--- language: - multilingual license: cc-by-nc-4.0 tags: - sentence-transformers - cross-encoder - reranker - generated_from_trainer - dataset_size:13717 - loss:BinaryCrossEntropyLoss base_model: jinaai/jina-reranker-v2-base-multilingual pipeline_tag: text-ranking library_name: sentence-transformers metrics: - map - mrr@10 - ndcg@10 model-index: - name: cometadata/jina-reranker-v2-multilingual-affiliations results: - task: type: cross-encoder-reranking name: Cross Encoder Reranking dataset: name: affiliation val type: affiliation-val metrics: - type: map value: 0.9294 name: Map - type: mrr@10 value: 0.9294 name: Mrr@10 - type: ndcg@10 value: 0.9564 name: Ndcg@10 --- # cometadata/jina-reranker-v2-multilingual-affiliations This is a [Cross Encoder](https://www.sbert.net/docs/cross_encoder/usage/usage.html) model finetuned from [jinaai/jina-reranker-v2-base-multilingual](https://huggingface.co/jinaai/jina-reranker-v2-base-multilingual) 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:** [jinaai/jina-reranker-v2-base-multilingual](https://huggingface.co/jinaai/jina-reranker-v2-base-multilingual) - **Maximum Sequence Length:** 1024 tokens - **Number of Output Labels:** 1 label - **Language:** multilingual - **License:** cc-by-nc-4.0 ### 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) ## 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("cometadata/jina-reranker-v2-multilingual-affiliations") # Get scores for pairs of texts pairs = [ ["Centre sur le handicap et l'intégration, School of Economics and Political Science Université de Saint‐Gall", "College of Saint Benedict and Saint John's University, Collegeville, MN, United States"], ['Swiss Federal Institute of Technology (ETH) Zurich, Institute of Quantum Electronics, Laser Spectroscopy and Sensing Laboratory, Hoenggerberg, HPF D19, CH-8093\u2009Zurich, Switzerland', 'Laboratory of Crystallography, ETH Zurich, CH-8093 Zurich, Switzerland'], ['Swiss Federal Institute of Technology (ETH) Zurich, Institute of Quantum Electronics, Laser Spectroscopy and Sensing Laboratory, Hoenggerberg, HPF D19, CH-8093\u2009Zurich, Switzerland', "Laboratoire d'Electrochimie Physique et Analytique, École Polytechnique Fédérale de Lausanne Station 6, CH-1015 Lausanne, Switzerland"], ['Institute for Advanced Study, Technische Universität München 2 , Lichtenbergstr. 2a, D-85748 Garching, Germany', 'Department of Surgery, Technical University of Munich, School of Medicine, Munich, Germany'], ['Institute for Advanced Study, Technische Universität München 2 , Lichtenbergstr. 2a, D-85748 Garching, Germany', 'Lehrstuhl für BioMolekulare Optik, Ludwig-Maximilians-Universität München, Oettingenstrasse 67, 80538 München (Germany)'], ] scores = model.predict(pairs) print(scores.shape) # (5,) # Or rank different texts based on similarity to a single text ranks = model.rank( "Centre sur le handicap et l'intégration, School of Economics and Political Science Université de Saint‐Gall", [ "College of Saint Benedict and Saint John's University, Collegeville, MN, United States", 'Laboratory of Crystallography, ETH Zurich, CH-8093 Zurich, Switzerland', "Laboratoire d'Electrochimie Physique et Analytique, École Polytechnique Fédérale de Lausanne Station 6, CH-1015 Lausanne, Switzerland", 'Department of Surgery, Technical University of Munich, School of Medicine, Munich, Germany', 'Lehrstuhl für BioMolekulare Optik, Ludwig-Maximilians-Universität München, Oettingenstrasse 67, 80538 München (Germany)', ] ) # [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...] ``` ## Evaluation ### Metrics #### Cross Encoder Reranking * Dataset: `affiliation-val` * 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, "always_rerank_positives": true } ``` | Metric | Value | |:------------|:---------------------| | map | 0.9294 (-0.0706) | | mrr@10 | 0.9294 (-0.0706) | | **ndcg@10** | **0.9564 (-0.0436)** | ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 13,717 training samples * Columns: query, document, and label * Approximate statistics based on the first 1000 samples: | | query | document | label | |:--------|:-----------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------|:------------------------------------------------| | type | string | string | int | | details | | | | * Samples: | query | document | label | |:-----------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------|:---------------| | Department of Otolaryngology-Head and Neck Surgery; National Defense Medical College; Saitama Japan | . Department of Otolaryngology-Head and Neck Surgery, National Defense Medical College, Japan. | 1 | | Department of Otolaryngology-Head and Neck Surgery; National Defense Medical College; Saitama Japan | EOG Resources, Inc | 0 | | School of Science and Engineering The Chinese University of Hong Kong,Shenzhen,China | School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, China, | 1 | * 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 } ``` ### Evaluation Dataset #### Unnamed Dataset * Size: 2,421 evaluation samples * Columns: query, document, and label * Approximate statistics based on the first 1000 samples: | | query | document | label | |:--------|:-------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------|:------------------------------------------------| | type | string | string | int | | details | | | | * Samples: | query | document | label | |:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------|:---------------| | Centre sur le handicap et l'intégration, School of Economics and Political Science Université de Saint‐Gall | College of Saint Benedict and Saint John's University, Collegeville, MN, United States | 0 | | Swiss Federal Institute of Technology (ETH) Zurich, Institute of Quantum Electronics, Laser Spectroscopy and Sensing Laboratory, Hoenggerberg, HPF D19, CH-8093 Zurich, Switzerland | Laboratory of Crystallography, ETH Zurich, CH-8093 Zurich, Switzerland | 1 | | Swiss Federal Institute of Technology (ETH) Zurich, Institute of Quantum Electronics, Laser Spectroscopy and Sensing Laboratory, Hoenggerberg, HPF D19, CH-8093 Zurich, Switzerland | Laboratoire d'Electrochimie Physique et Analytique, École Polytechnique Fédérale de Lausanne Station 6, CH-1015 Lausanne, Switzerland | 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 - `eval_strategy`: steps - `per_device_train_batch_size`: 16 - `per_device_eval_batch_size`: 16 - `learning_rate`: 2e-05 - `warmup_ratio`: 0.1 - `bf16`: True - `load_best_model_at_end`: True - `push_to_hub`: True - `hub_model_id`: cometadata/jina-reranker-v2-multilingual-affiliations #### All Hyperparameters
Click to expand - `overwrite_output_dir`: False - `do_predict`: False - `eval_strategy`: steps - `prediction_loss_only`: True - `per_device_train_batch_size`: 16 - `per_device_eval_batch_size`: 16 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 1 - `eval_accumulation_steps`: None - `torch_empty_cache_steps`: None - `learning_rate`: 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`: 3 - `max_steps`: -1 - `lr_scheduler_type`: linear - `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 - `bf16`: True - `fp16`: False - `fp16_opt_level`: O1 - `half_precision_backend`: auto - `bf16_full_eval`: False - `fp16_full_eval`: False - `tf32`: None - `local_rank`: 0 - `ddp_backend`: None - `tpu_num_cores`: None - `tpu_metrics_debug`: False - `debug`: [] - `dataloader_drop_last`: False - `dataloader_num_workers`: 0 - `dataloader_prefetch_factor`: None - `past_index`: -1 - `disable_tqdm`: False - `remove_unused_columns`: True - `label_names`: None - `load_best_model_at_end`: 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} - `parallelism_config`: None - `deepspeed`: None - `label_smoothing_factor`: 0.0 - `optim`: adamw_torch_fused - `optim_args`: None - `adafactor`: False - `group_by_length`: False - `length_column_name`: length - `project`: huggingface - `trackio_space_id`: trackio - `ddp_find_unused_parameters`: None - `ddp_bucket_cap_mb`: None - `ddp_broadcast_buffers`: False - `dataloader_pin_memory`: True - `dataloader_persistent_workers`: False - `skip_memory_metrics`: True - `use_legacy_prediction_loop`: False - `push_to_hub`: True - `resume_from_checkpoint`: None - `hub_model_id`: cometadata/jina-reranker-v2-multilingual-affiliations - `hub_strategy`: every_save - `hub_private_repo`: None - `hub_always_push`: False - `hub_revision`: None - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `include_for_metrics`: [] - `eval_do_concat_batches`: True - `fp16_backend`: auto - `push_to_hub_model_id`: None - `push_to_hub_organization`: None - `mp_parameters`: - `auto_find_batch_size`: False - `full_determinism`: False - `torchdynamo`: None - `ray_scope`: last - `ddp_timeout`: 1800 - `torch_compile`: False - `torch_compile_backend`: None - `torch_compile_mode`: None - `include_tokens_per_second`: False - `include_num_input_tokens_seen`: no - `neftune_noise_alpha`: None - `optim_target_modules`: None - `batch_eval_metrics`: False - `eval_on_start`: False - `use_liger_kernel`: False - `liger_kernel_config`: None - `eval_use_gather_object`: False - `average_tokens_across_devices`: True - `prompts`: None - `multi_dataset_batch_sampler`: proportional - `router_mapping`: {} - `learning_rate_mapping`: {}
### Training Logs | Epoch | Step | Training Loss | Validation Loss | affiliation-val_ndcg@10 | |:----------:|:--------:|:-------------:|:---------------:|:-----------------------:| | -1 | -1 | - | - | 0.8997 (-0.1003) | | 0.0012 | 1 | 0.0941 | - | - | | 0.1166 | 100 | 0.3775 | - | - | | 0.2331 | 200 | 0.2667 | - | - | | 0.3497 | 300 | 0.2155 | - | - | | 0.4662 | 400 | 0.212 | - | - | | 0.5828 | 500 | 0.2277 | 0.6306 | 0.9465 (-0.0535) | | 0.6993 | 600 | 0.2825 | - | - | | 0.8159 | 700 | 0.2932 | - | - | | 0.9324 | 800 | 0.3123 | - | - | | 1.0490 | 900 | 0.2608 | - | - | | 1.1655 | 1000 | 0.0833 | 0.5776 | 0.9543 (-0.0457) | | 1.2821 | 1100 | 0.0938 | - | - | | 1.3986 | 1200 | 0.1492 | - | - | | 1.5152 | 1300 | 0.1651 | - | - | | 1.6317 | 1400 | 0.1842 | - | - | | 1.7483 | 1500 | 0.2407 | 0.5891 | 0.9555 (-0.0445) | | 1.8648 | 1600 | 0.288 | - | - | | 1.9814 | 1700 | 0.3352 | - | - | | 2.0979 | 1800 | 0.1082 | - | - | | 2.2145 | 1900 | 0.0758 | - | - | | 2.3310 | 2000 | 0.1072 | 0.5725 | 0.9563 (-0.0437) | | 2.4476 | 2100 | 0.1437 | - | - | | 2.5641 | 2200 | 0.153 | - | - | | 2.6807 | 2300 | 0.2176 | - | - | | 2.7972 | 2400 | 0.2513 | - | - | | **2.9138** | **2500** | **0.2949** | **0.5721** | **0.9564 (-0.0436)** | | -1 | -1 | - | - | 0.9564 (-0.0436) | * The bold row denotes the saved checkpoint. ### Framework Versions - Python: 3.12.12 - Sentence Transformers: 5.2.0 - Transformers: 4.57.3 - PyTorch: 2.9.1+cu128 - Accelerate: 1.12.0 - Datasets: 4.4.2 - Tokenizers: 0.22.1 ## 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", } ```