Triple

T18528264
Position Surface form Disambiguated ID Type / Status
Subject Facebook Engineering E452772 entity
Predicate developedTechnology P5334 FINISHED
Object RocksDB NE NERFINISHED

How this triple was built (2 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: RocksDB | Statement: [Facebook Engineering, developedTechnology, RocksDB]
NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07
Target entity: RocksDB
Context triple: [Facebook Engineering, developedTechnology, RocksDB]
  • A. RocksDB chosen
    RocksDB is a high-performance, embeddable key–value store developed by Facebook, optimized for fast storage on flash and solid-state drives using a Log-Structured Merge-Tree (LSM) architecture.
  • B. LevelDB
    LevelDB is a fast, lightweight, open-source key–value storage library developed by Google, designed for high-performance reads and writes using a log-structured merge-tree architecture.
  • C. BSTDB
    BSTDB is a regional multilateral development bank that finances projects to promote economic development and cooperation among Black Sea region member countries.
  • D. Yama LSM
    Yama LSM is a Linux Security Module focused on strengthening process isolation and restricting ptrace and related debugging capabilities to harden the kernel against local attacks.
  • E. WiredTiger
    WiredTiger is a high-performance, concurrent storage engine designed for modern databases, known for its document-level locking, compression, and efficient memory management.
  • F. None of above.
  • G. Unsure - the case is ambiguous/there is not enough information to decide.

Provenance (2 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_69d8d387b5548190aa030dad2cb4947e completed April 10, 2026, 10:40 a.m.
NER Named-entity recognition batch_69e533fad2d081908395914d6a6b4eb1 completed April 19, 2026, 7:58 p.m.
Created at: April 10, 2026, 11:37 a.m.