--- tags: - sentence-transformers - cross-encoder - reranker - generated_from_trainer - dataset_size:203170 - loss:BinaryCrossEntropyLoss base_model: BAAI/bge-reranker-v2-m3 pipeline_tag: text-ranking library_name: sentence-transformers metrics: - map - mrr@10 - ndcg@10 model-index: - name: CrossEncoder based on BAAI/bge-reranker-v2-m3 results: - task: type: cross-encoder-reranking name: Cross Encoder Reranking dataset: name: gcc legal rerank type: gcc-legal-rerank metrics: - type: map value: 0.988572796934866 name: Map - type: mrr@10 value: 0.988572796934866 name: Mrr@10 - type: ndcg@10 value: 0.991518139889259 name: Ndcg@10 --- # CrossEncoder based on BAAI/bge-reranker-v2-m3 This is a [Cross Encoder](https://www.sbert.net/docs/cross_encoder/usage/usage.html) model finetuned from [BAAI/bge-reranker-v2-m3](https://huggingface.co/BAAI/bge-reranker-v2-m3) using the [sentence-transformers](https://www.SBERT.net) library. It computes scores for pairs of texts, which can be used for text reranking and semantic search. ## Model Details ### Model Description - **Model Type:** Cross Encoder - **Base model:** [BAAI/bge-reranker-v2-m3](https://huggingface.co/BAAI/bge-reranker-v2-m3) - **Maximum Sequence Length:** 1024 tokens - **Number of Output Labels:** 1 label - **Supported Modality:** Text ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Documentation:** [Cross Encoder Documentation](https://www.sbert.net/docs/cross_encoder/usage/usage.html) - **Repository:** [Sentence Transformers on GitHub](https://github.com/huggingface/sentence-transformers) - **Hugging Face:** [Cross Encoders on Hugging Face](https://huggingface.co/models?library=sentence-transformers&other=cross-encoder) ### Full Model Architecture ``` CrossEncoder( (0): Transformer({'transformer_task': 'sequence-classification', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'logits'}}, 'module_output_name': 'scores', 'architecture': 'XLMRobertaForSequenceClassification'}) ) ``` ## 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 CrossEncoder # Download from the 🤗 Hub model = CrossEncoder("cross_encoder_model_id") # Get scores for pairs of inputs pairs = [ ['AAOIFI conference Islamic finance AI artificial intelligence GCC', "[GCC | AAOIFI | AAOIFI Financial Accounting Standards (FAS)]\n\nSession two centered on the role of Islamic finance in AI-driven business and finance environment in which experts\nexamined the transformative potential of artificial intelligence across industries, highlighting both the opportunities it\npresents and the challenges it may pose for Islamic finance.\nThe following session tackled the all-important Islamic financial literacy and awareness, as it extends beyond capacity-\nbuilding and highlighted the strategies to elevate Islamic financial awareness and learning beyond its classic role for the\nnext generation of professionals and practitioners. It also explored the way by which technology can bridge the gaps in\nlearning accessibility and how Islamic financial literacy can surpass traditional capacity-building initiatives and individual\nlearning.\nThe day one of the event brought together a physical turnout exceeding 700 participants, in addition to over 900 online\nparticipations from around the world. To ensure accessibility and broader participation, all sessions were being streamed\nlive in English and Arabic on AAOIFI's official digital and social media pages.\nAAOIFI looks forward to the continuation of the conference on day two, which will cover a host of key issues such\nentrepreneurship development through Shari'ah compliant MSME financing and creation of integrated value chains\nand ecosystem, as well as Salam and Istisna and application of these modes of finance for economic growth and\ndevelopment. The session will feature discussion on practicality considerations, market reality and accounting\nchallenges.\nThis year's edition of the conference was supported by a number of key stakeholders including major partnership with\nJumhouria Bank (Libya), Kuwait Finance House (KFH), and Saudi National Bank (SNB), alongside golden partners such as\nKhaleeji Commercial Bank, Al Baraka Banking Group, Abu Dhabi Islamic Bank (ADIB), Bahrain Islamic Bank (BiSB), Abu Dhabi\nIslamic Bank (ADIB) and DDCAP Group.\nOn this special occasion, AAOIFI would like to extend its appreciation and gratitude to all participants, scholars, speakers,\nsponsors and media partners for their invaluable contributions towards making this event a success."], ['AAOIFI conference Islamic finance AI artificial intelligence GCC', "[GCC | IFSB | Islamic Financial Services Board (IFSB) Standards]\n\nPANELLIST 3\nAnggoro Eko Cahyo\nPresident Director, Bank Syariah Indonesia\nIndonesia's vast Muslim population presents immense potential for Islamic banking\ngrowth. Since its establishment in 2021, Bank Syariah Indonesia (BSI) has\nstrengthened market presence and positioned Indonesia as the second-largest\ncontributor to Islamic banking assets in Asia-Pacific. The shift in customer\nbehavior toward Sharia-compliant products combining spiritual, functional, and\nemotional values supports this progress. As global megatrends reshape the\nfinancial landscape—driven by technology, demographics, and sustainability—BSI\nis accelerating its digital transformation through initiatives such as BSI Mobile, BSI\nNet, BSI Agent, QRIS, and super apps like BYOND and BEWIZE. Moving forward,\nBSI integrates AI-based analytics for customer segmentation, risk management,\nand fraud detection, reflecting its ambition to compete globally while remaining\ninclusive and fully Sharia-compliant.\nIntroduction\nIndonesia, home to one of the world's largest Muslim populations, holds vast\nuntapped potential in Islamic finance. Despite its demographic advantage, the\npenetration of Sharia-compliant financial services has historically been modest.\nThe establishment of Bank Syariah Indonesia (BSI) in 2021 marked a milestone,"], ['AAOIFI conference Islamic finance AI artificial intelligence GCC', '[GCC | AAOIFI | AAOIFI Financial Accounting Standards (FAS)]\n\nEsteemed speakers and panelists\nH.E. Shaikh Ebrahim Bin H.E. Midkhat H.E. Mr. Yahya Shunnar Dr. Sami Al Suwailem\nKhalifa Al Khalifa Shagiakhmetov Governor Acting Director General, Islamic\nPalestine Monetary Authority, Palestine Development Bank Institute\nChairman, AAOIFI Board of Trustees Deputy Prime Minister, Minister of\nEconomy, Republic of Tatarstan\nMr. Omar Mustafa Ansari Mr. Farrukh Raza Mr. Roustam Vakhitov Mr. Ashar Nazim\nSecretary General, AAOIFI, Group CEO and Co-founder, IFAAS; Associated International Tax Partner, Chief Executive Officer, Aion Digital\nMember, AAOIFI Governance and Ethics Crowe, Kazakhstan\nBoard\nMr. Yahya Aleem Ur Rahman Mr. Naser Alawadhi Mr. Syed Amir Ali Mr. Noor ur Rahman Abid\nGlobal Lead, Islamic Finance Advisory Head of Digital Banking Deputy Chief Executive Officer, Former Managing Partner, Ernst &\nKnowledge Solutions, Islamic Bahrain Islamic Bank Meezan bank, Islamic Republic of Pakistan Young; Member, AAOIFI BOT\nDevelopment Bank Institute\nMr. Khaled Al Kayed Mr. Muhammad Nazir Mian Dr. Moutaz Abojeib Mr. Hamza Bawazir\nChief Executive Officer, Bank Nizwa Head of Group Internal Sharia Control, Director, IFAAS Operations Secretary General, Council for Islamic Banks\nDubai Islamic Bank Group, United Arab and Financial Institutions (CIBAFI), Kingdom\nIFAAS Group\nEmirates of Bahrain'], ['AAOIFI conference Islamic finance AI artificial intelligence GCC', '[GCC | AAOIFI | AAOIFI Shariah Standards (SS 1–60+)]\n\n20th AAOIFI-IsDB Annual Islamic Banking and Finance Conference 2025\n2-3 November 2025, Kingdom of Bahrain\nConference Overview\nThe 20th AAOIFI-IsDB Islamic Banking and Finance Conference, convened on 2-3 November 2025 in the Kingdom\nof Bahrain, brought together more than 700 in-person and over 800 virtual participants from across the globe.\nThe conference gathered leading industry experts, Shari\'ah scholars, regulators, policymakers, academicians, and\ninternational service providers to deliberate on a wide range of pertinent topics aligned with the theme "Islamic Finance\nin the Era of Artificial Intelligence (AI): Present Potentials and Future Prospects."\nThis edition witnessed high-level international participation, including a delegation from the Republic of Tatarstan,\nRussian Federation, led by H.E. Mr. Midkhat Shagiakhmetov, deputy prime minister and minister of economy. The\nconference also benefited from the strong support and cooperation of the Islamic Development Bank through the\nIslamic Development Bank Institute (IsDBI), under the leadership of its Director General, H.E. Dr. Sami Al-Suwailem, as\nwell as the active participation of the United Nations Industrial Development Organization (UNIDO).\nSenior representatives from around 25 regulatory and supervisory authorities worldwide took part in the conference,\nsuch as the Bank of Russia, Central Bank of Iraq, State Bank of Pakistan, Banque du Liban, Central Bank of Syria, Central\nBank of Nigeria, Securities Commission of Nigeria, Central Bank of Somalia, Securities and Exchange Commission of\nPakistan, Palestinian Monetary Authority, Bank of Tanzania, Central Bank of Uzbekistan, Financial Regulatory Authority\nof Oman, Bangladesh Bank, Indonesian Accounting Authority, Central Bank of Kosovo, Agency for the Republic of\nKazakhstan for Regulation and Development of the Financial Market and the Central Bank of Oman.\nThe flagship conference was held alongside the AAOIFI Capacity-Building Week (CBW) 2025 (3rd edition), which\nattracted more than 120 trainees from across the globe, further reinforcing AAOIFI\'s mandate in professional\ndevelopment and standardisation.\nThe conference addressed critical global economic developments reshaping the Islamic finance industry, with particular\nfocus on artificial intelligence and its implications for business operations, Shari\'ah compliance and the mobilisation of\nhuman and technological resources. It also underscored the importance of practical understanding of Islamic finance for\nempowering individuals and professionals to integrate ethical principles into daily operations and decision-making, while\nhighlighting the value proposition of Islamic finance for micro, small, and medium-sized enterprises (MSMEs).\nDay-one proceedings\nTheme: Islamic finance in the era of artificial intelligence: Present potentials and future prospects\nDate: Sunday, 2 November 2025\nVenue: Crowne Plaza Hotel, Kingdom of Bahrain\nOpening ceremony\n- Recitation from the Holy Qur\'an\n- Conference host\'s keynote speech'], ['AAOIFI conference Islamic finance AI artificial intelligence GCC', "[GCC | IFSB | Islamic Financial Services Board (IFSB) Standards]\n\nFuture Outlook\nLooking ahead, BSI is integrating artificial intelligence (AI) to enhance operational\nprecision and customer experience. AI tools are being applied for customer\nsegmentation, fraud detection, churn prevention, and MSME risk assessment,\nimproving responsiveness and efficiency. These initiatives reflect BSI's strategic\nvision to become a globally competitive Islamic digital bank, capable of blending\ninnovation with ethical finance.\nBSI's digital transformation demonstrates how Islamic banking can effectively\nnavigate global megatrends while upholding its ethical and spiritual foundations.\nAs emphasized by Anggoro Eko Cahyo, digital technology and inclusivity are key\nto unlocking Indonesia's financial potential. By integrating Sharia values with\ntechnological innovation, BSI is creating a resilient, inclusive, and future-ready\nIslamic financial ecosystem—one that not only strengthens Indonesia's economic\ngrowth but also contributes meaningfully to the global advancement of Islamic\nfinance.\n\nPANELLIST 4\nEka Nilam Dari\nPresident Director,\nPT Airpay International Indonesia\nShopeePay is advancing Indonesia's digital financial ecosystem by providing\nsecure, practical, and mobile-friendly solutions that align with Bank Indonesia's\nPayment System Blueprint 2030. In a country with high mobile phone penetration\nbut uneven financial literacy and fragmented providers, ShopeePay bridges\naccessibility gaps through inclusive digital services. Integrated with the Shopee\nmarketplace, QRIS, BI-FAST, and SeaBank, ShopeePay empowers millions of\nusers and MSMEs nationwide. Strong safeguards such as fraud detection, KYC,\nand ISO-certified security build user trust, while cross-border QRIS payments\nenhance regional connectivity. ShopeePay also supports sharia finance, micro-\ninvestments, and MSME literacy programs like Kelas TUNAI, promoting digital\nadoption, empowerment, and sustainable economic growth.\nIntroduction\nIndonesia's digital financial landscape is evolving rapidly, supported by high mobile\npenetration and expanding digital infrastructure. However, challenges remain in\nthe form of uneven financial literacy, low credit card adoption, and fragmented\nservice providers. Within this dynamic environment, ShopeePay has emerged as a"], ] scores = model.predict(pairs) print(scores) # [9.9914e-01 9.0280e-05 2.7803e-04 9.9711e-01 1.5012e-03] # Or rank different texts based on similarity to a single text ranks = model.rank( 'AAOIFI conference Islamic finance AI artificial intelligence GCC', [ "[GCC | AAOIFI | AAOIFI Financial Accounting Standards (FAS)]\n\nSession two centered on the role of Islamic finance in AI-driven business and finance environment in which experts\nexamined the transformative potential of artificial intelligence across industries, highlighting both the opportunities it\npresents and the challenges it may pose for Islamic finance.\nThe following session tackled the all-important Islamic financial literacy and awareness, as it extends beyond capacity-\nbuilding and highlighted the strategies to elevate Islamic financial awareness and learning beyond its classic role for the\nnext generation of professionals and practitioners. It also explored the way by which technology can bridge the gaps in\nlearning accessibility and how Islamic financial literacy can surpass traditional capacity-building initiatives and individual\nlearning.\nThe day one of the event brought together a physical turnout exceeding 700 participants, in addition to over 900 online\nparticipations from around the world. To ensure accessibility and broader participation, all sessions were being streamed\nlive in English and Arabic on AAOIFI's official digital and social media pages.\nAAOIFI looks forward to the continuation of the conference on day two, which will cover a host of key issues such\nentrepreneurship development through Shari'ah compliant MSME financing and creation of integrated value chains\nand ecosystem, as well as Salam and Istisna and application of these modes of finance for economic growth and\ndevelopment. The session will feature discussion on practicality considerations, market reality and accounting\nchallenges.\nThis year's edition of the conference was supported by a number of key stakeholders including major partnership with\nJumhouria Bank (Libya), Kuwait Finance House (KFH), and Saudi National Bank (SNB), alongside golden partners such as\nKhaleeji Commercial Bank, Al Baraka Banking Group, Abu Dhabi Islamic Bank (ADIB), Bahrain Islamic Bank (BiSB), Abu Dhabi\nIslamic Bank (ADIB) and DDCAP Group.\nOn this special occasion, AAOIFI would like to extend its appreciation and gratitude to all participants, scholars, speakers,\nsponsors and media partners for their invaluable contributions towards making this event a success.", "[GCC | IFSB | Islamic Financial Services Board (IFSB) Standards]\n\nPANELLIST 3\nAnggoro Eko Cahyo\nPresident Director, Bank Syariah Indonesia\nIndonesia's vast Muslim population presents immense potential for Islamic banking\ngrowth. Since its establishment in 2021, Bank Syariah Indonesia (BSI) has\nstrengthened market presence and positioned Indonesia as the second-largest\ncontributor to Islamic banking assets in Asia-Pacific. The shift in customer\nbehavior toward Sharia-compliant products combining spiritual, functional, and\nemotional values supports this progress. As global megatrends reshape the\nfinancial landscape—driven by technology, demographics, and sustainability—BSI\nis accelerating its digital transformation through initiatives such as BSI Mobile, BSI\nNet, BSI Agent, QRIS, and super apps like BYOND and BEWIZE. Moving forward,\nBSI integrates AI-based analytics for customer segmentation, risk management,\nand fraud detection, reflecting its ambition to compete globally while remaining\ninclusive and fully Sharia-compliant.\nIntroduction\nIndonesia, home to one of the world's largest Muslim populations, holds vast\nuntapped potential in Islamic finance. Despite its demographic advantage, the\npenetration of Sharia-compliant financial services has historically been modest.\nThe establishment of Bank Syariah Indonesia (BSI) in 2021 marked a milestone,", '[GCC | AAOIFI | AAOIFI Financial Accounting Standards (FAS)]\n\nEsteemed speakers and panelists\nH.E. Shaikh Ebrahim Bin H.E. Midkhat H.E. Mr. Yahya Shunnar Dr. Sami Al Suwailem\nKhalifa Al Khalifa Shagiakhmetov Governor Acting Director General, Islamic\nPalestine Monetary Authority, Palestine Development Bank Institute\nChairman, AAOIFI Board of Trustees Deputy Prime Minister, Minister of\nEconomy, Republic of Tatarstan\nMr. Omar Mustafa Ansari Mr. Farrukh Raza Mr. Roustam Vakhitov Mr. Ashar Nazim\nSecretary General, AAOIFI, Group CEO and Co-founder, IFAAS; Associated International Tax Partner, Chief Executive Officer, Aion Digital\nMember, AAOIFI Governance and Ethics Crowe, Kazakhstan\nBoard\nMr. Yahya Aleem Ur Rahman Mr. Naser Alawadhi Mr. Syed Amir Ali Mr. Noor ur Rahman Abid\nGlobal Lead, Islamic Finance Advisory Head of Digital Banking Deputy Chief Executive Officer, Former Managing Partner, Ernst &\nKnowledge Solutions, Islamic Bahrain Islamic Bank Meezan bank, Islamic Republic of Pakistan Young; Member, AAOIFI BOT\nDevelopment Bank Institute\nMr. Khaled Al Kayed Mr. Muhammad Nazir Mian Dr. Moutaz Abojeib Mr. Hamza Bawazir\nChief Executive Officer, Bank Nizwa Head of Group Internal Sharia Control, Director, IFAAS Operations Secretary General, Council for Islamic Banks\nDubai Islamic Bank Group, United Arab and Financial Institutions (CIBAFI), Kingdom\nIFAAS Group\nEmirates of Bahrain', '[GCC | AAOIFI | AAOIFI Shariah Standards (SS 1–60+)]\n\n20th AAOIFI-IsDB Annual Islamic Banking and Finance Conference 2025\n2-3 November 2025, Kingdom of Bahrain\nConference Overview\nThe 20th AAOIFI-IsDB Islamic Banking and Finance Conference, convened on 2-3 November 2025 in the Kingdom\nof Bahrain, brought together more than 700 in-person and over 800 virtual participants from across the globe.\nThe conference gathered leading industry experts, Shari\'ah scholars, regulators, policymakers, academicians, and\ninternational service providers to deliberate on a wide range of pertinent topics aligned with the theme "Islamic Finance\nin the Era of Artificial Intelligence (AI): Present Potentials and Future Prospects."\nThis edition witnessed high-level international participation, including a delegation from the Republic of Tatarstan,\nRussian Federation, led by H.E. Mr. Midkhat Shagiakhmetov, deputy prime minister and minister of economy. The\nconference also benefited from the strong support and cooperation of the Islamic Development Bank through the\nIslamic Development Bank Institute (IsDBI), under the leadership of its Director General, H.E. Dr. Sami Al-Suwailem, as\nwell as the active participation of the United Nations Industrial Development Organization (UNIDO).\nSenior representatives from around 25 regulatory and supervisory authorities worldwide took part in the conference,\nsuch as the Bank of Russia, Central Bank of Iraq, State Bank of Pakistan, Banque du Liban, Central Bank of Syria, Central\nBank of Nigeria, Securities Commission of Nigeria, Central Bank of Somalia, Securities and Exchange Commission of\nPakistan, Palestinian Monetary Authority, Bank of Tanzania, Central Bank of Uzbekistan, Financial Regulatory Authority\nof Oman, Bangladesh Bank, Indonesian Accounting Authority, Central Bank of Kosovo, Agency for the Republic of\nKazakhstan for Regulation and Development of the Financial Market and the Central Bank of Oman.\nThe flagship conference was held alongside the AAOIFI Capacity-Building Week (CBW) 2025 (3rd edition), which\nattracted more than 120 trainees from across the globe, further reinforcing AAOIFI\'s mandate in professional\ndevelopment and standardisation.\nThe conference addressed critical global economic developments reshaping the Islamic finance industry, with particular\nfocus on artificial intelligence and its implications for business operations, Shari\'ah compliance and the mobilisation of\nhuman and technological resources. It also underscored the importance of practical understanding of Islamic finance for\nempowering individuals and professionals to integrate ethical principles into daily operations and decision-making, while\nhighlighting the value proposition of Islamic finance for micro, small, and medium-sized enterprises (MSMEs).\nDay-one proceedings\nTheme: Islamic finance in the era of artificial intelligence: Present potentials and future prospects\nDate: Sunday, 2 November 2025\nVenue: Crowne Plaza Hotel, Kingdom of Bahrain\nOpening ceremony\n- Recitation from the Holy Qur\'an\n- Conference host\'s keynote speech', "[GCC | IFSB | Islamic Financial Services Board (IFSB) Standards]\n\nFuture Outlook\nLooking ahead, BSI is integrating artificial intelligence (AI) to enhance operational\nprecision and customer experience. AI tools are being applied for customer\nsegmentation, fraud detection, churn prevention, and MSME risk assessment,\nimproving responsiveness and efficiency. These initiatives reflect BSI's strategic\nvision to become a globally competitive Islamic digital bank, capable of blending\ninnovation with ethical finance.\nBSI's digital transformation demonstrates how Islamic banking can effectively\nnavigate global megatrends while upholding its ethical and spiritual foundations.\nAs emphasized by Anggoro Eko Cahyo, digital technology and inclusivity are key\nto unlocking Indonesia's financial potential. By integrating Sharia values with\ntechnological innovation, BSI is creating a resilient, inclusive, and future-ready\nIslamic financial ecosystem—one that not only strengthens Indonesia's economic\ngrowth but also contributes meaningfully to the global advancement of Islamic\nfinance.\n\nPANELLIST 4\nEka Nilam Dari\nPresident Director,\nPT Airpay International Indonesia\nShopeePay is advancing Indonesia's digital financial ecosystem by providing\nsecure, practical, and mobile-friendly solutions that align with Bank Indonesia's\nPayment System Blueprint 2030. In a country with high mobile phone penetration\nbut uneven financial literacy and fragmented providers, ShopeePay bridges\naccessibility gaps through inclusive digital services. Integrated with the Shopee\nmarketplace, QRIS, BI-FAST, and SeaBank, ShopeePay empowers millions of\nusers and MSMEs nationwide. Strong safeguards such as fraud detection, KYC,\nand ISO-certified security build user trust, while cross-border QRIS payments\nenhance regional connectivity. ShopeePay also supports sharia finance, micro-\ninvestments, and MSME literacy programs like Kelas TUNAI, promoting digital\nadoption, empowerment, and sustainable economic growth.\nIntroduction\nIndonesia's digital financial landscape is evolving rapidly, supported by high mobile\npenetration and expanding digital infrastructure. However, challenges remain in\nthe form of uneven financial literacy, low credit card adoption, and fragmented\nservice providers. Within this dynamic environment, ShopeePay has emerged as a", ] ) # [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...] ``` ## Evaluation ### Metrics #### Cross Encoder Reranking * Dataset: `gcc-legal-rerank` * Evaluated with [CrossEncoderRerankingEvaluator](https://sbert.net/docs/package_reference/cross_encoder/evaluation.html#sentence_transformers.cross_encoder.evaluation.CrossEncoderRerankingEvaluator) with these parameters: ```json { "at_k": 10 } ``` | Metric | Value | |:------------|:-----------| | map | 0.9886 | | mrr@10 | 0.9886 | | **ndcg@10** | **0.9915** | ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 203,170 training samples * Columns: query, doc, and label * Approximate statistics based on the first 1000 samples: | | query | doc | label | |:--------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------|:--------------------------------------------------------------| | type | string | string | float | | details | | | | * Samples: | query | doc | label | |:------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------| | AAOIFI conference Islamic finance AI artificial intelligence GCC | [GCC | AAOIFI | AAOIFI Financial Accounting Standards (FAS)] Session two centered on the role of Islamic finance in AI-driven business and finance environment in which experts examined the transformative potential of artificial intelligence across industries, highlighting both the opportunities it presents and the challenges it may pose for Islamic finance. The following session tackled the all-important Islamic financial literacy and awareness, as it extends beyond capacity- building and highlighted the strategies to elevate Islamic financial awareness and learning beyond its classic role for the next generation of professionals and practitioners. It also explored the way by which technology can bridge the gaps in learning accessibility and how Islamic financial literacy can surpass traditional capacity-building initiatives and individual learning. The day one of the event brought together a physical turnout exceeding 700 participants, in addition to over 900 online participations fr... | 1.0 | | AAOIFI conference Islamic finance AI artificial intelligence GCC | [GCC | IFSB | Islamic Financial Services Board (IFSB) Standards] PANELLIST 3 Anggoro Eko Cahyo President Director, Bank Syariah Indonesia Indonesia's vast Muslim population presents immense potential for Islamic banking growth. Since its establishment in 2021, Bank Syariah Indonesia (BSI) has strengthened market presence and positioned Indonesia as the second-largest contributor to Islamic banking assets in Asia-Pacific. The shift in customer behavior toward Sharia-compliant products combining spiritual, functional, and emotional values supports this progress. As global megatrends reshape the financial landscape—driven by technology, demographics, and sustainability—BSI is accelerating its digital transformation through initiatives such as BSI Mobile, BSI Net, BSI Agent, QRIS, and super apps like BYOND and BEWIZE. Moving forward, BSI integrates AI-based analytics for customer segmentation, risk management, and fraud detection, reflecting its ambition to compete globally while remainin... | 0.0 | | AAOIFI conference Islamic finance AI artificial intelligence GCC | [GCC | AAOIFI | AAOIFI Financial Accounting Standards (FAS)] Esteemed speakers and panelists H.E. Shaikh Ebrahim Bin H.E. Midkhat H.E. Mr. Yahya Shunnar Dr. Sami Al Suwailem Khalifa Al Khalifa Shagiakhmetov Governor Acting Director General, Islamic Palestine Monetary Authority, Palestine Development Bank Institute Chairman, AAOIFI Board of Trustees Deputy Prime Minister, Minister of Economy, Republic of Tatarstan Mr. Omar Mustafa Ansari Mr. Farrukh Raza Mr. Roustam Vakhitov Mr. Ashar Nazim Secretary General, AAOIFI, Group CEO and Co-founder, IFAAS; Associated International Tax Partner, Chief Executive Officer, Aion Digital Member, AAOIFI Governance and Ethics Crowe, Kazakhstan Board Mr. Yahya Aleem Ur Rahman Mr. Naser Alawadhi Mr. Syed Amir Ali Mr. Noor ur Rahman Abid Global Lead, Islamic Finance Advisory Head of Digital Banking Deputy Chief Executive Officer, Former Managing Partner, Ernst & Knowledge Solutions, Islamic Bahrain Islamic Bank Meezan bank, Islamic Republic of Pakistan Y... | 0.0 | * Loss: [BinaryCrossEntropyLoss](https://sbert.net/docs/package_reference/cross_encoder/losses.html#binarycrossentropyloss) with these parameters: ```json { "activation_fn": "torch.nn.modules.linear.Identity", "pos_weight": null } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `per_device_train_batch_size`: 16 - `num_train_epochs`: 1 - `learning_rate`: 2e-05 - `warmup_steps`: 0.1 - `gradient_accumulation_steps`: 2 - `bf16`: True - `load_best_model_at_end`: True #### All Hyperparameters
Click to expand - `per_device_train_batch_size`: 16 - `num_train_epochs`: 1 - `max_steps`: -1 - `learning_rate`: 2e-05 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: None - `warmup_steps`: 0.1 - `optim`: adamw_torch_fused - `optim_args`: None - `weight_decay`: 0.0 - `adam_beta1`: 0.9 - `adam_beta2`: 0.999 - `adam_epsilon`: 1e-08 - `optim_target_modules`: None - `gradient_accumulation_steps`: 2 - `average_tokens_across_devices`: True - `max_grad_norm`: 1.0 - `label_smoothing_factor`: 0.0 - `bf16`: True - `fp16`: False - `bf16_full_eval`: False - `fp16_full_eval`: False - `tf32`: None - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `torch_compile`: False - `torch_compile_backend`: None - `torch_compile_mode`: None - `use_liger_kernel`: False - `liger_kernel_config`: None - `use_cache`: False - `neftune_noise_alpha`: None - `torch_empty_cache_steps`: None - `auto_find_batch_size`: False - `log_on_each_node`: True - `logging_nan_inf_filter`: True - `include_num_input_tokens_seen`: no - `log_level`: passive - `log_level_replica`: warning - `disable_tqdm`: False - `project`: huggingface - `trackio_space_id`: None - `trackio_bucket_id`: None - `trackio_static_space_id`: None - `per_device_eval_batch_size`: 8 - `prediction_loss_only`: True - `eval_on_start`: False - `eval_do_concat_batches`: True - `eval_use_gather_object`: False - `eval_accumulation_steps`: None - `include_for_metrics`: [] - `batch_eval_metrics`: False - `save_only_model`: False - `save_on_each_node`: False - `enable_jit_checkpoint`: False - `push_to_hub`: False - `hub_private_repo`: None - `hub_model_id`: None - `hub_strategy`: every_save - `hub_always_push`: False - `hub_revision`: None - `load_best_model_at_end`: True - `ignore_data_skip`: False - `restore_callback_states_from_checkpoint`: False - `full_determinism`: False - `seed`: 42 - `data_seed`: None - `use_cpu`: False - `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 - `dataloader_drop_last`: False - `dataloader_num_workers`: 0 - `dataloader_pin_memory`: True - `dataloader_persistent_workers`: False - `dataloader_prefetch_factor`: None - `remove_unused_columns`: True - `label_names`: None - `train_sampling_strategy`: random - `length_column_name`: length - `ddp_find_unused_parameters`: None - `ddp_bucket_cap_mb`: None - `ddp_broadcast_buffers`: False - `ddp_static_graph`: None - `ddp_backend`: None - `ddp_timeout`: 1800 - `fsdp`: [] - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False} - `deepspeed`: None - `debug`: [] - `skip_memory_metrics`: True - `do_predict`: False - `resume_from_checkpoint`: None - `warmup_ratio`: None - `local_rank`: -1 - `prompts`: None - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: proportional - `router_mapping`: {} - `learning_rate_mapping`: {}
### Training Logs
Click to expand | Epoch | Step | Training Loss | gcc-legal-rerank_ndcg@10 | |:-------:|:--------:|:-------------:|:------------------------:| | 0.0031 | 20 | 0.4565 | - | | 0.0063 | 40 | 0.3304 | - | | 0.0094 | 60 | 0.2633 | - | | 0.0126 | 80 | 0.2625 | - | | 0.0157 | 100 | 0.2721 | - | | 0.0189 | 120 | 0.2419 | - | | 0.0220 | 140 | 0.2324 | - | | 0.0252 | 160 | 0.1969 | - | | 0.0283 | 180 | 0.1933 | - | | 0.0315 | 200 | 0.1613 | - | | 0.0346 | 220 | 0.2002 | - | | 0.0378 | 240 | 0.2209 | - | | 0.0409 | 260 | 0.2367 | - | | 0.0441 | 280 | 0.2223 | - | | 0.0472 | 300 | 0.1644 | - | | 0.0504 | 320 | 0.2042 | - | | 0.0535 | 340 | 0.2069 | - | | 0.0567 | 360 | 0.1988 | - | | 0.0598 | 380 | 0.1606 | - | | 0.0630 | 400 | 0.1696 | - | | 0.0661 | 420 | 0.2041 | - | | 0.0693 | 440 | 0.1858 | - | | 0.0724 | 460 | 0.1695 | - | | 0.0756 | 480 | 0.2009 | - | | 0.0787 | 500 | 0.1956 | - | | 0.0819 | 520 | 0.1845 | - | | 0.0850 | 540 | 0.2018 | - | | 0.0882 | 560 | 0.2156 | - | | 0.0913 | 580 | 0.2045 | - | | 0.0945 | 600 | 0.1961 | - | | 0.0976 | 620 | 0.1775 | - | | 0.1008 | 640 | 0.2142 | - | | 0.1039 | 660 | 0.1926 | - | | 0.1071 | 680 | 0.1742 | - | | 0.1102 | 700 | 0.2067 | - | | 0.1134 | 720 | 0.1626 | - | | 0.1165 | 740 | 0.2084 | - | | 0.1197 | 760 | 0.1854 | - | | 0.1228 | 780 | 0.1976 | - | | 0.1260 | 800 | 0.1159 | - | | 0.1291 | 820 | 0.1891 | - | | 0.1323 | 840 | 0.1840 | - | | 0.1354 | 860 | 0.2108 | - | | 0.1386 | 880 | 0.1741 | - | | 0.1417 | 900 | 0.1929 | - | | 0.1449 | 920 | 0.1915 | - | | 0.1480 | 940 | 0.1991 | - | | 0.1512 | 960 | 0.2171 | - | | 0.1543 | 980 | 0.2109 | - | | 0.1575 | 1000 | 0.1667 | - | | 0.1606 | 1020 | 0.1384 | - | | 0.1638 | 1040 | 0.1610 | - | | 0.1669 | 1060 | 0.2053 | - | | 0.1701 | 1080 | 0.2264 | - | | 0.1732 | 1100 | 0.2032 | - | | 0.1764 | 1120 | 0.1733 | - | | 0.1795 | 1140 | 0.1818 | - | | 0.1827 | 1160 | 0.1509 | - | | 0.1858 | 1180 | 0.1386 | - | | 0.1890 | 1200 | 0.1678 | - | | 0.1921 | 1220 | 0.1699 | - | | 0.1953 | 1240 | 0.1695 | - | | 0.1984 | 1260 | 0.1601 | - | | 0.2016 | 1280 | 0.1737 | - | | 0.2047 | 1300 | 0.1807 | - | | 0.2079 | 1320 | 0.1760 | - | | 0.2110 | 1340 | 0.1659 | - | | 0.2142 | 1360 | 0.1727 | - | | 0.2173 | 1380 | 0.1902 | - | | 0.2205 | 1400 | 0.1799 | - | | 0.2236 | 1420 | 0.1620 | - | | 0.2268 | 1440 | 0.1621 | - | | 0.2299 | 1460 | 0.1483 | - | | 0.2331 | 1480 | 0.1481 | - | | 0.2362 | 1500 | 0.1786 | - | | 0.2394 | 1520 | 0.1697 | - | | 0.2425 | 1540 | 0.1703 | - | | 0.2457 | 1560 | 0.1764 | - | | 0.2488 | 1580 | 0.1459 | - | | 0.2520 | 1600 | 0.1763 | - | | 0.2551 | 1620 | 0.1524 | - | | 0.2583 | 1640 | 0.1472 | - | | 0.2614 | 1660 | 0.1623 | - | | 0.2646 | 1680 | 0.1707 | - | | 0.2677 | 1700 | 0.1730 | - | | 0.2709 | 1720 | 0.1895 | - | | 0.2740 | 1740 | 0.1591 | - | | 0.2772 | 1760 | 0.1810 | - | | 0.2803 | 1780 | 0.1536 | - | | 0.2835 | 1800 | 0.1225 | - | | 0.2866 | 1820 | 0.1988 | - | | 0.2898 | 1840 | 0.1402 | - | | 0.2929 | 1860 | 0.1878 | - | | 0.2961 | 1880 | 0.1597 | - | | 0.2992 | 1900 | 0.1571 | - | | 0.3024 | 1920 | 0.1608 | - | | 0.3055 | 1940 | 0.1408 | - | | 0.3087 | 1960 | 0.1329 | - | | 0.3118 | 1980 | 0.1682 | - | | 0.3150 | 2000 | 0.1313 | - | | 0.3181 | 2020 | 0.1694 | - | | 0.3213 | 2040 | 0.1655 | - | | 0.3244 | 2060 | 0.1534 | - | | 0.3276 | 2080 | 0.1600 | - | | 0.3307 | 2100 | 0.1405 | - | | 0.3339 | 2120 | 0.1489 | - | | 0.3370 | 2140 | 0.1426 | - | | 0.3402 | 2160 | 0.2055 | - | | 0.3433 | 2180 | 0.1437 | - | | 0.3465 | 2200 | 0.1509 | - | | 0.3496 | 2220 | 0.1537 | - | | 0.3528 | 2240 | 0.1907 | - | | 0.3559 | 2260 | 0.1669 | - | | 0.3591 | 2280 | 0.1506 | - | | 0.3622 | 2300 | 0.1325 | - | | 0.3654 | 2320 | 0.1645 | - | | 0.3685 | 2340 | 0.1253 | - | | 0.3717 | 2360 | 0.1170 | - | | 0.3748 | 2380 | 0.1573 | - | | 0.3780 | 2400 | 0.1259 | - | | 0.3811 | 2420 | 0.1625 | - | | 0.3843 | 2440 | 0.1278 | - | | 0.3874 | 2460 | 0.1489 | - | | 0.3906 | 2480 | 0.1592 | - | | 0.3937 | 2500 | 0.1447 | - | | 0.3969 | 2520 | 0.1315 | - | | 0.4000 | 2540 | 0.1656 | - | | 0.4032 | 2560 | 0.1658 | - | | 0.4063 | 2580 | 0.1422 | - | | 0.4095 | 2600 | 0.1421 | - | | 0.4126 | 2620 | 0.1907 | - | | 0.4158 | 2640 | 0.1497 | - | | 0.4189 | 2660 | 0.1343 | - | | 0.4221 | 2680 | 0.1956 | - | | 0.4252 | 2700 | 0.1510 | - | | 0.4284 | 2720 | 0.1308 | - | | 0.4315 | 2740 | 0.1409 | - | | 0.4347 | 2760 | 0.1427 | - | | 0.4378 | 2780 | 0.1602 | - | | 0.4410 | 2800 | 0.1355 | - | | 0.4441 | 2820 | 0.1499 | - | | 0.4473 | 2840 | 0.1432 | - | | 0.4504 | 2860 | 0.1421 | - | | 0.4536 | 2880 | 0.1423 | - | | 0.4567 | 2900 | 0.1779 | - | | 0.4599 | 2920 | 0.1582 | - | | 0.4630 | 2940 | 0.1586 | - | | 0.4662 | 2960 | 0.1145 | - | | 0.4693 | 2980 | 0.1503 | - | | 0.4725 | 3000 | 0.1320 | - | | 0.4756 | 3020 | 0.1358 | - | | 0.4788 | 3040 | 0.1371 | - | | 0.4819 | 3060 | 0.1328 | - | | 0.4851 | 3080 | 0.1435 | - | | 0.4882 | 3100 | 0.1371 | - | | 0.4914 | 3120 | 0.1096 | - | | 0.4945 | 3140 | 0.1349 | - | | 0.4977 | 3160 | 0.1270 | - | | 0.5008 | 3180 | 0.1663 | - | | 0.5040 | 3200 | 0.1140 | - | | 0.5071 | 3220 | 0.1390 | - | | 0.5103 | 3240 | 0.1145 | - | | 0.5134 | 3260 | 0.1592 | - | | 0.5166 | 3280 | 0.1049 | - | | 0.5197 | 3300 | 0.1615 | - | | 0.5229 | 3320 | 0.1315 | - | | 0.5260 | 3340 | 0.1262 | - | | 0.5292 | 3360 | 0.1462 | - | | 0.5323 | 3380 | 0.1034 | - | | 0.5355 | 3400 | 0.1639 | - | | 0.5386 | 3420 | 0.1277 | - | | 0.5418 | 3440 | 0.1179 | - | | 0.5449 | 3460 | 0.1321 | - | | 0.5481 | 3480 | 0.1610 | - | | 0.5512 | 3500 | 0.1545 | - | | 0.5544 | 3520 | 0.1487 | - | | 0.5575 | 3540 | 0.1371 | - | | 0.5607 | 3560 | 0.1446 | - | | 0.5638 | 3580 | 0.1325 | - | | 0.5670 | 3600 | 0.1593 | - | | 0.5701 | 3620 | 0.1230 | - | | 0.5733 | 3640 | 0.1314 | - | | 0.5764 | 3660 | 0.1336 | - | | 0.5796 | 3680 | 0.1389 | - | | 0.5827 | 3700 | 0.1188 | - | | 0.5859 | 3720 | 0.1237 | - | | 0.5890 | 3740 | 0.1151 | - | | 0.5922 | 3760 | 0.1272 | - | | 0.5953 | 3780 | 0.1489 | - | | 0.5985 | 3800 | 0.1327 | - | | 0.6016 | 3820 | 0.1408 | - | | 0.6048 | 3840 | 0.1699 | - | | 0.6079 | 3860 | 0.1371 | - | | 0.6111 | 3880 | 0.1382 | - | | 0.6142 | 3900 | 0.1293 | - | | 0.6174 | 3920 | 0.1329 | - | | 0.6205 | 3940 | 0.1503 | - | | 0.6237 | 3960 | 0.1373 | - | | 0.6268 | 3980 | 0.1586 | - | | 0.6300 | 4000 | 0.1228 | - | | 0.6331 | 4020 | 0.1285 | - | | 0.6363 | 4040 | 0.1237 | - | | 0.6394 | 4060 | 0.1093 | - | | 0.6426 | 4080 | 0.1509 | - | | 0.6457 | 4100 | 0.1421 | - | | 0.6489 | 4120 | 0.1426 | - | | 0.6520 | 4140 | 0.1637 | - | | 0.6552 | 4160 | 0.1261 | - | | 0.6583 | 4180 | 0.1609 | - | | 0.6615 | 4200 | 0.1258 | - | | 0.6646 | 4220 | 0.1477 | - | | 0.6678 | 4240 | 0.0975 | - | | 0.6709 | 4260 | 0.1177 | - | | 0.6741 | 4280 | 0.1370 | - | | 0.6772 | 4300 | 0.1410 | - | | 0.6804 | 4320 | 0.1242 | - | | 0.6835 | 4340 | 0.1418 | - | | 0.6867 | 4360 | 0.1978 | - | | 0.6898 | 4380 | 0.1222 | - | | 0.6930 | 4400 | 0.1157 | - | | 0.6961 | 4420 | 0.1316 | - | | 0.6993 | 4440 | 0.1207 | - | | 0.7024 | 4460 | 0.1447 | - | | 0.7056 | 4480 | 0.1316 | - | | 0.7087 | 4500 | 0.1123 | - | | 0.7119 | 4520 | 0.1309 | - | | 0.7150 | 4540 | 0.1273 | - | | 0.7182 | 4560 | 0.1642 | - | | 0.7213 | 4580 | 0.1310 | - | | 0.7245 | 4600 | 0.1044 | - | | 0.7276 | 4620 | 0.1508 | - | | 0.7308 | 4640 | 0.1108 | - | | 0.7339 | 4660 | 0.1226 | - | | 0.7371 | 4680 | 0.1376 | - | | 0.7402 | 4700 | 0.1540 | - | | 0.7434 | 4720 | 0.1209 | - | | 0.7465 | 4740 | 0.1204 | - | | 0.7497 | 4760 | 0.1521 | - | | 0.7528 | 4780 | 0.0823 | - | | 0.7560 | 4800 | 0.1431 | - | | 0.7591 | 4820 | 0.1456 | - | | 0.7623 | 4840 | 0.0997 | - | | 0.7654 | 4860 | 0.1436 | - | | 0.7686 | 4880 | 0.1012 | - | | 0.7717 | 4900 | 0.1412 | - | | 0.7749 | 4920 | 0.1573 | - | | 0.7780 | 4940 | 0.1566 | - | | 0.7812 | 4960 | 0.1421 | - | | 0.7843 | 4980 | 0.1220 | - | | 0.7875 | 5000 | 0.1079 | - | | 0.7906 | 5020 | 0.1183 | - | | 0.7938 | 5040 | 0.0987 | - | | 0.7969 | 5060 | 0.1454 | - | | 0.8001 | 5080 | 0.1269 | - | | 0.8032 | 5100 | 0.1343 | - | | 0.8064 | 5120 | 0.1293 | - | | 0.8095 | 5140 | 0.1129 | - | | 0.8127 | 5160 | 0.1360 | - | | 0.8158 | 5180 | 0.1060 | - | | 0.8190 | 5200 | 0.1450 | - | | 0.8221 | 5220 | 0.1606 | - | | 0.8253 | 5240 | 0.1316 | - | | 0.8284 | 5260 | 0.1259 | - | | 0.8316 | 5280 | 0.1156 | - | | 0.8347 | 5300 | 0.1125 | - | | 0.8379 | 5320 | 0.0973 | - | | 0.8410 | 5340 | 0.1505 | - | | 0.8442 | 5360 | 0.1178 | - | | 0.8473 | 5380 | 0.0929 | - | | 0.8505 | 5400 | 0.1707 | - | | 0.8536 | 5420 | 0.1444 | - | | 0.8568 | 5440 | 0.1195 | - | | 0.8599 | 5460 | 0.1327 | - | | 0.8631 | 5480 | 0.0956 | - | | 0.8662 | 5500 | 0.1221 | - | | 0.8694 | 5520 | 0.1629 | - | | 0.8725 | 5540 | 0.1181 | - | | 0.8757 | 5560 | 0.1498 | - | | 0.8788 | 5580 | 0.1133 | - | | 0.8820 | 5600 | 0.1348 | - | | 0.8851 | 5620 | 0.1231 | - | | 0.8883 | 5640 | 0.0901 | - | | 0.8914 | 5660 | 0.0875 | - | | 0.8946 | 5680 | 0.1009 | - | | 0.8977 | 5700 | 0.1136 | - | | 0.9009 | 5720 | 0.1168 | - | | 0.9040 | 5740 | 0.1205 | - | | 0.9072 | 5760 | 0.1188 | - | | 0.9103 | 5780 | 0.1330 | - | | 0.9135 | 5800 | 0.1227 | - | | 0.9166 | 5820 | 0.1168 | - | | 0.9198 | 5840 | 0.1027 | - | | 0.9229 | 5860 | 0.0930 | - | | 0.9261 | 5880 | 0.1239 | - | | 0.9292 | 5900 | 0.1457 | - | | 0.9324 | 5920 | 0.1709 | - | | 0.9355 | 5940 | 0.1242 | - | | 0.9387 | 5960 | 0.1192 | - | | 0.9418 | 5980 | 0.1350 | - | | 0.9450 | 6000 | 0.1137 | - | | 0.9481 | 6020 | 0.1329 | - | | 0.9513 | 6040 | 0.1393 | - | | 0.9544 | 6060 | 0.1410 | - | | 0.9576 | 6080 | 0.1082 | - | | 0.9607 | 6100 | 0.1113 | - | | 0.9639 | 6120 | 0.1269 | - | | 0.9670 | 6140 | 0.0969 | - | | 0.9702 | 6160 | 0.0867 | - | | 0.9733 | 6180 | 0.0949 | - | | 0.9765 | 6200 | 0.1233 | - | | 0.9796 | 6220 | 0.1041 | - | | 0.9828 | 6240 | 0.1106 | - | | 0.9859 | 6260 | 0.0891 | - | | 0.9891 | 6280 | 0.1222 | - | | 0.9922 | 6300 | 0.1121 | - | | 0.9954 | 6320 | 0.1596 | - | | 0.9985 | 6340 | 0.1099 | - | | **1.0** | **6350** | **-** | **0.9915** | * The bold row denotes the saved checkpoint.
### Training Time - **Training**: 11.4 hours - **Evaluation**: 11.5 minutes - **Total**: 11.6 hours ### Framework Versions - Python: 3.12.3 - Sentence Transformers: 5.4.1 - Transformers: 5.6.2 - PyTorch: 2.12.0.dev20260407+cu128 - Accelerate: 1.13.0 - Datasets: 4.8.4 - Tokenizers: 0.22.2 ## 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", } ```