ToMMeR-bert-base-uncased_L5_R64

ToMMeR is a lightweight probing model extracting emergent mention detection capabilities from early layers representations of any LLM backbone, achieving high Zero Shot recall across a wide set of 13 NER benchmarks.

Checkpoint Details

Property	Value
Base LLM	bert-base-uncased
Layer	5
#Params	99.1K

Usage

Installation

Our code can be installed with pip+git, Please visit the repository for more details.

pip install git+https://github.com/VictorMorand/llm2ner.git

Fancy Outputs

import llm2ner
from llm2ner import ToMMeR

tommer = ToMMeR.from_pretrained("llm2ner/ToMMeR-bert-base-uncased_L5_R64")
# load Backbone llm, optionnally cut the unused layer to save GPU space.
llm = llm2ner.utils.load_llm( tommer.llm_name, cut_to_layer=tommer.layer,) 
tommer.to(llm.device)

text = "Large language models are awesome. While trained on language modeling, they exhibit emergent Zero Shot abilities that make them suitable for a wide range of tasks, including Named Entity Recognition (NER). "

#fancy interactive output
outputs = llm2ner.plotting.demo_inference( text, tommer, llm,
    decoding_strategy="threshold",  # or "greedy" for flat segmentation
    threshold=0.5, # default 50%
    show_attn=True,
)

Large PRED language PRED models are awesome . While trained on language PRED modeling , they exhibit emergent PRED abilities that make them suitable for a wide range of tasks PRED , including Named PRED Entity Recognition ( NER PRED ) .

Raw inference

By default, ToMMeR outputs span probabilities, but we also propose built-in options for decoding entities.

• Inputs:
  • tokens (batch, seq): tokens to process,
  • model: LLM to extract representation from.
• Outputs: (batch, seq, seq) matrix (masked outside valid spans)

tommer = ToMMeR.from_pretrained("llm2ner/ToMMeR-bert-base-uncased_L5_R64")
# load Backbone llm, optionnally cut the unused layer to save GPU space.
llm = llm2ner.utils.load_llm( tommer.llm_name, cut_to_layer=tommer.layer,) 
tommer.to(llm.device)

#### Raw Inference
text = ["Large language models are awesome"]
print(f"Input text: {text[0]}")

#tokenize in shape (1, seq_len)
tokens = model.tokenizer(text, return_tensors="pt")["input_ids"].to(device)
# Output raw scores
output = tommer.forward(tokens, model) # (batch_size, seq_len, seq_len)
print(f"Raw Output shape: {output.shape}")

#use given decoding strategy to infer entities
entities = tommer.infer_entities(tokens=tokens, model=model, threshold=0.5, decoding_strategy="greedy")
str_entities = [ model.tokenizer.decode(tokens[0,b:e+1]) for b, e in entities[0]]
print(f"Predicted entities: {str_entities}")

>>> Input text: Large language models are awesome
>>> Raw Output shape: torch.Size([1, 6, 6])
>>> Predicted entities: ['Large language models']

Please visit the repository for more details and a demo notebook.

Evaluation Results

dataset	precision	recall	f1	n_samples
MultiNERD	0.2291	0.9736	0.3709	154144
CoNLL 2003	0.35	0.9216	0.5074	16493
CrossNER_politics	0.3373	0.9538	0.4984	1389
CrossNER_AI	0.3378	0.9515	0.4985	879
CrossNER_literature	0.3465	0.9215	0.5037	916
CrossNER_science	0.3571	0.9392	0.5175	1193
CrossNER_music	0.3738	0.9217	0.5319	945
ncbi	0.1372	0.8794	0.2374	3952
FabNER	0.296	0.6694	0.4105	13681
WikiNeural	0.2168	0.9677	0.3542	92672
GENIA_NER	0.2517	0.9369	0.3968	16563
ACE 2005	0.2976	0.4088	0.3445	8230
Ontonotes	0.2576	0.7137	0.3786	42193
Aggregated	0.241	0.9103	0.3811	353250
Mean	0.2914	0.8584	0.4269	353250

Citation

If using this model or the approach, please cite the associated paper:

@misc{morand2025tommerefficiententity,
      title={ToMMeR -- Efficient Entity Mention Detection from Large Language Models}, 
      author={Victor Morand and Nadi Tomeh and Josiane Mothe and Benjamin Piwowarski},
      year={2025},
      eprint={2510.19410},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2510.19410}, 
}

License

Apache-2.0 (see repository for full text).