chingackgook/reranker-ModernBERT-base-gooaq-bce

ModernBERT-base trained on GooAQ

This is a Cross Encoder model finetuned from answerdotai/ModernBERT-base using the sentence-transformers 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: answerdotai/ModernBERT-base
• Maximum Sequence Length: 8192 tokens
• Number of Output Labels: 1 label
• Language: en
• License: apache-2.0

Model Sources

• Documentation: Sentence Transformers Documentation
• Documentation: Cross Encoder Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Cross Encoders on Hugging Face

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 CrossEncoder

# Download from the 🤗 Hub
model = CrossEncoder("chingackgook/reranker-ModernBERT-base-gooaq-bce")
# Get scores for pairs of texts
pairs = [
    ['how many days can i drive my car without mot?', "If your car fails its MOT you can only continue to drive it if the previous year's MOT is still valid - which might occur if you submitted the car for its test two weeks early. You can still drive it away from the testing centre or garage if no 'dangerous' problems were identified during the MOT."],
    ['how many days can i drive my car without mot?', '["Open File Explorer\', go to \'This PC\'. Check the status of the network drive. [00:11]", \'An error occurred while reconnecting or mapping Drive letter Z: to Network folder. [00:36]\', \'You can do either connect the drive (Map network drive) or Disconnect Network drive. [01:01]\', \'Disconnect Network Drive by this way. [01:28]\']'],
    ['what xbox 360 games are compatible with xbox 1?', "['0 day Attack on Earth.', '3D Ultra Minigolf.', 'A Kingdom for Keflings.', 'A World of Keflings.', 'Ace Combat 6: Fires of Liberation.', 'Aegis Wing.', 'Age of Booty.', 'Alan Wake (Tested by Digital Foundry)']"],
    ['what xbox 360 games are compatible with xbox 1?', "['1) Computer and Information Systems Managers.', '2) Computer and Information Research Scientists.', '3) Computer Network Architects.', '4) Software Development Engineer.', '5) Software Developers.', '6) Information Security Analysts.', '8) Computer Systems Analysts.']"],
    ['what does it mean when a guy asks for a picture of you?', 'He wants to confirm if he is talking to Priya or Angel Priya (I.e., if he is really talking to a girl or just a guy with fake profile) They are talking to you and want to see how you look. I found it normal but would say, be careful about whom do you share your picture with as they might misuse it. I hate this one.'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'how many days can i drive my car without mot?',
    [
        "If your car fails its MOT you can only continue to drive it if the previous year's MOT is still valid - which might occur if you submitted the car for its test two weeks early. You can still drive it away from the testing centre or garage if no 'dangerous' problems were identified during the MOT.",
        '["Open File Explorer\', go to \'This PC\'. Check the status of the network drive. [00:11]", \'An error occurred while reconnecting or mapping Drive letter Z: to Network folder. [00:36]\', \'You can do either connect the drive (Map network drive) or Disconnect Network drive. [01:01]\', \'Disconnect Network Drive by this way. [01:28]\']',
        "['0 day Attack on Earth.', '3D Ultra Minigolf.', 'A Kingdom for Keflings.', 'A World of Keflings.', 'Ace Combat 6: Fires of Liberation.', 'Aegis Wing.', 'Age of Booty.', 'Alan Wake (Tested by Digital Foundry)']",
        "['1) Computer and Information Systems Managers.', '2) Computer and Information Research Scientists.', '3) Computer Network Architects.', '4) Software Development Engineer.', '5) Software Developers.', '6) Information Security Analysts.', '8) Computer Systems Analysts.']",
        'He wants to confirm if he is talking to Priya or Angel Priya (I.e., if he is really talking to a girl or just a guy with fake profile) They are talking to you and want to see how you look. I found it normal but would say, be careful about whom do you share your picture with as they might misuse it. I hate this one.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Evaluation

Metrics

Cross Encoder Reranking

• Dataset: gooaq-dev
• Evaluated with CrossEncoderRerankingEvaluator with these parameters:

  {
      "at_k": 10,
      "always_rerank_positives": false
  }
  

Metric	Value
map	0.2500 (-0.7500)
mrr@10	0.2500 (-0.7500)
ndcg@10	0.4307 (-0.5693)

Cross Encoder Reranking

• Datasets: NanoMSMARCO_R100, NanoNFCorpus_R100 and NanoNQ_R100
• Evaluated with CrossEncoderRerankingEvaluator with these parameters:

  {
      "at_k": 10,
      "always_rerank_positives": true
  }
  

Metric	NanoMSMARCO_R100	NanoNFCorpus_R100	NanoNQ_R100
map	0.0358 (-0.4538)	0.2903 (+0.0293)	0.0379 (-0.3817)
mrr@10	0.0109 (-0.4666)	0.5069 (+0.0070)	0.0140 (-0.4127)
ndcg@10	0.0260 (-0.5145)	0.2900 (-0.0350)	0.0174 (-0.4833)

Cross Encoder Nano BEIR

• Dataset: NanoBEIR_R100_mean
• Evaluated with CrossEncoderNanoBEIREvaluator with these parameters:

  {
      "dataset_names": [
          "msmarco",
          "nfcorpus",
          "nq"
      ],
      "rerank_k": 100,
      "at_k": 10,
      "always_rerank_positives": true
  }
  

Metric	Value
map	0.1213 (-0.2688)
mrr@10	0.1772 (-0.2908)
ndcg@10	0.1111 (-0.3443)

Training Details

Training Dataset

Unnamed Dataset

• Size: 198 training samples
• Columns: question, answer, and label
• Approximate statistics based on the first 198 samples:

  	question	answer	label
  type	string	string	int
  details	min: 21 characters mean: 42.74 characters max: 73 characters	min: 55 characters mean: 257.56 characters max: 378 characters	0: ~50.00% 1: ~50.00%

• Samples:

  question	answer	label
  how many days can i drive my car without mot?	If your car fails its MOT you can only continue to drive it if the previous year's MOT is still valid - which might occur if you submitted the car for its test two weeks early. You can still drive it away from the testing centre or garage if no 'dangerous' problems were identified during the MOT.	1
  how many days can i drive my car without mot?	["Open File Explorer', go to 'This PC'. Check the status of the network drive. [00:11]", 'An error occurred while reconnecting or mapping Drive letter Z: to Network folder. [00:36]', 'You can do either connect the drive (Map network drive) or Disconnect Network drive. [01:01]', 'Disconnect Network Drive by this way. [01:28]']	0
  what xbox 360 games are compatible with xbox 1?	['0 day Attack on Earth.', '3D Ultra Minigolf.', 'A Kingdom for Keflings.', 'A World of Keflings.', 'Ace Combat 6: Fires of Liberation.', 'Aegis Wing.', 'Age of Booty.', 'Alan Wake (Tested by Digital Foundry)']	1

• Loss: BinaryCrossEntropyLoss with these parameters:

  {
      "activation_fn": "torch.nn.modules.linear.Identity",
      "pos_weight": 1
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• learning_rate: 2e-05
• num_train_epochs: 1
• warmup_ratio: 0.1
• seed: 12
• bf16: True
• dataloader_num_workers: 4
• load_best_model_at_end: True

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: 1
• 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: 12
• 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: 4
• 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: False
• resume_from_checkpoint: None
• hub_model_id: None
• 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
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	gooaq-dev_ndcg@10	NanoMSMARCO_R100_ndcg@10	NanoNFCorpus_R100_ndcg@10	NanoNQ_R100_ndcg@10	NanoBEIR_R100_mean_ndcg@10
-1	-1	-	0.4307 (-0.5693)	0.0220 (-0.5184)	0.2802 (-0.0448)	0.0200 (-0.4806)	0.1074 (-0.3479)
0.25	1	0.7171	-	-	-	-	-
-1	-1	-	0.4307 (-0.5693)	0.0260 (-0.5145)	0.2900 (-0.0350)	0.0174 (-0.4833)	0.1111 (-0.3443)

Framework Versions

• Python: 3.10.19
• Sentence Transformers: 5.2.0.dev0
• Transformers: 4.57.3
• PyTorch: 2.9.1+cu128
• Accelerate: 1.12.0
• Datasets: 4.4.1
• Tokenizers: 0.22.1

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",
}