bansalaman18/reranker-msmarco-v1.1-ettin-encoder-32m-listnet

CrossEncoder based on jhu-clsp/ettin-encoder-32m

This is a Cross Encoder model finetuned from jhu-clsp/ettin-encoder-32m on the ms_marco dataset 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: jhu-clsp/ettin-encoder-32m
• Maximum Sequence Length: 7999 tokens
• Number of Output Labels: 1 label
• Training Dataset:
  • ms_marco
• Language: en

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("bansalaman18/reranker-msmarco-v1.1-ettin-encoder-32m-listnet")
# Get scores for pairs of texts
pairs = [
    ['how to get rid of dark circles around mouth', '2. You can apply lemon juice to get rid of darkness around mouth due to melasma. Lemon juice is a natural bleaching agent. Massaging the dark around mouth with the lemon juice is the best home remedy for lightening the dark skin around mouth.'],
    ['how to get rid of dark circles around mouth', 'Apply it on your discolored skin around mouth for 15 minutes. This home remedy is best for removing the tan and making the mouth area fair and blemish-free. 9. A combination of curd, lemon juice, and honey can also prove to be beneficial for lightening the dark pigmentation around mouth and corners.'],
    ['how to get rid of dark circles around mouth', 'Apply this skin lightening pack over the dark spots and patches on the upper lip area and leave it on for 20 minutes. Rinse with warm water. Rubbing off this mask carefully when it is completely dry will also help to get rid of dead skin cells that make skin around lips appear dry, dark, and wrinkled.'],
    ['how to get rid of dark circles around mouth', '3. Know that the skin around your mouth is thin. This can lead to discoloration, dry skin, and mouth wrinkles. These problems do not go deep into the skin, so you will probably not need an invasive treatment. You may be able to easily get rid of your discoloration by treating or exfoliating the skin.'],
    ['how to get rid of dark circles around mouth', 'Step 1. Exfoliate daily with a mild and gentle facial scrub to lift dead skin cells and help fade dark areas around the mouth. Apply a pea-size amount of the exfoliant to a damp washcloth. Gently rub the product on your face to remove pigmented skin cells and cleanse the skin.'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'how to get rid of dark circles around mouth',
    [
        '2. You can apply lemon juice to get rid of darkness around mouth due to melasma. Lemon juice is a natural bleaching agent. Massaging the dark around mouth with the lemon juice is the best home remedy for lightening the dark skin around mouth.',
        'Apply it on your discolored skin around mouth for 15 minutes. This home remedy is best for removing the tan and making the mouth area fair and blemish-free. 9. A combination of curd, lemon juice, and honey can also prove to be beneficial for lightening the dark pigmentation around mouth and corners.',
        'Apply this skin lightening pack over the dark spots and patches on the upper lip area and leave it on for 20 minutes. Rinse with warm water. Rubbing off this mask carefully when it is completely dry will also help to get rid of dead skin cells that make skin around lips appear dry, dark, and wrinkled.',
        '3. Know that the skin around your mouth is thin. This can lead to discoloration, dry skin, and mouth wrinkles. These problems do not go deep into the skin, so you will probably not need an invasive treatment. You may be able to easily get rid of your discoloration by treating or exfoliating the skin.',
        'Step 1. Exfoliate daily with a mild and gentle facial scrub to lift dead skin cells and help fade dark areas around the mouth. Apply a pea-size amount of the exfoliant to a damp washcloth. Gently rub the product on your face to remove pigmented skin cells and cleanse the skin.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Training Details

Training Dataset

ms_marco

• Dataset: ms_marco at a47ee7a
• Size: 78,704 training samples
• Columns: query, docs, and labels
• Approximate statistics based on the first 1000 samples:

  	query	docs	labels
  type	string	list	list
  details	min: 12 characters mean: 33.84 characters max: 110 characters	min: 4 elements mean: 7.00 elements max: 10 elements	min: 4 elements mean: 7.00 elements max: 10 elements

• Samples:

  query	docs	labels
  what is the molecular weight of sucrose	['For sucrose (table sugar) we have the molecular formula C12H22O11 and the molar mass would be Molar mass of sucrose = 12x12.011 g + 22x31.008 g + 11x15.999g = 342.3 g Or, we would say that the molar mass of sucrose = 342.3 g/mol (i.e., a mole of sucrose molecules has a mass of 342.3 g). Molecular compounds consist of a large collection of molecules. The molar mass (or molecular weight) of a molecular compound is defined in the following way: Molar Mass = Mass of 1 mole of molecules Therefore, to get the molar mass of a molecular compound we need simply determine the molar mass of molecules. For sucrose (table sugar) we have the molecular formula C12H22O11 and the molar mass would be', 'The sucrose molecule (MW = 342) is almost twice as heavy as the glucose molecule (MW = 180). So a 35% solution of glucose would contain almost twice as many molecules as a 35% solution of sucrose. To correct the problem, you should make the solution with the weights of sucrose and glucose in a ratio of...	[1, 1, 0, 0, 0, ...]
  b&m bathgate contact number	['What is the number for B&M in bathgate west lothian. B&M Stores, Unit 2 Bathgate Retail Park, 24D Whitburn Rd, Bathgate, EH48 1HH. Contact details not listed. Please contact T: 01517285400 for more info. ', 'Please note, The-Shops.co.uk is a participative site where each person can indicate opening hours. Should you find any errors, please let us know. Usually the store is open Sunday. These times do not include public holidays. To see if your store is open these days, contact them. ', "B & M in Bathgate. Discover big name brands at amazing prices at the B&M store in Bathgate. See below to find the store's location and opening and closing hours. With over 400 stores nationwide, B&M Stores is now one of the UK's leading multi-brand retailers. Visit the Bathgate B&M store to find fantastic savings on household goods, food and drink, toys, home and garden furniture and much, much more.", 'You may be travelling, sightseeing, or simply passing through – it doesn’t matter – we look forward...	[1, 0, 0, 0, 0, ...]
  what is heart inflammation	['Heart inflammation: Introduction. Heart inflammation: Where there is inflammation of the muscle or lining of the heart. See detailed information below for a list of 5 causes of Heart inflammation, Symptom Checker, including diseases and drug side effect causes. Some of the comorbid or associated medical symptoms for Heart inflammation may include these symptoms: 1 Abdominal symptoms. 2 Cardiovascular symptoms. 3 Eczema. 4 Fever. 5 Heart symptoms. 6 Inflammatory symptoms. 7 Mouth symptoms.', 'Myocarditis is a disease marked by the inflammation of heart muscle. Learn about the symptoms, diagnosis, and treatment of myocarditis. Inflammation is a normal bodily response to any sort of wound or infection. Imagine when you cut your finger. ', 'Heart muscle disease or inflammation of the heart. Increase text size / Decrease text size	Print this page

• Loss: ListNetLoss with these parameters:

  {
      "activation_fn": "torch.nn.modules.linear.Identity",
      "mini_batch_size": 16
  }
  

Evaluation Dataset

ms_marco

• Dataset: ms_marco at a47ee7a
• Size: 1,000 evaluation samples
• Columns: query, docs, and labels
• Approximate statistics based on the first 1000 samples:

  	query	docs	labels
  type	string	list	list
  details	min: 8 characters mean: 33.7 characters max: 99 characters	min: 3 elements mean: 6.50 elements max: 10 elements	min: 3 elements mean: 6.50 elements max: 10 elements

• Samples:

  query	docs	labels
  how to get rid of dark circles around mouth	['2. You can apply lemon juice to get rid of darkness around mouth due to melasma. Lemon juice is a natural bleaching agent. Massaging the dark around mouth with the lemon juice is the best home remedy for lightening the dark skin around mouth.', 'Apply it on your discolored skin around mouth for 15 minutes. This home remedy is best for removing the tan and making the mouth area fair and blemish-free. 9. A combination of curd, lemon juice, and honey can also prove to be beneficial for lightening the dark pigmentation around mouth and corners.', 'Apply this skin lightening pack over the dark spots and patches on the upper lip area and leave it on for 20 minutes. Rinse with warm water. Rubbing off this mask carefully when it is completely dry will also help to get rid of dead skin cells that make skin around lips appear dry, dark, and wrinkled.', '3. Know that the skin around your mouth is thin. This can lead to discoloration, dry skin, and mouth wrinkles. These problems do not go deep i...	[1, 0, 0, 0, 0, ...]
  is it legal to sleep in your car nj	["Trespassing/Parking. For the most part, spending an extended period of time in your car isn't a problem -- you can sleep in your parked car in your driveway if you'd like. Rather, it's the fact that your car is in one spot for at least as long as you're sleeping in it. Parking your car on someone's private property can get you arrested for trespassing, unless you have the owner's consent. And if you do, you might as well as to just crash on a couch.", 'Even in areas where sleeping in the car is legal, police have the right to search you and your vehicle if they believe you are exhibiting suspicious behavior. Police are on the lookout for drug dealers, drug users, and other felonious activity.', 'Best Answer: yes its legal, it may be frowned upon sleeping in your car infront of wal-mart but its not illegal by any stretch.', "Is it legal for my dogs and I to sleep in my... show more I'm planning a short-ish road trip with my two German Shepherds, and I think it'd be a waste to pay for ...	[1, 0, 0, 0, 0, ...]
  how to insert calendar in excel cell	['To add a Calendar Control to Excel 2003, begin by clicking on a cell in the area of the spreadsheet where you want to make a calendar. Go to the Insert menu and click on Object. On the Create New tab of the Object dialog box, click Calendar Control and then click OK. In Excel 2007, go to the Developer tab of the ribbon and click Insert. Select ActiveX Controls and then More Controls. Select Calendar Control and click OK', "If you want to insert such a calendar into your spreadsheet, you can do so by enabling Developer Mode from the program's main menu. Step 1. Click the Office button and select the Excel Options option from the drop-down menu that appears. Step 2. Click the Popular tab in the Options menu and place a check mark next to the Show Developer Tab in the Ribbon option. Click OK to save the changes. Step 5. Click on the upper left corner of the cell to which you want to add the calendar and drag your mouse pointer to the cell's bottom-right corner. This will draw a rectangl...	[1, 0, 0, 0, 0, ...]

• Loss: ListNetLoss with these parameters:

  {
      "activation_fn": "torch.nn.modules.linear.Identity",
      "mini_batch_size": 16
  }
  

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: 5
• seed: 12
• bf16: True
• 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: 5
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.0
• 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
• use_ipex: 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}
• tp_size: 0
• fsdp_transformer_layer_cls_to_wrap: None
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch
• optim_args: None
• adafactor: False
• group_by_length: False
• length_column_name: length
• ddp_find_unused_parameters: None
• ddp_bucket_cap_mb: None
• ddp_broadcast_buffers: False
• dataloader_pin_memory: True
• dataloader_persistent_workers: False
• skip_memory_metrics: True
• use_legacy_prediction_loop: False
• push_to_hub: False
• resume_from_checkpoint: None
• hub_model_id: None
• hub_strategy: every_save
• hub_private_repo: None
• hub_always_push: False
• 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: False
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• eval_use_gather_object: False
• average_tokens_across_devices: False
• 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	Validation Loss
0.0002	1	2.5736	-
0.0203	100	2.1106	2.0826
0.0407	200	2.0919	2.0803
0.0610	300	2.0899	2.0795
0.0813	400	2.0906	2.0782
0.1016	500	2.0776	2.0759
0.1220	600	2.0782	2.0745
0.1423	700	2.08	2.0746
0.1626	800	2.0788	2.0738
0.1830	900	2.0751	2.0730
0.2033	1000	2.0786	2.0733
0.2236	1100	2.0851	2.0737
0.2440	1200	2.0804	2.0728
0.2643	1300	2.0847	2.0729
0.2846	1400	2.0742	2.0722
0.3049	1500	2.0782	2.0727
0.3253	1600	2.0761	2.0719
0.3456	1700	2.082	2.0722
0.3659	1800	2.0711	2.0716
0.3863	1900	2.0686	2.0716
0.4066	2000	2.0703	2.0719
0.4269	2100	2.0702	2.0715
0.4472	2200	2.0804	2.0717
0.4676	2300	2.073	2.0711
0.4879	2400	2.0665	2.0712
0.5082	2500	2.0757	2.0710
0.5286	2600	2.0798	2.0717
0.5489	2700	2.0734	2.0709
0.5692	2800	2.0697	2.0706
0.5896	2900	2.0669	2.0703
0.6099	3000	2.0634	2.0708
0.6302	3100	2.0759	2.0703
0.6505	3200	2.069	2.0700
0.6709	3300	2.0643	2.0706
0.6912	3400	2.0684	2.0708
0.7115	3500	2.0763	2.0702
0.7319	3600	2.0777	2.0702
0.7522	3700	2.079	2.0701
0.7725	3800	2.0671	2.0705
0.7928	3900	2.0782	2.0703
0.8132	4000	2.0627	2.0705
0.8335	4100	2.0696	2.0699
0.8538	4200	2.0833	2.0697
0.8742	4300	2.0752	2.0697
0.8945	4400	2.0795	2.0698
0.9148	4500	2.0665	2.0696
0.9351	4600	2.0692	2.0698
0.9555	4700	2.0683	2.0698
0.9758	4800	2.0703	2.0698
0.9961	4900	2.074	2.0694
1.0165	5000	2.0639	2.0698
1.0368	5100	2.0569	2.0700
1.0571	5200	2.0704	2.0704
1.0775	5300	2.0758	2.0696
1.0978	5400	2.0648	2.0697
1.1181	5500	2.0581	2.0699
1.1384	5600	2.0672	2.0698
1.1588	5700	2.064	2.0696
1.1791	5800	2.0589	2.0710
1.1994	5900	2.0679	2.0703
1.2198	6000	2.0709	2.0705
1.2401	6100	2.0677	2.0704
1.2604	6200	2.0733	2.0700
1.2807	6300	2.0725	2.0695
1.3011	6400	2.0696	2.0706
1.3214	6500	2.0624	2.0698
1.3417	6600	2.0683	2.0709
1.3621	6700	2.0685	2.0705
1.3824	6800	2.0643	2.0703
1.4027	6900	2.0707	2.0713
1.4231	7000	2.0632	2.0715
1.4434	7100	2.0746	2.0702
1.4637	7200	2.0575	2.0707
1.4840	7300	2.0713	2.0709
1.5044	7400	2.0754	2.0719
1.5247	7500	2.0663	2.0715
1.5450	7600	2.0772	2.0704
1.5654	7700	2.0705	2.0708
1.5857	7800	2.0646	2.0710
1.6060	7900	2.0689	2.0701
1.6263	8000	2.0623	2.0703
1.6467	8100	2.0772	2.0700
1.6670	8200	2.0693	2.0711
1.6873	8300	2.0642	2.0706
1.7077	8400	2.0674	2.0702
1.7280	8500	2.0747	2.0705
1.7483	8600	2.0729	2.0708
1.7687	8700	2.0692	2.0711
1.7890	8800	2.0612	2.0704
1.8093	8900	2.0625	2.0703
1.8296	9000	2.0665	2.0710
1.8500	9100	2.0606	2.0702
1.8703	9200	2.0684	2.0701
1.8906	9300	2.0617	2.0698
1.9110	9400	2.0636	2.0697
1.9313	9500	2.0606	2.0706
1.9516	9600	2.0617	2.0700
1.9719	9700	2.067	2.0705
1.9923	9800	2.0769	2.0700
2.0126	9900	2.0564	2.0727
2.0329	10000	2.0682	2.0714
2.0533	10100	2.0604	2.0730
2.0736	10200	2.0528	2.0726
2.0939	10300	2.0573	2.0742
2.1143	10400	2.0569	2.0738
2.1346	10500	2.0592	2.0731
2.1549	10600	2.0522	2.0745
2.1752	10700	2.0539	2.0744
2.1956	10800	2.0493	2.0736
2.2159	10900	2.0566	2.0736
2.2362	11000	2.0546	2.0735
2.2566	11100	2.0585	2.0750
2.2769	11200	2.0569	2.0732
2.2972	11300	2.0653	2.0739
2.3175	11400	2.0377	2.0742
2.3379	11500	2.0616	2.0735
2.3582	11600	2.0574	2.0733
2.3785	11700	2.0584	2.0745
2.3989	11800	2.0611	2.0751
2.4192	11900	2.0708	2.0737
2.4395	12000	2.0571	2.0758
2.4598	12100	2.0602	2.0731
2.4802	12200	2.0644	2.0739
2.5005	12300	2.0556	2.0750
2.5208	12400	2.0534	2.0746
2.5412	12500	2.0541	2.0740
2.5615	12600	2.0615	2.0745
2.5818	12700	2.0565	2.0746
2.6022	12800	2.0534	2.0751
2.6225	12900	2.0533	2.0732
2.6428	13000	2.0607	2.0735
2.6631	13100	2.0544	2.0750
2.6835	13200	2.058	2.0743
2.7038	13300	2.0576	2.0737
2.7241	13400	2.0546	2.0740
2.7445	13500	2.0521	2.0738
2.7648	13600	2.0633	2.0743
2.7851	13700	2.0586	2.0732
2.8054	13800	2.0618	2.0738
2.8258	13900	2.0547	2.0738
2.8461	14000	2.0589	2.0752
2.8664	14100	2.05	2.0742
2.8868	14200	2.0592	2.0753
2.9071	14300	2.0552	2.0748
2.9274	14400	2.0555	2.0729
2.9478	14500	2.0517	2.0744
2.9681	14600	2.0593	2.0743
2.9884	14700	2.0606	2.0738
3.0087	14800	2.0498	2.0790
3.0291	14900	2.0451	2.0793
3.0494	15000	2.0415	2.0782
3.0697	15100	2.0408	2.0793
3.0901	15200	2.049	2.0785
3.1104	15300	2.0534	2.0807
3.1307	15400	2.0518	2.0784
3.1510	15500	2.0458	2.0799
3.1714	15600	2.0436	2.0798
3.1917	15700	2.0497	2.0808
3.2120	15800	2.0407	2.0796
3.2324	15900	2.0436	2.0794
3.2527	16000	2.0463	2.0792
3.2730	16100	2.0403	2.0789
3.2934	16200	2.0442	2.0799
3.3137	16300	2.0447	2.0788
3.3340	16400	2.0385	2.0809
3.3543	16500	2.045	2.0804
3.3747	16600	2.0457	2.0806
3.3950	16700	2.0451	2.0794
3.4153	16800	2.0509	2.0795
3.4357	16900	2.0513	2.0794
3.4560	17000	2.0395	2.0797
3.4763	17100	2.0495	2.0805
3.4966	17200	2.0417	2.0804
3.5170	17300	2.0547	2.0797
3.5373	17400	2.0389	2.0810
3.5576	17500	2.0502	2.0794
3.5780	17600	2.0457	2.0814
3.5983	17700	2.0427	2.0804
3.6186	17800	2.0386	2.0804
3.6390	17900	2.0429	2.0802
3.6593	18000	2.0537	2.0809
3.6796	18100	2.0441	2.0794
3.6999	18200	2.0438	2.0803
3.7203	18300	2.0419	2.0799
3.7406	18400	2.0535	2.0813
3.7609	18500	2.046	2.0798
3.7813	18600	2.0454	2.0821
3.8016	18700	2.0469	2.0788
3.8219	18800	2.0428	2.0806
3.8422	18900	2.0374	2.0795
3.8626	19000	2.0472	2.0801
3.8829	19100	2.0434	2.0810
3.9032	19200	2.0392	2.0822
3.9236	19300	2.048	2.0809
3.9439	19400	2.0429	2.0799
3.9642	19500	2.0492	2.0825
3.9845	19600	2.0508	2.0811
4.0049	19700	2.0434	2.0805
4.0252	19800	2.0287	2.0836
4.0455	19900	2.0458	2.0830
4.0659	20000	2.0405	2.0851
4.0862	20100	2.0367	2.0849
4.1065	20200	2.0328	2.0851
4.1269	20300	2.0353	2.0843
4.1472	20400	2.0396	2.0840
4.1675	20500	2.0251	2.0842
4.1878	20600	2.0314	2.0854
4.2082	20700	2.0425	2.0847
4.2285	20800	2.0331	2.0851
4.2488	20900	2.0329	2.0858
4.2692	21000	2.0372	2.0852
4.2895	21100	2.0341	2.0847
4.3098	21200	2.046	2.0864
4.3301	21300	2.0375	2.0846
4.3505	21400	2.0379	2.0844
4.3708	21500	2.0444	2.0847
4.3911	21600	2.0328	2.0850
4.4115	21700	2.0391	2.0853
4.4318	21800	2.0352	2.0840
4.4521	21900	2.0406	2.0840
4.4725	22000	2.0309	2.0842
4.4928	22100	2.0292	2.0843
4.5131	22200	2.0303	2.0860
4.5334	22300	2.0292	2.0851
4.5538	22400	2.0475	2.0841
4.5741	22500	2.033	2.0846
4.5944	22600	2.0354	2.0850
4.6148	22700	2.0462	2.0851
4.6351	22800	2.0305	2.0847
4.6554	22900	2.028	2.0853
4.6757	23000	2.0377	2.0844
4.6961	23100	2.04	2.0846
4.7164	23200	2.0472	2.0852
4.7367	23300	2.033	2.0850
4.7571	23400	2.0302	2.0849
4.7774	23500	2.0372	2.0842
4.7977	23600	2.0369	2.0860
4.8181	23700	2.0349	2.0846
4.8384	23800	2.0354	2.0858
4.8587	23900	2.04	2.0841
4.8790	24000	2.0392	2.0845
4.8994	24100	2.0334	2.0851
4.9197	24200	2.0344	2.0850
4.9400	24300	2.039	2.0848
4.9604	24400	2.0317	2.0846
4.9807	24500	2.0419	2.0844

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.11.13
• Sentence Transformers: 5.0.0
• Transformers: 4.51.0
• PyTorch: 2.9.1+cu126
• Accelerate: 1.8.1
• Datasets: 3.6.0
• Tokenizers: 0.21.4-dev.0

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",
}

ListNetLoss

@inproceedings{cao2007learning,
    title={Learning to Rank: From Pairwise Approach to Listwise Approach},
    author={Cao, Zhe and Qin, Tao and Liu, Tie-Yan and Tsai, Ming-Feng and Li, Hang},
    booktitle={Proceedings of the 24th international conference on Machine learning},
    pages={129--136},
    year={2007}
}