gte-base-case-law-v2

SentenceTransformer based on Alibaba-NLP/gte-base-en-v1.5

This is a sentence-transformers model finetuned from Alibaba-NLP/gte-base-en-v1.5 on the json dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: Alibaba-NLP/gte-base-en-v1.5 <!-- at revision a8e4f3e0ee719c75bc30d12b8eae0f8440502718 -->
• Maximum Sequence Length: 8192 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • json
• Language: en
• License: apache-2.0

Model Sources

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

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 8192, 'do_lower_case': False}) with Transformer model: NewModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, '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': False, '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.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("Tejasw1/gte-base-case-law-v2")
# Run inference
sentences = [
    "What legal considerations are involved in determining a person's citizenship status when foreign nationality is claimed in criminal proceedings?",
    "**1. Key Legal Issues and Holdings:**\n\n* **Determination of Citizenship:** The main legal issue is the determination of the respondent's citizenship status, specifically whether he had acquired Pakistani nationality and lost Indian citizenship.\n* **Section 9 of the Citizenship Act:** The court considered the application of Section 9 of the Citizenship Act, 1955, which requires the Central Government to determine whether a person has acquired foreign citizenship.\n* **Jurisdiction of Criminal Courts:** The court held that the jurisdiction of criminal courts to record a conviction or acquittal is limited when the question of acquisition of foreign citizenship is involved.\n* **Burden of Proof under Section 9 of the Foreigners' Act:** The court noted that the burden of proof under Section 9 of the Foreigners' Act, 1946, is immaterial when an issue triable exclusively by the Central Government has properly arisen.\n\n**2. Significant Facts of the Case:**\n\n* The respondent was charged with overstaying in India after his permit expired, allegedly due to his foreigner status.\n* The respondent claimed to be an Indian citizen, born in Gujarat, and had gone to Pakistan as a minor after a quarrel with his father.\n* The respondent produced credible evidence to prove his Indian citizenship, including his birth certificate and school records.\n* The prosecution relied on the respondent's Pakistani passport and visa application, which stated that he had migrated to Pakistan in 1950.\n\n**3. Court's Ruling:**\n\n* The High Court had acquitted the respondent, finding that the prosecution had failed to prove that he had acquired Pakistani nationality.\n* The Supreme Court upheld the acquittal, holding that the prosecution had proceeded with the case despite knowing that the question of acquisition of foreign citizenship was involved.\n* The court quashed the charge against the respondent, allowing him to be discharged.\n\n**4. Citations:**\n\n* **Kulathil Mammu v. State of Kerala**, (1966) 3 SCR 706 : AIR 1966 SC 1614 : 1966 Cri LJ 1217\n* **Shanno Devi v. Mangal Sain**, (1961) 1 SCR 576 : AIR 1961 SC 58 : (1961) 1 SCJ 201\n* **State of A.P. v. Abdul Khader**, (1962) 1 SCR 737 : AIR 1961 SC 1467 : 1961 (2) Cri LJ 573\n* **Abdul Sattar v. State of Gujarat**, AIR 1965 SC 810 : (1964) 2 SCJ 461 : (1965) 1 Cri LJ 759\n* **Akbar Khan v. Union of India**, (1962) 1 SCR 779 : AIR 1962 SC 70 : (1962) 1 SCJ 665",
    "**1. Key Legal Issues and Holdings:**\n\n* **Acquiescence in Arbitrator's Appointment:** The Supreme Court held that if parties to a reference agree beforehand to the method of appointment or acquiesce in the appointment made with full knowledge of the circumstances, they cannot object to such appointment as invalidating subsequent proceedings.\n* **Succession of Arbitrator:** The court ruled that the successor arbitrator was competent to complete the arbitration proceedings, despite the predecessor arbitrator's entry into reference.\n* **Principles of Natural Justice:** The court rejected the contention that there was a violation of the principles of natural justice, as the parties had knowledge of the change of the incumbent and had acquiesced in the proceedings.\n* **Award's Validity:** The court held that the award was unassailable, as it did not contain any patent mistake of law or gross misstatement of facts resulting in miscarriage of justice or equity.\n\n**2. Significant Facts of the Case:**\n\n* The petitioner company undertook the work of widening and strengthening pavements in National Highways No. 7, Madurai-Kanyakumari Road.\n* The work was divided into fourteen Reaches, and fourteen separate agreements were entered into between the petitioner and the Superintending Engineer, National Highways, Tirunelveli.\n* The predecessor arbitrator, Thiru Mohan, entered into reference but was transferred before completing the award.\n* The successor arbitrator, Thiru J.R. Cornelius, completed the arbitration proceedings with the knowledge and consent of the petitioner.\n\n**3. Court's Ruling:**\n\n* The Supreme Court dismissed the special leave petitions and upheld the High Court's judgment, which had upheld the District Judge's dismissal of the challenge to the award.\n* The court held that the award was unassailable and that the parties had acquiesced in the appointment of the successor arbitrator.\n* The court rejected the contentions that the successor arbitrator had no jurisdiction to complete the arbitration proceedings and that there was a violation of the principles of natural justice.\n\n**4. Citations:**\n\n* **Chowdhury Murtaza Hossein v. Mussumat Bibi Bechunnissa**, 3 IA 209 : 3 Sar 663 : 3 Suther 342\n* **Jupiter General Insurance Co. Ltd. v. Corporation of Calcutta**, AIR 1956 Cal 470 : 1955 Cri LJ 1243\n* **N. Challappan v. Secretary, Kerala State Electricity Board**, (1975) 1 SCC 289\n* **Prasun Roy v. Calcutta Metropolitan Development Authority**, (1987) 4 SCC 217\n* **Firm Madanlal Roshanlal Mahajan v. Hukumchand Mills Ltd.**, (1967) 1 SCR 105 : AIR 1967 SC 1030",
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

<!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* -->

Evaluation

Metrics

Information Retrieval

• Datasets: dim_768 and dim_512
• Evaluated with <code>InformationRetrievalEvaluator</code>

<!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* -->

Training Details

Training Dataset

json

• Dataset: json
• Size: 16,465 training samples
• Columns: <code>anchor</code> and <code>positive</code>
• Approximate statistics based on the first 1000 samples:

  	anchor	positive
  type	string	string
  details	<ul><li>min: 2 tokens</li><li>mean: 26.38 tokens</li><li>max: 72 tokens</li></ul>	<ul><li>min: 333 tokens</li><li>mean: 563.51 tokens</li><li>max: 1134 tokens</li></ul>

• Samples:

  anchor	positive
  <code>What factors do courts consider when evaluating the reliability of eyewitness testimonies in murder trials?</code>	<code>1. Key Legal Issues and Holdings:<br><br>* Culpable Homicide: The court considered the application of Section 302 of the Indian Penal Code, 1860, which deals with punishment for culpable homicide not amounting to murder.<br>* Section 302 IPC: The court upheld the conviction of the accused under Section 302 IPC for the death of Ishwardeen.<br>* Section 34 IPC: The court also upheld the conviction of the accused under Section 34 IPC for the death of Ishwardeen, considering the common intention of the accused.<br><br>2. Significant Facts of the Case:<br><br>* The deceased, Ishwardeen, was killed in a alleged incident involving the accused, Bhagwan Das, Sheo Mohan, and Shanker @ Sheo Shanker.<br>* The incident occurred on August 18, 1983, at the house of Bhagwan Das, where Ishwardeen and his family were residing.<br>* The accused had allegedly demanded rent from Ishwardeen and had threatened to evict him from the house.<br>* Ishwardeen was killed by the accused, who allegedly gave him 8-9 knife blows.<br>...</code>
  <code>What principles guide the court's decisions on wage fixation in cases involving government undertakings?</code>	<code>1. Key Legal Issues and Holdings:<br><br>* Wage Structure: The main legal issue is whether the wage structure of a government undertaking in the public sector should be different from that of an undertaking in the private sector.<br>* Section 10(1)(d) of the Industrial Disputes Act, 1947: The court considered the applicability of this provision in the context of wage fixation.<br>* Article 39 and 43 of the Directive Principles of State Policy: The court examined the constitutional implications of making a distinction between laborers in the public and private sectors.<br>* Region-cum-Industry Principle: The court upheld the principle of region-cum-industry in wage fixation.<br>* Gratuity Scheme: The court considered the validity of the gratuity scheme introduced by the Tribunal, which provided for payment of gratuity to workmen upon retirement or termination of service.<br>* Fair Wages: The court examined the issue of fair wages and whether the Tribunal had violated the accepted...</code>
  <code>- What role does the recommendation of a State Government play in the land exemption process under the Punjab Towns Improvement Act, 1922?</code>	<code>1. Key Legal Issues and Holdings:<br><br>* Acquisition of Land: The main legal issue is the acquisition of land by the Improvement Trust under the Punjab Towns Improvement Act, 1922.<br>* Exemption of Orchards: The court considered whether the Improvement Trust was bound to exempt orchards from acquisition under Section 56 of the Act.<br>* Article 14 of the Constitution: The court held that the Improvement Trust did not violate Article 14 of the Constitution by exempting some orchards while acquiring others.<br>* Quasi-Judicial Capacity: The court held that the Improvement Trust acts in a quasi-judicial capacity when determining claims under Section 56 of the Act, but is not bound to give reasons for its decisions.<br>* Locus Standi: The court observed that the appellants had no locus standi to invoke Section 56 of the Act, as the acquisition of their land had not been discovered to be unnecessary for the execution of the scheme.<br>* Power to Exempt Lands: The court held that t...</code>

• Loss: <code>MatryoshkaLoss</code> with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512
      ],
      "matryoshka_weights": [
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Evaluation Dataset

json

• Dataset: json
• Size: 867 evaluation samples
• Columns: <code>anchor</code> and <code>positive</code>
• Approximate statistics based on the first 867 samples:

  	anchor	positive
  type	string	string
  details	<ul><li>min: 2 tokens</li><li>mean: 26.91 tokens</li><li>max: 94 tokens</li></ul>	<ul><li>min: 318 tokens</li><li>mean: 568.71 tokens</li><li>max: 1120 tokens</li></ul>

• Samples:

  anchor	positive
  <code></code>	<code>1. Key Legal Issues and Holdings:<br><br>* Deemed Tenancy: The main legal issue is the interpretation of Section 6 of the Bombay Tenancy and Agricultural Lands (Vidarbha Region) Act, 1958, which deals with deemed tenancy.<br>* Right to Cultivate Land: The case revolves around the right of the appellant to cultivate the land and appropriate the produce, despite being a servant of the respondent.<br>* Employer-Employee Relationship: The court considered the nature of the relationship between the appellant and the respondent, whether it was that of an employer-employee or a landlord-tenant.<br><br>2. Significant Facts of the Case:<br><br>* The respondent owns three agricultural lands in the Vidarbha region of Maharashtra.<br>* The appellant's father, Umaji, was appointed a pujari by the landlord to worship the idols in the temple and look after the management of the dharamshalas.<br>* The appellant was allowed to cultivate the lands and take the crops in lieu of payment for his services as a pujari...</code>
  <code>What is the significance of the burden of proof in tax cases involving the failure to file returns on time?</code>	<code>1. Key Legal Issues and Holdings:<br><br>* Burden of Proof: The main legal issue is the burden of proof required to establish the absence of "reasonable cause" for the assessee's failure to file returns within the specified time.<br>* Penalty under Section 271(1)(a): The court considered the nature of penalty under Section 271(1)(a) of the Income-tax Act, 1961, and whether it requires the element of "mens rea" (guilty mind).<br>* Civil Obligation vs. Criminal Liability: The court distinguished between civil obligation and criminal liability, holding that penalty under Section 271(1)(a) is civil in nature and does not require proof of "mens rea".<br><br>2. Significant Facts of the Case:<br><br>* The assessee, I.M. Patel and Co., failed to file returns for the assessment years 1964-65, 1965-66, and 1966-67, leading to imposition of penalty under Section 271(1)(a).<br>* The assessee argued that the burden was on the Revenue to prove absence of "reasonable cause" for the delay.<br>* The court held t...</code>
  <code>In what circumstances can an industrial tribunal adjudicate on issues related to retrenchment compensation versus reinstatement of employees?</code>	<code>1. Key Legal Issues and Holdings:<br><br>* Transfer of Employment: The main legal issue is whether Respondent 3 was transferred from the appellant corporation to the subsidiary company Sindhu Hotchief, and whether he continued to be an employee of the appellant corporation.<br>* Retrenchment Compensation: The court considered the entitlement of Respondent 3 to retrenchment compensation from the appellant corporation.<br>* Jurisdiction of Industrial Tribunal: The court examined whether the State Government had the jurisdiction to refer a question of reinstatement as an industrial dispute for adjudication by the Tribunal.<br>* Competency of State Government Reference: The court held that the State Government was not competent to refer a question of reinstatement as an industrial dispute for adjudication by the Tribunal.<br><br>2. Significant Facts of the Case:<br><br>* Respondent 3 was employed by the appellant corporation as an Accounts Clerk and was later transferred to Sindhu Hotchief.<br>*...</code>

• Loss: <code>MatryoshkaLoss</code> with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512
      ],
      "matryoshka_weights": [
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: epoch
• per_device_train_batch_size: 16
• gradient_accumulation_steps: 8
• learning_rate: 2e-05
• num_train_epochs: 4
• lr_scheduler_type: cosine
• warmup_ratio: 0.1
• bf16: True
• tf32: True
• load_best_model_at_end: True
• optim: adamw_torch_fused
• batch_sampler: no_duplicates

All Hyperparameters

<details><summary>Click to expand</summary>

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: epoch
• prediction_loss_only: True
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 8
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 8
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 2e-05
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 4
• max_steps: -1
• lr_scheduler_type: cosine
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.1
• 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
• use_ipex: False
• bf16: True
• fp16: False
• fp16_opt_level: O1
• half_precision_backend: auto
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: True
• 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: True
• 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}
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch_fused
• optim_args: None
• adafactor: False
• group_by_length: False
• length_column_name: length
• 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: False
• hub_always_push: False
• 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
• dispatch_batches: None
• split_batches: None
• include_tokens_per_second: False
• include_num_input_tokens_seen: False
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• eval_use_gather_object: False
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

</details>

Training Logs

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.11.5
• Sentence Transformers: 3.3.0.dev0
• Transformers: 4.46.1
• PyTorch: 2.5.1+cu124
• Accelerate: 0.34.2
• Datasets: 3.0.0
• Tokenizers: 0.20.1

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning},
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

<!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->