Triple

T1146142
Position Surface form Disambiguated ID Type / Status
Subject Chengdu E23568 entity
Predicate famousDish P6863 FINISHED
Object mapo tofu
Mapo tofu is a classic Sichuan Chinese dish featuring soft tofu and minced meat in a spicy, numbing chili and Sichuan peppercorn sauce.
 E131341 NE FINISHED

How this triple was built (4 steps)

Every LLM step that produced this triple, in pipeline order — named-entity classification, the disambiguation choices (the exact options shown, with the pick highlighted), and the generated description. The batch + timestamp of each is in the Provenance table below.

NER Named-entity recognition gpt-5-mini
Instruction
Given a phrase, classify it is english named entity (e.g., persons, organizations, works of art) in Latin script, or not (e.g., literals, dates, URLs, verbose phrases). For disambiguation, the statement where the phrase occurs as object is also given. Please return a JSON object with `phrase` (string, the phrase being analyzed) and `is_ne` (boolean, indicating whether the phrase is a Named Entity).
Input
Phrase: mapo tofu | Statement: [Chengdu, famousDish, mapo tofu]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07 
Target entity: mapo tofu
Context triple: [Chengdu, famousDish, mapo tofu]
  • A. Mandu
    Mandu is a historic fortified city in central India renowned for its Afghan-era architecture, romantic legends, and scenic hilltop setting.
  • B. Okonomiyaki
    Okonomiyaki is a savory Japanese pancake made from a batter of flour, cabbage, and various mix-ins, typically grilled and topped with a rich sauce, mayonnaise, and bonito flakes.
  • C. bibimbap
    Bibimbap is a traditional Korean rice dish topped with assorted vegetables, meat, egg, and spicy gochujang sauce, all mixed together before eating.
  • D. Paippalada
    Paippalada is one of the principal ancient recensions (śākhās) of the Atharvaveda, preserved mainly in the Paippalāda tradition of Vedic scholarship.
  • E. Taro
    Taro is a common Japanese male given name, often written with kanji meaning "eldest son" or similar traditional connotations.
  • F. None of above. chosen
  • G. Unsure - the case is ambiguous/there is not enough information to decide.
NEDg Description generation gpt-5.1
Instruction
Generate a one-sentence description of the target entity. 
You are given a context triple in the form (subject, predicate, object), where the object is the target entity. 
# Instructions
Use the triple to infer relevant information about the entity. Describe the entity based on what is most defining, well-known. 
Avoid repeating the information from the triple, unless really essential.
# Response Format
Return only the sentence: "Description: [one-sentence description of the target entity]"
Input
Entity: mapo tofu
Triple: [Chengdu, famousDish, mapo tofu]
Generated description
Mapo tofu is a classic Sichuan Chinese dish featuring soft tofu and minced meat in a spicy, numbing chili and Sichuan peppercorn sauce.
NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07 
Target entity: mapo tofu
Target entity description: Mapo tofu is a classic Sichuan Chinese dish featuring soft tofu and minced meat in a spicy, numbing chili and Sichuan peppercorn sauce.
  • A. Mandu
    Mandu is a historic fortified city in central India renowned for its Afghan-era architecture, romantic legends, and scenic hilltop setting.
  • B. Okonomiyaki
    Okonomiyaki is a savory Japanese pancake made from a batter of flour, cabbage, and various mix-ins, typically grilled and topped with a rich sauce, mayonnaise, and bonito flakes.
  • C. bibimbap
    Bibimbap is a traditional Korean rice dish topped with assorted vegetables, meat, egg, and spicy gochujang sauce, all mixed together before eating.
  • D. Paippalada
    Paippalada is one of the principal ancient recensions (śākhās) of the Atharvaveda, preserved mainly in the Paippalāda tradition of Vedic scholarship.
  • E. Taro
    Taro is a common Japanese male given name, often written with kanji meaning "eldest son" or similar traditional connotations.
  • F. None of above. chosen

Provenance (5 batches)

The batch behind each pipeline step, in order, with when it ran. Timestamps are batch-level — stages were processed in waves, so the object chain (NER → NED1 → NEDg → NED2) reads in order, but predicate / elicitation batches can sit in a different wave.

Step Stage Batch ID Status When
creating Elicitation batch_69a493ef399c8190b04b9146d2314f59 completed March 1, 2026, 7:30 p.m.
NER Named-entity recognition batch_69a4bc6e8c2081909fb3534413b7aacb completed March 1, 2026, 10:23 p.m.
NED1 Entity disambiguation (via context triple) batch_69ac5eb1f7d08190ba722dcbbc8a6799 completed March 7, 2026, 5:21 p.m.
NEDg Description generation batch_69ac5f5ac2f88190a1d27b6ff457ddd3 completed March 7, 2026, 5:24 p.m.
NED2 Entity disambiguation (via description) batch_69ac5fbd55d88190a21f0e288bd9ae0a completed March 7, 2026, 5:26 p.m.
Created at: March 1, 2026, 7:44 p.m.