textfacebook/kilt_tasksnlpbenchmarkquestion-answeringfact-checkingentity-linkingragwikipediaretrievalevaluationkilt

KILT: Knowledge-Intensive Language Tasks Benchmark

Name: KILT: Knowledge-Intensive Language Tasks Benchmark
Creator: DataBazaar
Keywords: facebook/kilt_tasks, nlp, benchmark, question-answering, fact-checking, entity-linking, rag, wikipedia, retrieval, evaluation, kilt

About this data

Facebook AI's KILT benchmark — 11 datasets across fact-checking, entity linking, slot filling, open-domain QA, and dialog generation, all grounded in a unified Wikipedia snapshot. MIT licensed, parquet format, 1M-10M examples.

Schema

Name	Type	Description
id	VARCHAR	Unique example identifier string
input	VARCHAR	Query, claim, or dialog context text
meta	STRUCT(left_context VARCHAR, mention VARCHAR, right_context VARCHAR, partial_evidence STRUCT(start_paragraph_id INTEGER, end_paragraph_id INTEGER, title VARCHAR, section VARCHAR, wikipedia_id VARCHAR, meta STRUCT(evidence_span VARCHAR[]))[], obj_surface VARCHAR[], sub_surface VARCHAR[], subj_aliases VARCHAR[], template_questions VARCHAR[])	Task-specific metadata including entity mention context, surface forms, aliases, template questions, and partial Wikipedia evidence references
output	STRUCT(answer VARCHAR, meta STRUCT(score INTEGER), provenance STRUCT(bleu_score FLOAT, start_character INTEGER, start_paragraph_id INTEGER, end_character INTEGER, end_paragraph_id INTEGER, meta STRUCT(fever_page_id VARCHAR, fever_sentence_id INTEGER, annotation_id VARCHAR, yes_no_answer VARCHAR, evidence_span VARCHAR[]), section VARCHAR, title VARCHAR, wikipedia_id VARCHAR)[])[]	List of gold answers with text and supporting Wikipedia provenance (document ID, section, character spans, evidence metrics)

Sample Data

Preview a sample of the data before downloading.

Free

Open dataset

Quality: No ratings

0 downloads

Seller: DataBazaar

Agent? No sign-up needed →

For AI Agents

Via MCP Server

# 1. Add to your agent's MCP config (claude_desktop_config.json or similar):
{
  "mcpServers": {
    "databazaar": { "command": "npx", "args": ["databazaar-mcp"] }
  }
}

# 2. Your agent can then call:
search_datasets({ query: "KILT: Knowledge-Intensive Lang" })
// Found: 8997fb69-ff6d-48af-926e-b3840702fd18
get_download_url({ dataset_id: "8997fb69-ff6d-48af-926e-b3840702fd18" })  // free — no API key needed

Via REST API

# Free dataset — no API key required:
curl https://api.databazaar.io/datasets/8997fb69-ff6d-48af-926e-b3840702fd18/download-url