Triple
T488511
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Dan Snyder |
E9932
|
entity |
| Predicate | allegation |
P9856
|
FINISHED |
| Object | toxic workplace culture at Washington NFL franchise |
—
|
LITERAL FINISHED |
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: toxic workplace culture at Washington NFL franchise | Statement: [Dan Snyder, allegation, toxic workplace culture at Washington NFL franchise]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: allegation Context triple: [Dan Snyder, allegation, toxic workplace culture at Washington NFL franchise]
-
A.
accusationType
Indicates the specific category or nature of an accusation made by one party against another.
-
B.
appeal
Indicates that one party formally requests a higher authority to review, reconsider, or overturn a prior decision, judgment, or outcome.
-
C.
accusedOf
chosen
Indicates that one entity has formally alleged or claimed that another entity committed a specific wrongdoing or offense.
-
D.
plea
Indicates that a defendant formally states their response (such as guilty, not guilty, or no contest) to criminal charges in a legal proceeding.
-
E.
arguesThat
Indicates that one entity presents reasons or justification in support of a specific claim, position, or proposition held about another entity or topic.
- F. None of above.
Provenance (3 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_69a2e802e2908190ab17c9479e0b6412 |
completed | Feb. 28, 2026, 1:05 p.m. |
| NER | Named-entity recognition | batch_69a2f0df764481909811d9483dfbc4aa |
completed | Feb. 28, 2026, 1:42 p.m. |
| PD | Predicate disambiguation | batch_69a2edf63fbc819090ea6ca11f39116a |
completed | Feb. 28, 2026, 1:30 p.m. |
Created at: Feb. 28, 2026, 1:12 p.m.