Berk/span-marker-xlm-roberta-base-fewnerd-onnx

SpanMarker XLM-RoBERTa Base FewNERD (ONNX)

ONNX export of tomaarsen/span-marker-xlm-roberta-base-fewnerd-fine-super for browser-based inference.

Model Details

• Base model: XLM-RoBERTa Base (multilingual, 10+ languages)
• Training data: FewNERD fine-grained supervised
• Entity types: 66 types + O (including building-hotel, building-airport, location-GPE, location-park, location-bodiesofwater, location-mountain, etc.)
• Quantization: INT8 dynamic quantization on encoder

Files

File	Size	Description
encoder.onnx	~266 MB	XLM-RoBERTa encoder (INT8 quantized)
classifier.onnx	0.4 MB	Linear classifier (1536 -> 67 classes)
labels.json	2 KB	ID to label mapping
tokenizer.json	17 MB	SentencePiece tokenizer

Inference

SpanMarker uses a custom architecture:

1. Tokenize input and enumerate all spans (up to entity_max_length=8 tokens)
2. Insert start/end marker tokens into the sequence
3. Run encoder to get hidden states
4. Extract start marker embeddings from [start_idx : start_idx + n_pairs]
5. Extract end marker embeddings from [start_idx + n_pairs : start_idx + n_pairs*2]
6. Concatenate start+end -> (n_pairs, 1536) features
7. Pad to (256, 1536) and run classifier -> (256, 67) logits
8. Softmax + filter non-O predictions above threshold

Limitations

• Fixed sequence length of 512 (batch=1)
• Longer texts need to be chunked (the original model does this internally with batching)
• INT8 quantization slightly reduces scores (~0.05-0.1 lower) but maintains correct predictions

License

CC-BY-SA-4.0 (same as original model)