mjbommar/opengloss-v2.1-examples

Superseded by [OpenGloss v2.2](https://huggingface.co/datasets/mjbommar/opengloss-v2.2-examples) (2026-09-08): 148,292 live lexemes and 288,304 senses — tier 5 closes the WordNet gap (38,100 entries imported from Princeton WordNet 3.0 and enriched), inflected-form headwords are folded onto their lemmas, and every inherited field carries a migrate provenance record. v2.1 stays published for reproducibility.

OpenGloss v2.1 — Examples

Every example sentence in OpenGloss v2.1, one row at a time, each tagged to the sense it illustrates and carrying the [span_start, span_end) character offsets of the headword occurrence inside it. That combination — a sentence, the sense it uses, and where the word is — is what a word-in-context or sense-disambiguation task needs and is normally paid for by annotation. source distinguishes the per-sense examples stage's verified sentences from reading-level and register rewrites of an existing example.

Part of the OpenGloss v2.1 release family — 16 datasets built from one store of 109,633 lexemes and 250,003 live senses, all joinable on derived ids. See Related datasets for the rest.

What's new in v2.1 vs v1.3

1. 1.Schema v3. Every lexeme carries a kind discriminator (simplex, compound, phrasal verb, idiom, proper noun, abbreviation, affix, function word); every sense carries a controlled domain leaf from a fixed ~160-leaf taxonomy instead of free text; every example carries the character span of the headword occurrence inside it.
2. 2.Renditions, not one string. A definition is a set: the canonical one plus rewrites at four reading levels and in four registers, each produced in a single call from the canonical text so they say the same thing at different altitudes.
3. 3.A sense graph, not a word graph. Typed relations resolve to sense ids wherever the target's entry exists in the release, so bank --hypernym--> financial institution points at a meaning rather than at a string.
4. 4.Retrieval data is first-class. Synthetic per-sense queries in eight styles, grounded QA pairs, mined word-in-context pairs, MS MARCO-style triples with graph-derived hard negatives, and graded TREC qrels — all derivable from, and consistent with, the same entries.
5. 5.Derivable identifiers everywhere. v1.3 published a positional id for lexemes and senses (3d_model_noun_0) and nothing below that. v2.1 gives every rendition, edge, query, QA pair and provenance record an id computable from the row alone, and never renumbers: a retired sense is tombstoned, so the ids after it keep their meaning.
6. 6.Per-field provenance. Which model wrote a field, how many tokens it took, what it cost — published as its own dataset.

What changed since v2.0

v2.0 (2026-09-05) covered the frequency-ranked single words. v2.1 adds tier 4: the function words the core ranking had excluded on purpose, and every remaining v1.3 entry at Wikipedia frequency ≥ 10 — mostly multiword compounds ("natural selection", "catalog number"), plus names and rarer single words. That doubles the lexeme count and changes the mix: v2.0 was 99.8% single words; a third of v2.1 is multiword.

Schema. No column was added, removed or retyped in any existing dataset. Three things did change:

• —tier gains the value tier4 (it was core, tier2 or tier3).
• —One new dataset, opengloss-v2.1-inflections: a flat surface-form → lemma lookup (plural, past tense, participles, comparative, superlative, derivations) built from the morphology that the lexicon already carried nested.
• —New provenance note prefixes on tombstones and edges, all reversible and all counted in the store audit: phantom_pos: (a v1.3 part-of-speech block whose glosses defined a component word rather than the compound — 11,440 blocks retired), regen: (relations regenerated for senses that had lost every edge to judging), and retyped: contrast (synonym edges the contrast paragraphs showed to be hypernym or hyponym).

Not row-compatible with v2.0. Lexeme, sense, rendition, edge, query and QA ids are stable for every entry v2.0 had. The derived training sets (retrieval-pairs, retrieval-triples, qrels) re-sample negatives over the larger pool, so their rows differ; and the store-wide quality passes run for v2.1 retired ~3,000 senses of the v2.0 entries (phantom part-of-speech blocks and near-duplicate senses), so those senses are now tombstoned rather than live. Treat v2.1 as a new release, not a delta.

Scope: fewer headwords, far more per headword

v2.1 is not a superset of v1.3. It covers 109,633 of v1.3's 205,988 lexemes — every frequency-ranked single word, plus the compounds and names at Wikipedia frequency ≥ 10 — and spends the difference on depth. If you need breadth of vocabulary, use v1.3; if you need graded renditions, resolved relations, spans, or retrieval supervision, use v2.1.

Key statistics

By tier

• —core — top 10K by composite frequency
• —tier2 — ranks to ~42K
• —tier3 — the rest of the frequency-ranked single words
• —tier4 — stopwords, plus compounds and names at Wikipedia frequency ≥ 10

Coverage by tier

The release was built in 4 frequency-ranked passes (core, tier2, tier3 and tier4) and they did not all receive the same stages. This table is per-field and per-tier so the gaps are visible rather than averaged away.

Files

Fields

2,163,329 rows, one row per example rendition.

One real row:

{
  "sense_id": "a:determiner:0",
  "lexeme_id": "a",
  "headword": "a",
  "pos": "determiner",
  "sense_index": 0,
  "domain": "language.grammar",
  "tier": "tier4",
  "reading_level": "neutral",
  "register": "plain",
  "text": "I saw a bird in the yard.",
  "span_start": 6,
  "span_end": 7,
  "readability_grade": null,
  "source": "renditions"
}

Loading it

from datasets import load_dataset

ds = load_dataset("mjbommar/opengloss-v2.1-examples", split="train")
print(ds)
print(ds[0])

The shards are plain parquet, so nothing forces you through datasets — read them straight, locally or over hf://:

import polars as pl

df = pl.read_parquet("hf://datasets/mjbommar/opengloss-v2.1-examples/data/train-*.parquet")
print(df.head())

import duckdb

duckdb.sql("SELECT count(*) FROM 'hf://datasets/mjbommar/opengloss-v2.1-examples/data/train-*.parquet'").show()

Word-in-context items: the sentence, the sense, the span

from datasets import load_dataset

ex = load_dataset("mjbommar/opengloss-v2.1-examples", split="train")
row = ex[0]
text, start, end = row["text"], row["span_start"], row["span_end"]
print(text[:start] + "[" + text[start:end] + "]" + text[end:])
print("sense:", row["sense_id"], "|", row["reading_level"], "/", row["register"])

Identifiers, and how they compose

Every id is derived from structure, never randomly minted, so a consumer can recompute one from a row and join across the whole family without a lookup table. Sense positions are stable across regenerations: a retired sense is tombstoned, not removed, so the indices after it never shift.

An edge id keys on the target's slug, not on the target's sense, so resolving a target never changes the id of the edge that found it.

Reading levels and registers

A rendition is keyed on a (reading_level, register) pair. The canonical rendition of every field is (neutral, plain); everything else is a rewrite of it.

marketing sits on the register axis for convenience but is a genre value rather than a point on the formality scale — worth remembering if you train a formality classifier on this column.

Related datasets

Everything below is built from the same store and joins on lexeme_id / sense_id.

Known limitations

• —It is synthetic. Every string here was written by a language model against a schema, not transcribed from a corpus or checked by a lexicographer. It is well-formed and internally consistent; it is not attested usage, and it will contain confident errors. Do not use it as ground truth about what a word means.
• —Judge scores 70.2/100 (core + tier 2) and 66.7/100 (tier 3). A different model family (Claude Opus) scored fixed 40-entry stratified samples at the close of each build. Sample statistics, not per-entry guarantees, and the judge is itself a model.
• —Relation precision is the weakest axis. Relations were judged for validity and the ones that failed were demoted rather than asserted; symmetric reciprocity finished at 94.3% for synonyms and 95.3% for antonyms, and 1,890 senses were left with no relation at all. Treat a single edge as a hypothesis, not a fact; treat the aggregate graph as usable.
• —`core`, `tier2`, `tier3` and `tier4` are deliberately partial. 109,633 lexemes across core, tier2, tier3 and tier4 received the text stages (glosses, examples, encyclopedia) but not the queries, QA pairs, contrasts or register renditions. The coverage table above gives the exact per-field share; nothing is hidden behind an average.
• —The encyclopedia is entry-level. One article per headword, about the headword as a whole. On a polysemous entry it is not a description of any one sense, and it is never used as a positive for one (D-71). It is entry-level reference prose, not a specialist article.

Citation

@misc{bommarito2025opengloss,
  title  = {OpenGloss: A Synthetic Encyclopedic Dictionary and Semantic Knowledge Graph},
  author = {Bommarito, Michael J., II},
  year   = {2025},
  eprint = {2511.18622},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/2511.18622}
}

License

Released under Creative Commons Attribution 4.0 International (CC-BY 4.0). Attribution to the OpenGloss project is required; commercial use is permitted.