": "Lang",
      "type": "string"
    },
    "max": {
      "default": 1.0,
      "title": "Max",
      "type": "number",
      "x-aip-param-role": "scoring_metadata"
    },
    "metric_name": {
      "default": "context_recall",
      "title": "Metric Name",
      "type": "string",
      "x-aip-param-role": "scoring_metadata"
    },
    "min": {
      "default": 0.0,
      "title": "Min",
      "type": "number",
      "x-aip-param-role": "scoring_metadata"
    },
    "model": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": "gpt-4o",
      "title": "Model"
    },
    "system_type": {
      "const": "rag",
      "default": "rag",
      "title": "System Type",
      "type": "string",
      "x-aip-param-role": "judge_wiring"
    },
    "weight": {
      "default": 1.0,
      "title": "Weight",
      "type": "number",
      "x-aip-param-role": "scoring_metadata"
    }
  },
  "title": "InitializationConfig",
  "type": "object"
}

Execution

{
  "emits_metric_family": false,
  "function_name": null,
  "max_concurrency": 8,
  "processing_kind": "network",
  "stream_batch_size": 5000,
  "timeout_seconds": 21600
}

Complete manifest

accepts:
- name: gdi_text_v1
  task_types:
  - multi_turn_rag
  - single_turn_rag
config_schema:
  properties:
    concurrency_limit:
      default: 10
      title: Concurrency Limit
      type: integer
      x-aip-param-role: operational
    inverted:
      default: false
      title: Inverted
      type: boolean
      x-aip-param-role: scoring_metadata
    lang:
      default: en
      enum:
      - en
      - de
      - fr
      title: Lang
      type: string
    max:
      default: 1.0
      title: Max
      type: number
      x-aip-param-role: scoring_metadata
    metric_name:
      default: context_recall
      title: Metric Name
      type: string
      x-aip-param-role: scoring_metadata
    min:
      default: 0.0
      title: Min
      type: number
      x-aip-param-role: scoring_metadata
    model:
      anyOf:
      - type: string
      - type: 'null'
      default: gpt-4o
      title: Model
    system_type:
      const: rag
      default: rag
      title: System Type
      type: string
      x-aip-param-role: judge_wiring
    weight:
      default: 1.0
      title: Weight
      type: number
      x-aip-param-role: scoring_metadata
  title: InitializationConfig
  type: object
dependencies: []
description: How fully does the retrieved material cover the evidence needed for the
  answer?
direction: higher_is_better
display_name: Context Recall
entrypoint: aip_metrics_rag.answered.context_recall
execution:
  emits_metric_family: false
  function_name: null
  max_concurrency: 8
  processing_kind: network
  stream_batch_size: 5000
  timeout_seconds: 21600
kind: metric
manifest_version: '1'
max_prompt_slots: 0
metric_metadata:
  methodology:
  - An LLM judge (Ragas context_recall) receives the question, the reference answer,
    and the retrieved contexts.
  - Ragas breaks the reference answer into individual claims.
  - The judge decides, per claim, whether it is attributable to the retrieved contexts.
  - The score is the fraction of attributable claims, normalised to 0-1.
  score_semantics: Scores range 0-1, the fraction of reference-answer claims that
    are attributable to the retrieved context. A high score means the retrieval brought
    back the evidence the answer relies on; a low score means ground-truth information
    is missing. Higher is better.
  summary: Measures how fully the retrieved material covers the evidence needed to
    support the reference answer.
  worked_example: A reference answer with 4 claims where the retrieved context supports
    3 of them scores 3/4 = 0.75.
metric_type: pointwise
name: rag.context_recall
partition_types: []
required_columns:
- input_id
- prompt
- expected_output
- sut_response
- retrieved_context
required_kinds: []
scorer_contract: per_row
target_kind: null
unsupported_trace_shapes: []
version: 1.0.0