PyLate model based on antoinelouis/colbert-xm

This is a PyLate model finetuned from antoinelouis/colbert-xm on the train dataset. It maps sentences & paragraphs to sequences of 128-dimensional dense vectors and can be used for semantic textual similarity using the MaxSim operator.

Model Details

Model Description

• Model Type: PyLate model
• Base model: antoinelouis/colbert-xm
• Document Length: 256 tokens
• Query Length: 32 tokens
• Output Dimensionality: 128 tokens
• Similarity Function: MaxSim
• Training Dataset:
  • train
• Language: en

Model Sources

• Documentation: PyLate Documentation
• Repository: PyLate on GitHub
• Hugging Face: PyLate models on Hugging Face

Full Model Architecture

ColBERT(
 (0): Transformer({'max_seq_length': 255, 'do_lower_case': False}) with Transformer model: XmodModel 
 (1): Dense({'in_features': 768, 'out_features': 128, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity'})
)

Evaluation Results

nDCG and Recall scores for late interaction retrieval models on Turkish NanoBEIR. 👁 drawing

MrTyDi: 👁 drawing

nDCG@10 change from v1 to v2 on English NanoBEIR.

Usage

First install the PyLate library:

pip install -U pylate

Retrieval

PyLate provides a streamlined interface to index and retrieve documents using ColBERT models. The index leverages the Voyager HNSW index to efficiently handle document embeddings and enable fast retrieval.

Indexing documents

First, load the ColBERT model and initialize the Voyager index, then encode and index your documents:

from pylate import indexes, models, retrieve

# Step 1: Load the ColBERT model
model = models.ColBERT(
 model_name_or_path="99eren99/ColBERT-XM-v2",
 #document_length=512
)

language = "tr_TR" # Use a code from https://huggingface.co/facebook/xmod-base#languages
backbone = model[0].auto_model
if backbone.__class__.__name__.lower().startswith("xmod"):
 backbone.set_default_language(language)

# Step 2: Initialize the Voyager index
index = indexes.Voyager(
 index_folder="pylate-index",
 index_name="index",
 override=True, # This overwrites the existing index if any
)

# Step 3: Encode the documents
documents_ids = ["1", "2", "3"]
documents = ["document 1 text", "document 2 text", "document 3 text"]

documents_embeddings = model.encode(
 documents,
 batch_size=32,
 is_query=False, # Ensure that it is set to False to indicate that these are documents, not queries
 show_progress_bar=True,
)

# Step 4: Add document embeddings to the index by providing embeddings and corresponding ids
index.add_documents(
 documents_ids=documents_ids,
 documents_embeddings=documents_embeddings,
)

Note that you do not have to recreate the index and encode the documents every time. Once you have created an index and added the documents, you can re-use the index later by loading it:

# To load an index, simply instantiate it with the correct folder/name and without overriding it
index = indexes.Voyager(
 index_folder="pylate-index",
 index_name="index",
)

Retrieving top-k documents for queries

Once the documents are indexed, you can retrieve the top-k most relevant documents for a given set of queries. To do so, initialize the ColBERT retriever with the index you want to search in, encode the queries and then retrieve the top-k documents to get the top matches ids and relevance scores:

# Step 1: Initialize the ColBERT retriever
retriever = retrieve.ColBERT(index=index)

# Step 2: Encode the queries
queries_embeddings = model.encode(
 ["query for document 3", "query for document 1"],
 batch_size=32,
 is_query=True, # # Ensure that it is set to False to indicate that these are queries
 show_progress_bar=True,
)

# Step 3: Retrieve top-k documents
scores = retriever.retrieve(
 queries_embeddings=queries_embeddings,
 k=10, # Retrieve the top 10 matches for each query
)

Reranking

If you only want to use the ColBERT model to perform reranking on top of your first-stage retrieval pipeline without building an index, you can simply use rank function and pass the queries and documents to rerank:

from pylate import rank, models

queries = [
 "query A",
 "query B",
]

documents = [
 ["document A", "document B"],
 ["document 1", "document C", "document B"],
]

documents_ids = [
 [1, 2],
 [1, 3, 2],
]

model = models.ColBERT(
 model_name_or_path=pylate_model_id,
)

queries_embeddings = model.encode(
 queries,
 is_query=True,
)

documents_embeddings = model.encode(
 documents,
 is_query=False,
)

reranked_documents = rank.rerank(
 documents_ids=documents_ids,
 queries_embeddings=queries_embeddings,
 documents_embeddings=documents_embeddings,
)

Training Details

Checkpoints from 5000th, 7500th and 10000th steps were averaged.

Training Dataset

train

• Dataset: train at d8bad49
• Size: 640,000 training samples
• Loss: pylate.losses.distillation.Distillation

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 4
• gradient_accumulation_steps: 32
• learning_rate: 3e-05
• num_train_epochs: 2
• lr_scheduler_type: constant_with_warmup
• warmup_ratio: 0.1
• bf16: True
• tf32: True

All Hyperparameters

Training Logs

Framework Versions

• Python: 3.10.16
• Sentence Transformers: 4.0.2
• PyLate: 1.1.7
• Transformers: 4.48.2
• PyTorch: 2.5.1+cu124
• Accelerate: 1.2.1
• Datasets: 2.21.0
• Tokenizers: 0.21.0

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"
}

PyLate

@misc{PyLate,
title={PyLate: Flexible Training and Retrieval for Late Interaction Models},
author={Chaffin, Antoine and Sourty, Raphaël},
url={https://github.com/lightonai/pylate},
year={2024}
}

Base model

facebook/xmod-base

antoinelouis/colbert-xm

(4)