--- 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:6300 - loss:MatryoshkaLoss - loss:MultipleNegativesRankingLoss widget: - source_sentence: R&D expense increased by $304 million, or 14.9%, led by Intelligent Edge, HPC & AI and Storage in fiscal 2023. sentences: - What was the growth rate of Visa Inc.'s overall total nominal volume from 2021 to 2022? - How much did Hewlett Packard Enterprise's R&D expenses increase in fiscal 2023? - What is the purpose of the Global Day of Joy at Hasbro? - source_sentence: In 2022 and continuing into 2023, the Russia-Ukraine conflict was a catalyst for an energy crisis in Europe. Government interventions related to the energy crisis resulting from the Russia-Ukraine conflict, such as the Market Correction Mechanism (price cap), or interventions that may be proposed in the future related to the Russia-Ukraine conflict or the conflict in Israel and Gaza could also have a negative impact on our business. sentences: - What are Garmin's core strategies for reducing its environmental impact? - What are the potential consequences of the Russia-Ukraine conflict on a company's business? - What factors influence HP's critical accounting estimates? - source_sentence: The increase in other income, net was primarily due to an increase in interest income as a result of higher cash balances and higher interest rates. sentences: - What was the primary reason for the increase in other income, net during the noted period? - What led to the increase in room expenses at Las Vegas Sands Corp. in 2023? - What was the provision for income taxes for the year ended June 30, 2023? - source_sentence: When an investment declines below cost basis, management evaluates whether the decline in fair value is other than temporary. If deemed other than temporary, an impairment charge is recorded. sentences: - What are the requirements for Gilead's cell therapy products under the FDA's Risk Evaluation and Mitigation Strategy program? - What are the four focus areas declared by the company to strengthen their performance going forward? - What triggers the requirement for management to record an impairment charge for investments? - source_sentence: The total gross fair value of derivatives was listed as $422,232 million as per the latest financial data without adjustments for counterparty netting or collateral. sentences: - What was the total gross fair value of derivatives as of December 2023 before netting adjustments in the consolidated financial statements? - How does the company handle the recording and disclosure of contingent liabilities? - What is the significance of reporting financial results on a constant currency basis? 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: 0.7071428571428572 name: Cosine Accuracy@1 - type: cosine_accuracy@3 value: 0.8214285714285714 name: Cosine Accuracy@3 - type: cosine_accuracy@5 value: 0.8614285714285714 name: Cosine Accuracy@5 - type: cosine_accuracy@10 value: 0.9042857142857142 name: Cosine Accuracy@10 - type: cosine_precision@1 value: 0.7071428571428572 name: Cosine Precision@1 - type: cosine_precision@3 value: 0.2738095238095238 name: Cosine Precision@3 - type: cosine_precision@5 value: 0.17228571428571426 name: Cosine Precision@5 - type: cosine_precision@10 value: 0.09042857142857141 name: Cosine Precision@10 - type: cosine_recall@1 value: 0.7071428571428572 name: Cosine Recall@1 - type: cosine_recall@3 value: 0.8214285714285714 name: Cosine Recall@3 - type: cosine_recall@5 value: 0.8614285714285714 name: Cosine Recall@5 - type: cosine_recall@10 value: 0.9042857142857142 name: Cosine Recall@10 - type: cosine_ndcg@10 value: 0.8050065074948352 name: Cosine Ndcg@10 - type: cosine_mrr@10 value: 0.7732902494331064 name: Cosine Mrr@10 - type: cosine_map@100 value: 0.776990609765374 name: Cosine Map@100 - task: type: information-retrieval name: Information Retrieval dataset: name: dim 512 type: dim_512 metrics: - type: cosine_accuracy@1 value: 0.7014285714285714 name: Cosine Accuracy@1 - type: cosine_accuracy@3 value: 0.8214285714285714 name: Cosine Accuracy@3 - type: cosine_accuracy@5 value: 0.8657142857142858 name: Cosine Accuracy@5 - type: cosine_accuracy@10 value: 0.9057142857142857 name: Cosine Accuracy@10 - type: cosine_precision@1 value: 0.7014285714285714 name: Cosine Precision@1 - type: cosine_precision@3 value: 0.2738095238095238 name: Cosine Precision@3 - type: cosine_precision@5 value: 0.17314285714285713 name: Cosine Precision@5 - type: cosine_precision@10 value: 0.09057142857142855 name: Cosine Precision@10 - type: cosine_recall@1 value: 0.7014285714285714 name: Cosine Recall@1 - type: cosine_recall@3 value: 0.8214285714285714 name: Cosine Recall@3 - type: cosine_recall@5 value: 0.8657142857142858 name: Cosine Recall@5 - type: cosine_recall@10 value: 0.9057142857142857 name: Cosine Recall@10 - type: cosine_ndcg@10 value: 0.8035496957871646 name: Cosine Ndcg@10 - type: cosine_mrr@10 value: 0.7707964852607707 name: Cosine Mrr@10 - type: cosine_map@100 value: 0.7744696266512991 name: Cosine Map@100 - task: type: information-retrieval name: Information Retrieval dataset: name: dim 256 type: dim_256 metrics: - type: cosine_accuracy@1 value: 0.6885714285714286 name: Cosine Accuracy@1 - type: cosine_accuracy@3 value: 0.8157142857142857 name: Cosine Accuracy@3 - type: cosine_accuracy@5 value: 0.86 name: Cosine Accuracy@5 - type: cosine_accuracy@10 value: 0.9014285714285715 name: Cosine Accuracy@10 - type: cosine_precision@1 value: 0.6885714285714286 name: Cosine Precision@1 - type: cosine_precision@3 value: 0.27190476190476187 name: Cosine Precision@3 - type: cosine_precision@5 value: 0.172 name: Cosine Precision@5 - type: cosine_precision@10 value: 0.09014285714285714 name: Cosine Precision@10 - type: cosine_recall@1 value: 0.6885714285714286 name: Cosine Recall@1 - type: cosine_recall@3 value: 0.8157142857142857 name: Cosine Recall@3 - type: cosine_recall@5 value: 0.86 name: Cosine Recall@5 - type: cosine_recall@10 value: 0.9014285714285715 name: Cosine Recall@10 - type: cosine_ndcg@10 value: 0.7959304086509564 name: Cosine Ndcg@10 - type: cosine_mrr@10 value: 0.7620759637188204 name: Cosine Mrr@10 - type: cosine_map@100 value: 0.7656989001700307 name: Cosine Map@100 - task: type: information-retrieval name: Information Retrieval dataset: name: dim 128 type: dim_128 metrics: - type: cosine_accuracy@1 value: 0.6871428571428572 name: Cosine Accuracy@1 - type: cosine_accuracy@3 value: 0.7871428571428571 name: Cosine Accuracy@3 - type: cosine_accuracy@5 value: 0.8257142857142857 name: Cosine Accuracy@5 - type: cosine_accuracy@10 value: 0.8828571428571429 name: Cosine Accuracy@10 - type: cosine_precision@1 value: 0.6871428571428572 name: Cosine Precision@1 - type: cosine_precision@3 value: 0.2623809523809524 name: Cosine Precision@3 - type: cosine_precision@5 value: 0.16514285714285712 name: Cosine Precision@5 - type: cosine_precision@10 value: 0.08828571428571427 name: Cosine Precision@10 - type: cosine_recall@1 value: 0.6871428571428572 name: Cosine Recall@1 - type: cosine_recall@3 value: 0.7871428571428571 name: Cosine Recall@3 - type: cosine_recall@5 value: 0.8257142857142857 name: Cosine Recall@5 - type: cosine_recall@10 value: 0.8828571428571429 name: Cosine Recall@10 - type: cosine_ndcg@10 value: 0.7805054661054854 name: Cosine Ndcg@10 - type: cosine_mrr@10 value: 0.7483526077097503 name: Cosine Mrr@10 - type: cosine_map@100 value: 0.7524860233992903 name: Cosine Map@100 - task: type: information-retrieval name: Information Retrieval dataset: name: dim 64 type: dim_64 metrics: - type: cosine_accuracy@1 value: 0.64 name: Cosine Accuracy@1 - type: cosine_accuracy@3 value: 0.7557142857142857 name: Cosine Accuracy@3 - type: cosine_accuracy@5 value: 0.7828571428571428 name: Cosine Accuracy@5 - type: cosine_accuracy@10 value: 0.8428571428571429 name: Cosine Accuracy@10 - type: cosine_precision@1 value: 0.64 name: Cosine Precision@1 - type: cosine_precision@3 value: 0.25190476190476185 name: Cosine Precision@3 - type: cosine_precision@5 value: 0.15657142857142856 name: Cosine Precision@5 - type: cosine_precision@10 value: 0.08428571428571427 name: Cosine Precision@10 - type: cosine_recall@1 value: 0.64 name: Cosine Recall@1 - type: cosine_recall@3 value: 0.7557142857142857 name: Cosine Recall@3 - type: cosine_recall@5 value: 0.7828571428571428 name: Cosine Recall@5 - type: cosine_recall@10 value: 0.8428571428571429 name: Cosine Recall@10 - type: cosine_ndcg@10 value: 0.7386047605712329 name: Cosine Ndcg@10 - type: cosine_mrr@10 value: 0.7057772108843535 name: Cosine Mrr@10 - type: cosine_map@100 value: 0.7112870933540941 name: Cosine Map@100 --- # BGE base Financial Matryoshka This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-base-en-v1.5](https://huggingface.co./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](https://huggingface.co./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](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co./models?library=sentence-transformers) ### 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: ```bash 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("moritzglnr/bge-base-financial-matryoshka") # Run inference sentences = [ 'The total gross fair value of derivatives was listed as $422,232 million as per the latest financial data without adjustments for counterparty netting or collateral.', 'What was the total gross fair value of derivatives as of December 2023 before netting adjustments in the consolidated financial statements?', 'How does the company handle the recording and disclosure of contingent liabilities?', ] 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](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) | Metric | Value | |:--------------------|:----------| | cosine_accuracy@1 | 0.7071 | | cosine_accuracy@3 | 0.8214 | | cosine_accuracy@5 | 0.8614 | | cosine_accuracy@10 | 0.9043 | | cosine_precision@1 | 0.7071 | | cosine_precision@3 | 0.2738 | | cosine_precision@5 | 0.1723 | | cosine_precision@10 | 0.0904 | | cosine_recall@1 | 0.7071 | | cosine_recall@3 | 0.8214 | | cosine_recall@5 | 0.8614 | | cosine_recall@10 | 0.9043 | | cosine_ndcg@10 | 0.805 | | cosine_mrr@10 | 0.7733 | | **cosine_map@100** | **0.777** | #### Information Retrieval * Dataset: `dim_512` * Evaluated with [InformationRetrievalEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) | Metric | Value | |:--------------------|:-----------| | cosine_accuracy@1 | 0.7014 | | cosine_accuracy@3 | 0.8214 | | cosine_accuracy@5 | 0.8657 | | cosine_accuracy@10 | 0.9057 | | cosine_precision@1 | 0.7014 | | cosine_precision@3 | 0.2738 | | cosine_precision@5 | 0.1731 | | cosine_precision@10 | 0.0906 | | cosine_recall@1 | 0.7014 | | cosine_recall@3 | 0.8214 | | cosine_recall@5 | 0.8657 | | cosine_recall@10 | 0.9057 | | cosine_ndcg@10 | 0.8035 | | cosine_mrr@10 | 0.7708 | | **cosine_map@100** | **0.7745** | #### Information Retrieval * Dataset: `dim_256` * Evaluated with [InformationRetrievalEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) | Metric | Value | |:--------------------|:-----------| | cosine_accuracy@1 | 0.6886 | | cosine_accuracy@3 | 0.8157 | | cosine_accuracy@5 | 0.86 | | cosine_accuracy@10 | 0.9014 | | cosine_precision@1 | 0.6886 | | cosine_precision@3 | 0.2719 | | cosine_precision@5 | 0.172 | | cosine_precision@10 | 0.0901 | | cosine_recall@1 | 0.6886 | | cosine_recall@3 | 0.8157 | | cosine_recall@5 | 0.86 | | cosine_recall@10 | 0.9014 | | cosine_ndcg@10 | 0.7959 | | cosine_mrr@10 | 0.7621 | | **cosine_map@100** | **0.7657** | #### Information Retrieval * Dataset: `dim_128` * Evaluated with [InformationRetrievalEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) | Metric | Value | |:--------------------|:-----------| | cosine_accuracy@1 | 0.6871 | | cosine_accuracy@3 | 0.7871 | | cosine_accuracy@5 | 0.8257 | | cosine_accuracy@10 | 0.8829 | | cosine_precision@1 | 0.6871 | | cosine_precision@3 | 0.2624 | | cosine_precision@5 | 0.1651 | | cosine_precision@10 | 0.0883 | | cosine_recall@1 | 0.6871 | | cosine_recall@3 | 0.7871 | | cosine_recall@5 | 0.8257 | | cosine_recall@10 | 0.8829 | | cosine_ndcg@10 | 0.7805 | | cosine_mrr@10 | 0.7484 | | **cosine_map@100** | **0.7525** | #### Information Retrieval * Dataset: `dim_64` * Evaluated with [InformationRetrievalEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator) | Metric | Value | |:--------------------|:-----------| | cosine_accuracy@1 | 0.64 | | cosine_accuracy@3 | 0.7557 | | cosine_accuracy@5 | 0.7829 | | cosine_accuracy@10 | 0.8429 | | cosine_precision@1 | 0.64 | | cosine_precision@3 | 0.2519 | | cosine_precision@5 | 0.1566 | | cosine_precision@10 | 0.0843 | | cosine_recall@1 | 0.64 | | cosine_recall@3 | 0.7557 | | cosine_recall@5 | 0.7829 | | cosine_recall@10 | 0.8429 | | cosine_ndcg@10 | 0.7386 | | cosine_mrr@10 | 0.7058 | | **cosine_map@100** | **0.7113** | ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 6,300 training samples * Columns: positive and anchor * Approximate statistics based on the first 1000 samples: | | positive | anchor | |:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------| | type | string | string | | details | | | * Samples: | positive | anchor | |:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------| | The 2023 Form 10-K for Delta Air Lines, Inc. includes various types of financial statements such as consolidated balance sheets, consolidated statements of operations, comprehensive income, cash flows, and stockholders' equity. | What are the primary types of financial statements included in Delta Air Lines, Inc.'s 2023 Form 10-K? | | Critical accounting estimates are those that involve a significant level of estimation uncertainty and have had or are reasonably likely to have a material impact on HP's financial condition or results of operations. | What factors influence HP's critical accounting estimates? | | The requisite service period for both employee stock options and RSUs is generally four years from the grant date. | What is the recognition period for Etsy's stock options and RSUs granted to employees? | * Loss: [MatryoshkaLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters: ```json { "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 - `per_device_train_batch_size`: 32 - `per_device_eval_batch_size`: 16 - `gradient_accumulation_steps`: 16 - `learning_rate`: 2e-05 - `num_train_epochs`: 1 - `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 - `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`: 1 - `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 - `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, '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_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 | |:----------:|:------:|:-------------:|:----------------------:|:----------------------:|:----------------------:|:---------------------:|:----------------------:| | 0.8122 | 10 | 1.4747 | - | - | - | - | - | | **0.9746** | **12** | **-** | **0.7525** | **0.7657** | **0.7745** | **0.7113** | **0.777** | * The bold row denotes the saved checkpoint. ### Framework Versions - Python: 3.11.9 - Sentence Transformers: 3.0.1 - Transformers: 4.40.2 - PyTorch: 2.3.1 - Accelerate: 0.32.1 - Datasets: 2.20.0 - Tokenizers: 0.19.1 ## Citation ### BibTeX #### Sentence Transformers ```bibtex @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 ```bibtex @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 ```bibtex @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} } ```