metadata

base_model: BAAI/bge-base-en-v1.5
datasets: []
language:
  - en
library_name: sentence-transformers
license: apache-2.0
metrics:
  - cosine_accuracy@1
  - cosine_accuracy@3
  - cosine_accuracy@5
  - cosine_accuracy@10
  - cosine_precision@1
  - cosine_precision@3
  - cosine_precision@5
  - cosine_precision@10
  - cosine_recall@1
  - cosine_recall@3
  - cosine_recall@5
  - cosine_recall@10
  - cosine_ndcg@10
  - cosine_mrr@10
  - cosine_map@100
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:90
  - loss:MatryoshkaLoss
  - loss:MultipleNegativesRankingLoss
widget:
  - source_sentence: >-
      Ownership of NVIDIA Securities Information regarding ownership of NVIDIA
      securities required by this item will be contained in our 2023 Proxy
      Statement under the caption “Security Ownership of Certain Beneficial
      Owners and Management,” and is hereby incorporated by reference.
    sentences:
      - What are the two operating segments of NVIDIA as mentioned in the text?
      - >-
        What major factors contributed to the decrease in cash provided by
        operating activities in fiscal year 2023?
      - >-
        Where can information regarding the ownership of NVIDIA securities be
        found?
  - source_sentence: >-
      Development and Retention To support employee development, we provide
      opportunities to learn on-the-job through training programs, one on one
      coaching and ongoing feedback. We have a library of live and on-demand
      learning experiences that include workshops, panel discussions, and
      speaker forums. We curate learning paths focused on our most common
      development needs and constantly upgrade our offerings to ensure that our
      employees are exposed to the most current programs and technologies
      available.
    sentences:
      - >-
        How much is authorized for the repurchase of additional shares of common
        stock as of January 29, 2023?
      - What position did Timothy S. Teter acquire at NVIDIA in 2018?
      - >-
        What types of learning opportunities does the company provide to support
        employee development?
  - source_sentence: >-
      Data Center The NVIDIA computing platform is focused on accelerating the
      most compute-intensive workloads, such as AI, data analytics, graphics and
      scientific computing, across hyperscale, cloud, enterprise, public sector,
      and edge data centers. The platform consists of our energy efficient GPUs,
      data processing units, or DPUs, interconnects and systems, our CUDA
      programming model, and a growing body of software libraries, software
      development kits, or SDKs, application frameworks and services, which are
      either available as part of the platform or packaged and sold separately.
    sentences:
      - What position did Colette M. Kress hold before joining NVIDIA in 2013?
      - Where can NVIDIA's financial reports be accessed?
      - What are the key components of the NVIDIA computing platform?
  - source_sentence: >-
      Human Capital Management We believe that our employees are our greatest
      assets, and they play a key role in creating long-term value for our
      stakeholders. As of the end of fiscal year 2023, we had 26,196 employees
      in 35 countries, 19,532 were engaged in research and development and 6,664
      were engaged in sales, marketing, operations, and administrative
      positions.
    sentences:
      - What industries use NVIDIA's GPUs and software for automation?
      - >-
        How many employees did the company have at the end of fiscal year 2023,
        and in how many countries were they located?
      - How does NVIDIA's platform strategy contribute to the markets it serves?
  - source_sentence: >-
      Equity Compensation Plan Information Information regarding our equity
      compensation plans required by this item will be contained in our 2023
      Proxy Statement under the caption "Equity Compensation Plan Information,"
      and is hereby incorporated by reference.
    sentences:
      - >-
        What amount is recorded as unrecognized tax benefits at the end of
        fiscal year 2023?
      - >-
        What is the total amount authorized for the repurchase of common stock
        up to December 2023?
      - What document contains details about NVIDIA's equity compensation plans?
model-index:
  - name: BGE base Financial Matryoshka
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 768
          type: dim_768
        metrics:
          - type: cosine_accuracy@1
            value: 0.7
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.9
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 1
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.7
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.3
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.2
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.7
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.9
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 1
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.869253606521631
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.825
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.825
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 512
          type: dim_512
        metrics:
          - type: cosine_accuracy@1
            value: 0.7
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.9
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 1
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.7
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.3
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.2
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.7
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.9
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 1
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.869253606521631
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.825
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.825
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 256
          type: dim_256
        metrics:
          - type: cosine_accuracy@1
            value: 0.7
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.9
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 1
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.7
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.3
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.2
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.7
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.9
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 1
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.869253606521631
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.825
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.825
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 128
          type: dim_128
        metrics:
          - type: cosine_accuracy@1
            value: 0.7
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.9
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 1
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.7
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.3
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.2
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.7
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.9
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 1
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.869253606521631
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.825
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.825
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 64
          type: dim_64
        metrics:
          - type: cosine_accuracy@1
            value: 0.6
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.8
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.9
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.6
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.2666666666666667
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.18
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.1
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.6
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.8
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.9
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.7993566060880289
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.736111111111111
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.736111111111111
            name: Cosine Map@100

BGE base Financial Matryoshka

This is a sentence-transformers model finetuned from BAAI/bge-base-en-v1.5. 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: BAAI/bge-base-en-v1.5
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 tokens
Similarity Function: Cosine Similarity
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': 512, 'do_lower_case': True}) with Transformer model: BertModel 
  (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})
  (2): Normalize()
)

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("anikulkar/bge-base-financial-matryoshka-nvda-iter20")
# Run inference
sentences = [
    'Equity Compensation Plan Information Information regarding our equity compensation plans required by this item will be contained in our 2023 Proxy Statement under the caption "Equity Compensation Plan Information," and is hereby incorporated by reference.',
    "What document contains details about NVIDIA's equity compensation plans?",
    'What is the total amount authorized for the repurchase of common stock up to December 2023?',
]
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]

Evaluation

Metrics

Information Retrieval

Dataset: dim_768
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.7
cosine_accuracy@3	0.9
cosine_accuracy@5	1.0
cosine_accuracy@10	1.0
cosine_precision@1	0.7
cosine_precision@3	0.3
cosine_precision@5	0.2
cosine_precision@10	0.1
cosine_recall@1	0.7
cosine_recall@3	0.9
cosine_recall@5	1.0
cosine_recall@10	1.0
cosine_ndcg@10	0.8693
cosine_mrr@10	0.825
cosine_map@100	0.825

Information Retrieval

Dataset: dim_512
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.7
cosine_accuracy@3	0.9
cosine_accuracy@5	1.0
cosine_accuracy@10	1.0
cosine_precision@1	0.7
cosine_precision@3	0.3
cosine_precision@5	0.2
cosine_precision@10	0.1
cosine_recall@1	0.7
cosine_recall@3	0.9
cosine_recall@5	1.0
cosine_recall@10	1.0
cosine_ndcg@10	0.8693
cosine_mrr@10	0.825
cosine_map@100	0.825

Information Retrieval

Dataset: dim_256
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.7
cosine_accuracy@3	0.9
cosine_accuracy@5	1.0
cosine_accuracy@10	1.0
cosine_precision@1	0.7
cosine_precision@3	0.3
cosine_precision@5	0.2
cosine_precision@10	0.1
cosine_recall@1	0.7
cosine_recall@3	0.9
cosine_recall@5	1.0
cosine_recall@10	1.0
cosine_ndcg@10	0.8693
cosine_mrr@10	0.825
cosine_map@100	0.825

Information Retrieval

Dataset: dim_128
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.7
cosine_accuracy@3	0.9
cosine_accuracy@5	1.0
cosine_accuracy@10	1.0
cosine_precision@1	0.7
cosine_precision@3	0.3
cosine_precision@5	0.2
cosine_precision@10	0.1
cosine_recall@1	0.7
cosine_recall@3	0.9
cosine_recall@5	1.0
cosine_recall@10	1.0
cosine_ndcg@10	0.8693
cosine_mrr@10	0.825
cosine_map@100	0.825

Information Retrieval

Dataset: dim_64
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.6
cosine_accuracy@3	0.8
cosine_accuracy@5	0.9
cosine_accuracy@10	1.0
cosine_precision@1	0.6
cosine_precision@3	0.2667
cosine_precision@5	0.18
cosine_precision@10	0.1
cosine_recall@1	0.6
cosine_recall@3	0.8
cosine_recall@5	0.9
cosine_recall@10	1.0
cosine_ndcg@10	0.7994
cosine_mrr@10	0.7361
cosine_map@100	0.7361

Training Details

Training Dataset

Unnamed Dataset

Size: 90 training samples
Columns: positive and anchor
Approximate statistics based on the first 1000 samples:
positive anchor
type string string
details
min: 22 tokens
mean: 56.66 tokens
max: 142 tokens

min: 11 tokens
mean: 19.33 tokens
max: 32 tokens

	positive	anchor
type	string	string
details	min: 22 tokens mean: 56.66 tokens max: 142 tokens	min: 11 tokens mean: 19.33 tokens max: 32 tokens

Samples:

positive	anchor
`We also offer the NVIDIA GPU Cloud registry, or NGC, a comprehensive catalog of easy-to-use, optimized software stacks across a range of domains including scientific computing, deep learning, and machine learning. With NGC, AI developers, researchers and data scientists can get started with the development of AI and HPC applications and deploy them on DGX systems, NVIDIA-Certified systems from our partners, or with NVIDIA’s cloud partners.`	`What does the NVIDIA GPU Cloud registry offer?`
`To the extent realization of the deferred tax assets becomes more-likely-than-not, we would recognize such deferred tax assets as income tax benefits during the period.`	`What will be recognized as income tax benefits if the realization of deferred tax assets becomes more-likely-than-not?`
`Fueled by the sustained demand for exceptional 3D graphics and the scale of the gaming market, NVIDIA has leveraged its GPU architecture to create platforms for scientific computing, AI, data science, AV, robotics, metaverse and 3D internet applications.`	`How did NVIDIA pivot its GPU architecture usage beyond PC graphics?`

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        768,
        512,
        256,
        128,
        64
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: epoch
per_device_train_batch_size: 32
per_device_eval_batch_size: 16
gradient_accumulation_steps: 16
learning_rate: 2e-05
num_train_epochs: 20
lr_scheduler_type: cosine
warmup_ratio: 0.1
tf32: False
load_best_model_at_end: True
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: epoch
prediction_loss_only: True
per_device_train_batch_size: 32
per_device_eval_batch_size: 16
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 16
eval_accumulation_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: 20
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: False
fp16: False
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: False
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
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
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
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	dim_128_cosine_map@100	dim_256_cosine_map@100	dim_512_cosine_map@100	dim_64_cosine_map@100	dim_768_cosine_map@100
1.0	1	-	0.6952	0.6617	0.725	0.5966	0.7167
2.0	2	-	0.7060	0.75	0.8	0.6086	0.8
3.0	3	-	0.72	0.75	0.8	0.6277	0.75
4.0	4	-	0.8	0.75	0.8	0.6283	0.75
1.0	1	-	0.8	0.75	0.8	0.6283	0.75
2.0	2	-	0.8	0.75	0.8	0.6424	0.775
3.0	3	-	0.8	0.75	0.825	0.71	0.775
4.3333	5	-	0.825	0.825	0.825	0.735	0.775
5.0	6	-	0.825	0.825	0.825	0.735	0.775
6.0	7	-	0.825	0.825	0.825	0.735	0.825
7.0	8	-	0.825	0.825	0.825	0.735	0.825
8.3333	10	0.1558	0.825	0.825	0.825	0.7375	0.825
9.0	11	-	0.825	0.825	0.825	0.7375	0.825
10.0	12	-	0.825	0.825	0.825	0.7375	0.825
11.0	13	-	0.825	0.825	0.825	0.7361	0.825
12.3333	15	-	0.825	0.825	0.825	0.7361	0.825
13.0	16	-	0.825	0.825	0.825	0.7361	0.825
14.0	17	-	0.825	0.825	0.825	0.7361	0.825
15.0	18	-	0.825	0.825	0.825	0.7361	0.825
16.3333	20	0.0272	0.825	0.825	0.825	0.7361	0.825

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.10.12
Sentence Transformers: 3.0.1
Transformers: 4.41.2
PyTorch: 2.3.0+cu121
Accelerate: 0.32.1
Datasets: 2.20.0
Tokenizers: 0.19.1

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}
}