README.md

metadata

language: en
license: apache-2.0
tags:
  - pytorch
  - token-classification
  - ner
  - dei
  - equibert
metrics:
  - f1

EquiBERT — DEI Named Entity Recognition

Model ID: SallySims/equibert-ner

Token-level BIO tagger for 13 DEI-specific entity types in organisational text.

Entity Types

Entity	Description	Example
DEMOGRAPHIC_GROUP	Identity group reference	"BIPOC employees", "women"
PROTECTED_CHAR	Protected characteristic	"gender", "disability"
DEI_CONCEPT	DEI domain concept	"unconscious bias", "intersectionality"
ORG_ROLE	Organisational role	"CHRO", "hiring manager"
POLICY_ARTIFACT	Policy or document	"pay equity review", "DEI report"
METRIC	Measurable DEI outcome	"9% pay gap", "27% diverse hiring"
BIAS_INDICATOR	Bias signal language	"rock star", "cultural fit"
COMMITMENT	DEI commitment statement	"we will close this by Q2"
BARRIER	Identified barrier	"no lift at venue"
ACTION	Concrete DEI action	"blind CV screening"
ORGANISATION	Named organisation	company names
TIMEFRAME	Time reference	"Q2 2024", "by year end"
O	Outside — no entity	—

Usage (same as bert-base-NER)

from transformers import pipeline
ner = pipeline("ner", model="SallySims/equibert-ner")
ner("Our CHRO published the annual DEI report with measurable targets.")

Model Description

EquiBERT is a multi-task DEI (Diversity, Equity and Inclusion) transformer built on a dual-encoder backbone that fuses RoBERTa-base and DeBERTa-v3-base via a learned weighted sum (α parameter). The fused representation is fed into task-specific heads covering 17 distinct DEI analysis tasks.

Organisation: SallySims Framework: PyTorch + HuggingFace Transformers Backbone: RoBERTa-base + DeBERTa-v3-base (dual encoder, fused) Language: English Domain: Organisational DEI text — HR communications, policies, job descriptions, performance reviews, leadership statements, reports

Architecture

Input Text
    │
    ├──▶ RoBERTa-base encoder ──▶ Linear projection
    │                                     │
    └──▶ DeBERTa-v3-base encoder ──▶ Linear projection
                                          │
                              Weighted fusion (learned α)
                                          │
                                   Layer Norm + Dropout
                                          │
                              Task-specific head (see below)

Training Data

Trained on synthetic DEI organisational text generated by the EquiBERT synthetic data pipeline, covering 20 DEI categories across HR, policy, leadership, and workforce analytics domains. For production use, fine-tune on real labelled DEI data.

Limitations

• Trained on synthetic data — predictions should be validated before use in real HR or policy decisions.
• English-only.
• Not a substitute for qualified DEI practitioners or legal advice.
• May reflect biases present in the training corpus.

Citation

If you use EquiBERT in your research, please cite:

@misc{equibert2024,
  author    = {SallySims},
  title     = {EquiBERT: A Multi-Task DEI Transformer},
  year      = {2024},
  publisher = {HuggingFace},
  url       = {https://huggingface.co/SallySims}
}

All EquiBERT Models

Model	Task	Primary Metric
equibert-bias-classifier	Bias Detection	Macro F1
equibert-microaggression	Microaggression Detection	Macro F1
equibert-category-tagger	DEI Category Tagging	Macro F1
equibert-event-exclusion	Event Exclusion Classification	Macro F1
equibert-inclusive-language	Inclusive Language Scoring	Span F1
equibert-review-auditor	Performance Review Auditing	Span F1
equibert-washing-detector	DEI Washing Detection	MAE
equibert-framing-scorer	Report Framing Scoring	MAE
equibert-awareness-scorer	DEI Awareness Scoring	MAE
equibert-similarity	Semantic Similarity	Accuracy
equibert-ner	DEI Entity Recognition	Span F1
equibert-relation-extraction	Relation Extraction	Macro F1
equibert-qa	Extractive QA	Span EM
equibert-search	Semantic Search	MRR@10
equibert-nli	NLI / Textual Entailment	Macro F1
equibert-generator	DEI Text Generation	ROUGE-L