Este modelo é uma versão fine-tuned do Qwen/Qwen3-Embedding-4B, especializado no domínio jurídico brasileiro. Ele foi desenvolvido como parte de uma pesquisa de doutorado focada no Enriquecimento de Embeddings Neurais para a Linguagem Jurídica Brasileira.

O BumbaLM-Embedding mapeia sentenças e parágrafos (como peças jurídicas, jurisprudência e consultas processuais) para um espaço vetorial denso de 2560 dimensões, sendo ideal para tarefas de:

Recuperação de Informação Legal (Legal IR)
Busca Semântica em processos jurídicos(TJMA e outros tribunais)
Aplicações de RAG (Retrieval-Augmented Generation) para Direito

Model Description

Model Type: Sentence Transformer
Base model: Qwen3-Embedding-4B -->
Maximum Sequence Length: 1024 tokens
Output Dimensionality: 2560 dimensions
Similarity Function: Cosine Similarity
Training Dataset: Legal Documents
Language: Portuguese
License: Unknown

Métricas de Avaliação

O modelo foi avaliado em um conjunto de teste reservado ("bumba_test_eval") composto por pares de consultas e trechos jurídicos reais.

Métrica	Valor	Descrição
cosine_ndcg@10	0.3687	Normalized Discounted Cumulative Gain (Métrica principal de ranking)
cosine_mrr@10	0.3363	Mean Reciprocal Rank
cosine_map@100	0.3426	Mean Average Precision
cosine_accuracy@10	0.4700	Acurácia no top-10 resultados

Nota: O modelo base foi refinado utilizando a função de perda TripletLoss, focando na distinção entre passagens juridicamente relevantes e "negativos difíceis" (trechos com termos similares mas semanticamente incorretos para a consulta).

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

bash

SentenceTransformer(
  (0): Transformer({'max_seq_length': 1024, 'do_lower_case': False, 'architecture': 'Qwen3Model'})
  (1): Pooling({'word_embedding_dimension': 2560, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': True, 'include_prompt': True})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("fabricioalmeida/BumbaLM-Embedding-4B-v0.1")
# Run inference
frases = [
    "O réu apresentou habeas corpus preventivo.",
    "A jurisprudência do STJ é pacífica nesse sentido.",
    "Receita de bolo de cenoura com chocolate."
]

embeddings = model.encode(frases)
print(embeddings.shape)
# [3, 2560]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.2075, 0.2365],
#        [0.2075, 1.0000, 0.1745],
#        [0.2365, 0.1745, 1.0000]])

Evaluation

Metrics

Information Retrieval

Dataset: bumba_test_eval
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.268
cosine_accuracy@3	0.388
cosine_accuracy@5	0.424
cosine_accuracy@10	0.47
cosine_precision@1	0.268
cosine_precision@3	0.1293
cosine_precision@5	0.0848
cosine_precision@10	0.047
cosine_recall@1	0.268
cosine_recall@3	0.388
cosine_recall@5	0.424
cosine_recall@10	0.47
cosine_ndcg@10	0.3687
cosine_mrr@10	0.3363
cosine_map@100	0.3426

Training Dataset

Legal Dataset

Size: 19,500 training samples
Columns: sentence_0, sentence_1, and sentence_2
Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1 sentence_2
type string string string
details
min: 31 tokens
mean: 47.57 tokens
max: 92 tokens
min: 6 tokens
mean: 242.04 tokens
max: 463 tokens
min: 6 tokens
mean: 224.16 tokens
max: 457 tokens
Loss: TripletLoss with these parameters:

json
1{ 2 "distance_metric": "TripletDistanceMetric.EUCLIDEAN", 3 "triplet_margin": 5 4}

Training Details

Training Hyperparameters

eval_strategy: steps
num_train_epochs: 1
multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 8
per_device_eval_batch_size: 8
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 5e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1
num_train_epochs: 1
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.0
warmup_steps: 0
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: True
save_on_each_node: False
save_only_model: False
restore_callback_states_from_checkpoint: False
no_cuda: False
use_cpu: False
use_mps_device: False
seed: 42
data_seed: None
jit_mode_eval: False
bf16: False
fp16: False
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: False
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
parallelism_config: None
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch_fused
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
project: huggingface
trackio_space_id: trackio
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: None
hub_always_push: False
hub_revision: None
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
include_for_metrics: []
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
include_tokens_per_second: False
include_num_input_tokens_seen: no
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
liger_kernel_config: None
eval_use_gather_object: False
average_tokens_across_devices: True
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: round_robin
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	bumba_test_eval_cosine_ndcg@10
0.0820	200	0.3687

Framework Versions

Python: 3.12.12
Sentence Transformers: 5.1.2
Transformers: 4.57.2
PyTorch: 2.9.0+cu126
Accelerate: 1.12.0
Datasets: 4.4.1
Tokenizers: 0.22.1

Citation

BibTeX

latex

@misc{bumbalm2025,
    title={BumbaLM-Embeddings: Enriquecimento de Embeddings Neurais para a Linguagem Jurídica Brasileira},
    author={Almeida, Fabrício},
    year={2025},
    description={Modelo de embedding fine-tuned para o domínio jurídico brasileiro (TJMA).}
}

	sentence_0	sentence_1	sentence_2
type	string	string	string
details	min: 31 tokens mean: 47.57 tokens max: 92 tokens	min: 6 tokens mean: 242.04 tokens max: 463 tokens	min: 6 tokens mean: 224.16 tokens max: 457 tokens