LateOn-Code

The LateOn-Code collection is composed of PyLate models optimized for code retrieval. These late interaction models are first pre-trained following the methodology of CoRNStack. These pre-trained models are then further fine-tuned on train sets of CoIR using the nv-retriever methodology to mine hard negatives while preventing false negatives.

We started from the two best ColBERT models on the BEIR benchmark for their respective sizes. The first one, LateOn-Code is based on in-house LateOn model, a new version of GTE-ModernColBERT-v1 built on ModernBERT-base (also developed at LightOn). This version underwent significantly deeper training, crossing the 57 mark on BEIR, almost a 2.5-point improvement and is thus SOTA by a large margin. We'll release this base model along with training data and boilerplates in the near future, so stay tuned! The second, LateOn-Code-edge is a smaller model based on the edge-colbert model family from mixedbread, using the smallest variant (Ettin-17M) for maximum efficiency. For more details on the training setup, please refer to our blogpost.

The original CoRNStack data in a format compatible with PyLate can be found here while the fine-tuning data can be found here. Training boilerplates can be found here in the PyLate repository

MTEB (Code, v1) benchmark results

Pre-trained models achieve very competitive results as the 17M model outperforms the very strong granite-embedding-small-english-r2 by an average of 1.7. This is truly impressive, as the granite model is almost three times bigger (17M vs 48M), but is also a beast on its own in the <100M parameters range. It also outperforms the larger granite variant (149M). The larger version nicely scales by improving over the performance of its little sibling by 6.5 on average.

Although the pre-training results are already very impressive given that they are mostly out-of-domain, running a proper fine-tuning using the training data of CoIR significantly boost the performance of the models. Notably, the 17M model increases from 57.50 to 66.64 (+9.14), getting pretty close to EmbeddingGemma-300M while being 17 times smaller. The larger one increases from 63.77 to 74.12 (+10.35), strongly outperforming EmbeddingGemma-300M and getting closer to strong LLM models such as Qwen3-Embedding-0.6B and C2LLM-0.5B while being much smaller.

Best result across all sizes is underlined. Best within each size category is bolded.

Colgrep

The LateOn-Code family model can easily be used within ColGrep, an easy-to-use search tool that give their powerful search capabilities to coding agent. It has been designed to extend grep capabilities to get the best of both world and is very effective to enhance the quality of the answer while diminishing answer time and tokens consumption. Given the performance of the very light-weight 17M model, it can easily run quickly on any computer.

Install ColGrep

# macOS / Linux
curl --proto '=https' --tlsv1.2 -LsSf https://github.com/lightonai/next-plaid/releases/latest/download/colgrep-installer.sh | sh

# Windows (PowerShell)
powershell -c "irm https://github.com/lightonai/next-plaid/releases/latest/download/colgrep-installer.ps1 | iex"

Search

# Semantic search — find code by meaning
colgrep "function that retries HTTP requests"

# Regex search
colgrep -e "async fn\s+\w+"

# Hybrid — regex narrows candidates, semantics ranks them
colgrep -e "Result<" "error handling" --include="*.rs"

Install for Claude Code

colgrep --install-claude-code

Choose a Model

# Set the model
colgrep set-model lightonai/LateOn-Code  # default: lightonai/LateOn-Code-edge

For more information about ColGrep, please refer to the official documentation

PyLate model based on lightonai/LateOn (unreleased yet)

This is a PyLate model finetuned from lightonai/LateOn (the base model has not been released (yet) but is a strong model hitting 67 on BEIR, stay tuned!) on the python, php, go, ruby, javascript and java datasets. 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: lightonai/LateOn-v0
• Document Length: 2048 tokens
• Query Length: 256 tokens
• Output Dimensionality: 128 tokens
• Similarity Function: MaxSim
• Training Datasets:
  • python
  • php
  • go
  • ruby
  • javascript
  • java
• Language: English, code
• License: Apache 2.0

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': 2047, 'do_lower_case': False, 'architecture': 'ModernBertModel'})
  (1): Dense({'in_features': 768, 'out_features': 128, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity', 'use_residual': False})
)

Usage

First install the PyLate library:

pip install -U pylate

Retrieval

Use this model with PyLate to index and retrieve documents. The index uses FastPLAID for efficient similarity search.

Indexing documents

Load the ColBERT model and initialize the PLAID 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="pylate_model_id",
)

# Step 2: Initialize the PLAID index
index = indexes.PLAID(
    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.PLAID(
    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,
)

Evaluation

Metrics

Py Late Information Retrieval

• Dataset: ['CodeSearchNetPython', 'CodeSearchNetJavascript', 'CodeSearchNetGo', 'CodeSearchNetRuby', 'CodeSearchNetJava', 'CodeSearchNetPhp']
• Evaluated with pylate.evaluation.pylate_information_retrieval_evaluator.PyLateInformationRetrievalEvaluator

Code Search Network

• Dataset: CodeSearchNet_mean
• Evaluated with pylate.evaluation.code_stack_network_evaluator.CodeSearchNetworkEvaluator

Training Details

Training Datasets

python

• Dataset: python at d821c55
• Size: 6,889,731 training samples
• Approximate statistics based on the first 1000 samples:

  	query	document	negative_0	negative_1	negative_2	negative_3	negative_4	negative_5	negative_6	negative_7	negative_8	negative_9	negative_10	negative_11	negative_12	negative_13	negative_14	negative_15	negative_16	negative_17	negative_18	negative_19	negative_20	negative_21	negative_22	negative_23	negative_24	negative_25	negative_26	negative_27	negative_28	negative_29	negative_30	negative_31	negative_32	negative_33	negative_34	negative_35	negative_36	negative_37	negative_38	negative_39	negative_40	negative_41	negative_42	negative_43	negative_44	negative_45	negative_46	negative_47	negative_48	negative_49	negative_50	negative_51	negative_52	negative_53	negative_54	negative_55	negative_56	negative_57	negative_58	negative_59	negative_60	negative_61	negative_62	negative_63	negative_64	negative_65	negative_66	negative_67	negative_68	negative_69	negative_70	negative_71	negative_72	negative_73	negative_74	negative_75	negative_76	negative_77	negative_78	negative_79	negative_80	negative_81	negative_82	negative_83	negative_84	negative_85	negative_86	negative_87	negative_88	negative_89	negative_90	negative_91	negative_92	negative_93	negative_94	negative_95	negative_96	negative_97	negative_98	negative_99	negative_scores
  type	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	list
  details	min: 7 tokens, mean: 24.12 tokens, max: 256 tokens	min: 13 tokens, mean: 124.24 tokens, max: 256 tokens	min: 6 tokens, mean: 102.93 tokens, max: 256 tokens	min: 6 tokens, mean: 100.95 tokens, max: 256 tokens	min: 6 tokens, mean: 98.81 tokens, max: 256 tokens	min: 6 tokens, mean: 97.66 tokens, max: 256 tokens	min: 7 tokens, mean: 100.24 tokens, max: 256 tokens	min: 8 tokens, mean: 99.31 tokens, max: 256 tokens	min: 6 tokens, mean: 99.38 tokens, max: 256 tokens	min: 6 tokens, mean: 97.76 tokens, max: 256 tokens	min: 6 tokens, mean: 100.62 tokens, max: 256 tokens	min: 7 tokens, mean: 102.2 tokens, max: 256 tokens	min: 8 tokens, mean: 99.57 tokens, max: 256 tokens	min: 6 tokens, mean: 105.36 tokens, max: 256 tokens	min: 6 tokens, mean: 105.1 tokens, max: 256 tokens	min: 8 tokens, mean: 97.96 tokens, max: 256 tokens	min: 7 tokens, mean: 102.75 tokens, max: 256 tokens	min: 6 tokens, mean: 100.27 tokens, max: 256 tokens	min: 8 tokens, mean: 97.89 tokens, max: 256 tokens	min: 6 tokens, mean: 100.83 tokens, max: 256 tokens	min: 6 tokens, mean: 100.24 tokens, max: 256 tokens	min: 8 tokens, mean: 94.86 tokens, max: 256 tokens	min: 6 tokens, mean: 102.06 tokens, max: 256 tokens	min: 7 tokens, mean: 97.33 tokens, max: 256 tokens	min: 8 tokens, mean: 99.28 tokens, max: 256 tokens	min: 6 tokens, mean: 99.07 tokens, max: 256 tokens	min: 6 tokens, mean: 104.46 tokens, max: 256 tokens	min: 6 tokens, mean: 100.14 tokens, max: 256 tokens	min: 6 tokens, mean: 104.86 tokens, max: 256 tokens	min: 6 tokens, mean: 104.23 tokens, max: 256 tokens	min: 6 tokens, mean: 101.24 tokens, max: 256 tokens	min: 6 tokens, mean: 102.47 tokens, max: 256 tokens	min: 6 tokens, mean: 103.95 tokens, max: 256 tokens	min: 8 tokens, mean: 100.49 tokens, max: 256 tokens	min: 6 tokens, mean: 101.04 tokens, max: 256 tokens	min: 6 tokens, mean: 101.99 tokens, max: 256 tokens	min: 6 tokens, mean: 104.11 tokens, max: 256 tokens	min: 7 tokens, mean: 102.69 tokens, max: 256 tokens	min: 6 tokens, mean: 104.26 tokens, max: 256 tokens	min: 6 tokens, mean: 104.61 tokens, max: 256 tokens	min: 7 tokens, mean: 105.69 tokens, max: 256 tokens	min: 6 tokens, mean: 102.66 tokens, max: 256 tokens	min: 6 tokens, mean: 100.26 tokens, max: 256 tokens	min: 6 tokens, mean: 104.47 tokens, max: 256 tokens	min: 6 tokens, mean: 104.86 tokens, max: 256 tokens	min: 6 tokens, mean: 104.49 tokens, max: 256 tokens	min: 6 tokens, mean: 100.12 tokens, max: 256 tokens	min: 6 tokens, mean: 105.86 tokens, max: 256 tokens	min: 8 tokens, mean: 103.38 tokens, max: 256 tokens	min: 6 tokens, mean: 107.83 tokens, max: 256 tokens	min: 6 tokens, mean: 104.13 tokens, max: 256 tokens	min: 6 tokens, mean: 102.61 tokens, max: 256 tokens	min: 6 tokens, mean: 106.11 tokens, max: 256 tokens	min: 8 tokens, mean: 107.79 tokens, max: 256 tokens	min: 6 tokens, mean: 104.54 tokens, max: 256 tokens	min: 6 tokens, mean: 106.19 tokens, max: 256 tokens	min: 6 tokens, mean: 104.62 tokens, max: 256 tokens	min: 6 tokens, mean: 101.92 tokens, max: 256 tokens	min: 6 tokens, mean: 99.12 tokens, max: 256 tokens	min: 8 tokens, mean: 102.54 tokens, max: 256 tokens	min: 8 tokens, mean: 103.7 tokens, max: 256 tokens	min: 8 tokens, mean: 104.09 tokens, max: 256 tokens	min: 8 tokens, mean: 101.61 tokens, max: 256 tokens	min: 7 tokens, mean: 104.18 tokens, max: 256 tokens	min: 6 tokens, mean: 104.56 tokens, max: 256 tokens	min: 6 tokens, mean: 103.59 tokens, max: 256 tokens	min: 6 tokens, mean: 104.55 tokens, max: 256 tokens	min: 6 tokens, mean: 102.95 tokens, max: 256 tokens	min: 6 tokens, mean: 103.91 tokens, max: 256 tokens	min: 6 tokens, mean: 107.13 tokens, max: 256 tokens	min: 6 tokens, mean: 106.22 tokens, max: 256 tokens	min: 8 tokens, mean: 103.66 tokens, max: 256 tokens	min: 6 tokens, mean: 102.49 tokens, max: 256 tokens	min: 9 tokens, mean: 101.41 tokens, max: 256 tokens	min: 6 tokens, mean: 102.56 tokens, max: 256 tokens	min: 6 tokens, mean: 105.9 tokens, max: 256 tokens	min: 6 tokens, mean: 104.3 tokens, max: 256 tokens	min: 6 tokens, mean: 101.44 tokens, max: 256 tokens	min: 6 tokens, mean: 103.99 tokens, max: 256 tokens	min: 6 tokens, mean: 104.28 tokens, max: 256 tokens	min: 6 tokens, mean: 104.46 tokens, max: 256 tokens	min: 6 tokens, mean: 105.45 tokens, max: 256 tokens	min: 6 tokens, mean: 103.9 tokens, max: 256 tokens	min: 6 tokens, mean: 103.97 tokens, max: 256 tokens	min: 8 tokens, mean: 103.85 tokens, max: 256 tokens	min: 7 tokens, mean: 105.92 tokens, max: 256 tokens	min: 6 tokens, mean: 102.82 tokens, max: 256 tokens	min: 6 tokens, mean: 101.99 tokens, max: 256 tokens	min: 6 tokens, mean: 103.84 tokens, max: 256 tokens	min: 6 tokens, mean: 101.51 tokens, max: 256 tokens	min: 6 tokens, mean: 105.28 tokens, max: 256 tokens	min: 6 tokens, mean: 105.18 tokens, max: 256 tokens	min: 6 tokens, mean: 107.3 tokens, max: 256 tokens	min: 6 tokens, mean: 108.62 tokens, max: 256 tokens	min: 8 tokens, mean: 108.81 tokens, max: 256 tokens	min: 6 tokens, mean: 101.51 tokens, max: 256 tokens	min: 6 tokens, mean: 105.4 tokens, max: 256 tokens	min: 7 tokens, mean: 105.98 tokens, max: 256 tokens	min: 8 tokens, mean: 105.64 tokens, max: 256 tokens	min: 8 tokens, mean: 103.76 tokens, max: 256 tokens	min: 6 tokens, mean: 102.87 tokens, max: 256 tokens	min: 6 tokens, mean: 102.58 tokens, max: 256 tokens	size: 100 elements

• Samples:

  query	document	negative_0
  Write the concordance entries to the output file(filename) See sample output files for format.	def write_concordance(self, filename):<br> all_keys = self.concordance_table.get_all_keys()<br> lines = []<br> for i in all_keys:<br> a = ""<br> a += i + ":"<br> f = self.concordance_table.get_value(i)<br> if f != None:<br> for s in f:<br> a += " " + str(s)<br> a += "\n"<br> lines.append(a)<br> a = open(filename, "w+")<br> for i in lines:<br> a.write(i)<br> a.close()	def write_concordance(self, filename):<br> out = ''<br> values = [x for x in self.concordance_table.hash_table if x is not None]<br> values.sort(key=lambda x: x[0])<br> for v in values:<br> out += f'{v[0]}: {" ".join(str(x) for x in sorted(set(v[1])))}\n' <br> with open(filename, 'w') as f:<br> f.write(out.rstrip())

• Loss: pylate.losses.cached_contrastive.CachedContrastive

php

• Dataset: php at d821c55
• Size: 2,676,409 training samples
• Approximate statistics based on the first 1000 samples:

  	query	document	negative_0	negative_1	negative_2	negative_3	negative_4	negative_5	negative_6	negative_7	negative_8	negative_9	negative_10	negative_11	negative_12	negative_13	negative_14	negative_15	negative_16	negative_17	negative_18	negative_19	negative_20	negative_21	negative_22	negative_23	negative_24	negative_25	negative_26	negative_27	negative_28	negative_29	negative_30	negative_31	negative_32	negative_33	negative_34	negative_35	negative_36	negative_37	negative_38	negative_39	negative_40	negative_41	negative_42	negative_43	negative_44	negative_45	negative_46	negative_47	negative_48	negative_49	negative_50	negative_51	negative_52	negative_53	negative_54	negative_55	negative_56	negative_57	negative_58	negative_59	negative_60	negative_61	negative_62	negative_63	negative_64	negative_65	negative_66	negative_67	negative_68	negative_69	negative_70	negative_71	negative_72	negative_73	negative_74	negative_75	negative_76	negative_77	negative_78	negative_79	negative_80	negative_81	negative_82	negative_83	negative_84	negative_85	negative_86	negative_87	negative_88	negative_89	negative_90	negative_91	negative_92	negative_93	negative_94	negative_95	negative_96	negative_97	negative_98	negative_99	negative_scores
  type	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	list
  details	min: 7 tokens, mean: 18.02 tokens, max: 219 tokens	min: 8 tokens, mean: 97.31 tokens, max: 256 tokens	min: 7 tokens, mean: 83.05 tokens, max: 256 tokens	min: 7 tokens, mean: 84.18 tokens, max: 256 tokens	min: 7 tokens, mean: 82.44 tokens, max: 256 tokens	min: 7 tokens, mean: 82.74 tokens, max: 256 tokens	min: 7 tokens, mean: 85.61 tokens, max: 256 tokens	min: 6 tokens, mean: 80.61 tokens, max: 256 tokens	min: 7 tokens, mean: 85.39 tokens, max: 256 tokens	min: 7 tokens, mean: 85.96 tokens, max: 256 tokens	min: 7 tokens, mean: 83.7 tokens, max: 256 tokens	min: 7 tokens, mean: 83.25 tokens, max: 256 tokens	min: 7 tokens, mean: 86.36 tokens, max: 256 tokens	min: 7 tokens, mean: 79.83 tokens, max: 256 tokens	min: 7 tokens, mean: 85.96 tokens, max: 256 tokens	min: 7 tokens, mean: 83.3 tokens, max: 256 tokens	min: 7 tokens, mean: 82.56 tokens, max: 256 tokens	min: 7 tokens, mean: 84.44 tokens, max: 256 tokens	min: 7 tokens, mean: 84.95 tokens, max: 256 tokens	min: 7 tokens, mean: 85.52 tokens, max: 256 tokens	min: 7 tokens, mean: 84.9 tokens, max: 256 tokens	min: 7 tokens, mean: 82.11 tokens, max: 256 tokens	min: 7 tokens, mean: 84.76 tokens, max: 256 tokens	min: 7 tokens, mean: 85.61 tokens, max: 256 tokens	min: 7 tokens, mean: 87.78 tokens, max: 256 tokens	min: 7 tokens, mean: 79.7 tokens, max: 256 tokens	min: 7 tokens, mean: 88.52 tokens, max: 256 tokens	min: 7 tokens, mean: 88.94 tokens, max: 256 tokens	min: 7 tokens, mean: 84.75 tokens, max: 256 tokens	min: 7 tokens, mean: 87.28 tokens, max: 256 tokens	min: 7 tokens, mean: 86.87 tokens, max: 256 tokens	min: 7 tokens, mean: 89.19 tokens, max: 256 tokens	min: 7 tokens, mean: 87.12 tokens, max: 256 tokens	min: 7 tokens, mean: 88.61 tokens, max: 256 tokens	min: 7 tokens, mean: 89.19 tokens, max: 256 tokens	min: 7 tokens, mean: 88.56 tokens, max: 256 tokens	min: 7 tokens, mean: 85.75 tokens, max: 256 tokens	min: 7 tokens, mean: 85.16 tokens, max: 256 tokens	min: 7 tokens, mean: 87.35 tokens, max: 256 tokens	min: 7 tokens, mean: 90.14 tokens, max: 256 tokens	min: 7 tokens, mean: 86.28 tokens, max: 256 tokens	min: 7 tokens, mean: 86.32 tokens, max: 256 tokens	min: 7 tokens, mean: 84.63 tokens, max: 256 tokens	min: 7 tokens, mean: 88.19 tokens, max: 256 tokens	min: 7 tokens, mean: 87.46 tokens, max: 256 tokens	min: 7 tokens, mean: 86.83 tokens, max: 256 tokens	min: 7 tokens, mean: 89.91 tokens, max: 256 tokens	min: 7 tokens, mean: 90.59 tokens, max: 256 tokens	min: 7 tokens, mean: 87.58 tokens, max: 256 tokens	min: 7 tokens, mean: 89.3 tokens, max: 256 tokens	min: 7 tokens, mean: 93.99 tokens, max: 256 tokens	min: 7 tokens, mean: 88.55 tokens, max: 256 tokens	min: 7 tokens, mean: 86.46 tokens, max: 256 tokens	min: 7 tokens, mean: 83.97 tokens, max: 256 tokens	min: 7 tokens, mean: 86.73 tokens, max: 256 tokens	min: 7 tokens, mean: 88.11 tokens, max: 256 tokens	min: 7 tokens, mean: 85.57 tokens, max: 256 tokens	min: 7 tokens, mean: 87.64 tokens, max: 256 tokens	min: 7 tokens, mean: 88.58 tokens, max: 256 tokens	min: 7 tokens, mean: 89.99 tokens, max: 256 tokens	min: 7 tokens, mean: 85.44 tokens, max: 256 tokens	min: 7 tokens, mean: 88.96 tokens, max: 256 tokens	min: 7 tokens, mean: 90.66 tokens, max: 256 tokens	min: 7 tokens, mean: 88.72 tokens, max: 256 tokens	min: 7 tokens, mean: 93.31 tokens, max: 256 tokens	min: 7 tokens, mean: 87.37 tokens, max: 256 tokens	min: 7 tokens, mean: 91.06 tokens, max: 256 tokens	min: 7 tokens, mean: 90.74 tokens, max: 256 tokens	min: 6 tokens, mean: 85.83 tokens, max: 256 tokens	min: 7 tokens, mean: 87.6 tokens, max: 256 tokens	min: 7 tokens, mean: 87.71 tokens, max: 256 tokens	min: 7 tokens, mean: 90.29 tokens, max: 256 tokens	min: 7 tokens, mean: 91.09 tokens, max: 256 tokens	min: 7 tokens, mean: 87.94 tokens, max: 256 tokens	min: 7 tokens, mean: 90.81 tokens, max: 256 tokens	min: 6 tokens, mean: 89.77 tokens, max: 256 tokens	min: 7 tokens, mean: 84.67 tokens, max: 256 tokens	min: 6 tokens, mean: 88.34 tokens, max: 256 tokens	min: 7 tokens, mean: 87.25 tokens, max: 256 tokens	min: 7 tokens, mean: 91.56 tokens, max: 256 tokens	min: 7 tokens, mean: 90.43 tokens, max: 256 tokens	min: 6 tokens, mean: 86.3 tokens, max: 256 tokens	min: 6 tokens, mean: 92.18 tokens, max: 256 tokens	min: 7 tokens, mean: 90.68 tokens, max: 256 tokens	min: 6 tokens, mean: 90.08 tokens, max: 256 tokens	min: 7 tokens, mean: 94.62 tokens, max: 256 tokens	min: 7 tokens, mean: 89.4 tokens, max: 256 tokens	min: 7 tokens, mean: 82.08 tokens, max: 256 tokens	min: 7 tokens, mean: 87.92 tokens, max: 256 tokens	min: 7 tokens, mean: 88.84 tokens, max: 256 tokens	min: 7 tokens, mean: 89.72 tokens, max: 256 tokens	min: 7 tokens, mean: 92.3 tokens, max: 256 tokens	min: 6 tokens, mean: 87.56 tokens, max: 256 tokens	min: 6 tokens, mean: 88.55 tokens, max: 256 tokens	min: 7 tokens, mean: 90.84 tokens, max: 256 tokens	min: 7 tokens, mean: 84.04 tokens, max: 256 tokens	min: 7 tokens, mean: 91.26 tokens, max: 256 tokens	min: 7 tokens, mean: 89.11 tokens, max: 256 tokens	min: 6 tokens, mean: 93.41 tokens, max: 256 tokens	min: 7 tokens, mean: 86.29 tokens, max: 256 tokens	min: 7 tokens, mean: 87.78 tokens, max: 256 tokens	min: 7 tokens, mean: 87.01 tokens, max: 256 tokens	size: 100 elements

• Samples:

  query	document	negative_0
  return boolean as string 'true' / 'false'	function bool2str($bool) {<br> if($bool ===false)<br> return 'false';<br> else<br> return 'true';<br>}	function bool_s($boolean) {<br> return ($boolean ? 'true' : 'false');<br>}

• Loss: pylate.losses.cached_contrastive.CachedContrastive

go

• Dataset: go at d821c55
• Size: 5,815,734 training samples
• Approximate statistics based on the first 1000 samples:

  	query	document	negative_0	negative_1	negative_2	negative_3	negative_4	negative_5	negative_6	negative_7	negative_8	negative_9	negative_10	negative_11	negative_12	negative_13	negative_14	negative_15	negative_16	negative_17	negative_18	negative_19	negative_20	negative_21	negative_22	negative_23	negative_24	negative_25	negative_26	negative_27	negative_28	negative_29	negative_30	negative_31	negative_32	negative_33	negative_34	negative_35	negative_36	negative_37	negative_38	negative_39	negative_40	negative_41	negative_42	negative_43	negative_44	negative_45	negative_46	negative_47	negative_48	negative_49	negative_50	negative_51	negative_52	negative_53	negative_54	negative_55	negative_56	negative_57	negative_58	negative_59	negative_60	negative_61	negative_62	negative_63	negative_64	negative_65	negative_66	negative_67	negative_68	negative_69	negative_70	negative_71	negative_72	negative_73	negative_74	negative_75	negative_76	negative_77	negative_78	negative_79	negative_80	negative_81	negative_82	negative_83	negative_84	negative_85	negative_86	negative_87	negative_88	negative_89	negative_90	negative_91	negative_92	negative_93	negative_94	negative_95	negative_96	negative_97	negative_98	negative_99	negative_scores
  type	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	list
  details	min: 7 tokens, mean: 23.67 tokens, max: 217 tokens	min: 13 tokens, mean: 109.68 tokens, max: 256 tokens	min: 16 tokens, mean: 107.33 tokens, max: 256 tokens	min: 12 tokens, mean: 102.9 tokens, max: 256 tokens	min: 14 tokens, mean: 103.89 tokens, max: 256 tokens	min: 10 tokens, mean: 106.78 tokens, max: 256 tokens	min: 9 tokens, mean: 106.23 tokens, max: 256 tokens	min: 9 tokens, mean: 106.45 tokens, max: 256 tokens	min: 9 tokens, mean: 107.99 tokens, max: 256 tokens	min: 9 tokens, mean: 104.84 tokens, max: 256 tokens	min: 12 tokens, mean: 106.47 tokens, max: 256 tokens	min: 11 tokens, mean: 105.55 tokens, max: 256 tokens	min: 10 tokens, mean: 108.57 tokens, max: 256 tokens	min: 12 tokens, mean: 108.62 tokens, max: 256 tokens	min: 12 tokens, mean: 110.55 tokens, max: 256 tokens	min: 12 tokens, mean: 106.26 tokens, max: 256 tokens	min: 12 tokens, mean: 106.64 tokens, max: 256 tokens	min: 14 tokens, mean: 106.96 tokens, max: 256 tokens	min: 11 tokens, mean: 103.05 tokens, max: 256 tokens	min: 9 tokens, mean: 107.45 tokens, max: 256 tokens	min: 13 tokens, mean: 105.3 tokens, max: 256 tokens	min: 15 tokens, mean: 105.44 tokens, max: 256 tokens	min: 12 tokens, mean: 104.67 tokens, max: 256 tokens	min: 13 tokens, mean: 111.42 tokens, max: 256 tokens	min: 15 tokens, mean: 107.38 tokens, max: 256 tokens	min: 10 tokens, mean: 107.34 tokens, max: 256 tokens	min: 10 tokens, mean: 102.53 tokens, max: 256 tokens	min: 10 tokens, mean: 108.49 tokens, max: 256 tokens	min: 10 tokens, mean: 111.58 tokens, max: 256 tokens	min: 12 tokens, mean: 105.18 tokens, max: 256 tokens	min: 12 tokens, mean: 108.69 tokens, max: 256 tokens	min: 15 tokens, mean: 108.0 tokens, max: 256 tokens	min: 12 tokens, mean: 105.84 tokens, max: 256 tokens	min: 10 tokens, mean: 106.44 tokens, max: 256 tokens	min: 13 tokens, mean: 105.24 tokens, max: 256 tokens	min: 13 tokens, mean: 104.68 tokens, max: 256 tokens	min: 11 tokens, mean: 106.39 tokens, max: 256 tokens	min: 14 tokens, mean: 105.06 tokens, max: 256 tokens	min: 13 tokens, mean: 107.31 tokens, max: 256 tokens	min: 11 tokens, mean: 110.77 tokens, max: 256 tokens	min: 11 tokens, mean: 106.06 tokens, max: 256 tokens	min: 16 tokens, mean: 109.77 tokens, max: 256 tokens	min: 16 tokens, mean: 109.91 tokens, max: 256 tokens	min: 10 tokens, mean: 108.52 tokens, max: 256 tokens	min: 11 tokens, mean: 110.54 tokens, max: 256 tokens	min: 15 tokens, mean: 107.61 tokens, max: 256 tokens	min: 13 tokens, mean: 108.65 tokens, max: 256 tokens	min: 12 tokens, mean: 106.42 tokens, max: 256 tokens	min: 10 tokens, mean: 105.84 tokens, max: 256 tokens	min: 11 tokens, mean: 111.49 tokens, max: 256 tokens	min: 11 tokens, mean: 108.21 tokens, max: 256 tokens	min: 11 tokens, mean: 104.42 tokens, max: 256 tokens	min: 8 tokens, mean: 112.23 tokens, max: 256 tokens	min: 14 tokens, mean: 109.97 tokens, max: 256 tokens	min: 8 tokens, mean: 108.53 tokens, max: 256 tokens	min: 14 tokens, mean: 103.8 tokens, max: 256 tokens	min: 14 tokens, mean: 108.26 tokens, max: 256 tokens	min: 13 tokens, mean: 104.47 tokens, max: 256 tokens	min: 13 tokens, mean: 109.63 tokens, max: 256 tokens	min: 10 tokens, mean: 107.78 tokens, max: 256 tokens	min: 12 tokens, mean: 107.51 tokens, max: 256 tokens	min: 14 tokens, mean: 106.38 tokens, max: 256 tokens	min: 10 tokens, mean: 111.95 tokens, max: 256 tokens	min: 11 tokens, mean: 108.62 tokens, max: 256 tokens	min: 13 tokens, mean: 108.69 tokens, max: 256 tokens	min: 12 tokens, mean: 110.8 tokens, max: 256 tokens	min: 14 tokens, mean: 105.22 tokens, max: 256 tokens	min: 14 tokens, mean: 108.6 tokens, max: 256 tokens	min: 14 tokens, mean: 111.24 tokens, max: 256 tokens	min: 13 tokens, mean: 106.55 tokens, max: 256 tokens	min: 13 tokens, mean: 110.18 tokens, max: 256 tokens	min: 12 tokens, mean: 110.22 tokens, max: 256 tokens	min: 12 tokens, mean: 111.2 tokens, max: 256 tokens	min: 12 tokens, mean: 110.16 tokens, max: 256 tokens	min: 14 tokens, mean: 108.52 tokens, max: 256 tokens	min: 13 tokens, mean: 110.53 tokens, max: 256 tokens	min: 15 tokens, mean: 111.13 tokens, max: 256 tokens	min: 14 tokens, mean: 104.19 tokens, max: 256 tokens	min: 9 tokens, mean: 108.67 tokens, max: 256 tokens	min: 11 tokens, mean: 111.0 tokens, max: 256 tokens	min: 14 tokens, mean: 110.76 tokens, max: 256 tokens	min: 13 tokens, mean: 109.73 tokens, max: 256 tokens	min: 12 tokens, mean: 105.15 tokens, max: 256 tokens	min: 14 tokens, mean: 111.64 tokens, max: 256 tokens	min: 6 tokens, mean: 108.8 tokens, max: 256 tokens	min: 13 tokens, mean: 110.11 tokens, max: 256 tokens	min: 7 tokens, mean: 105.51 tokens, max: 256 tokens	min: 11 tokens, mean: 108.64 tokens, max: 256 tokens	min: 15 tokens, mean: 105.54 tokens, max: 256 tokens	min: 10 tokens, mean: 107.4 tokens, max: 256 tokens	min: 12 tokens, mean: 108.55 tokens, max: 256 tokens	min: 13 tokens, mean: 108.38 tokens, max: 256 tokens	min: 16 tokens, mean: 110.22 tokens, max: 256 tokens	min: 15 tokens, mean: 112.5 tokens, max: 256 tokens	min: 12 tokens, mean: 108.49 tokens, max: 256 tokens	min: 15 tokens, mean: 109.87 tokens, max: 256 tokens	min: 12 tokens, mean: 108.58 tokens, max: 256 tokens	min: 14 tokens, mean: 111.7 tokens, max: 256 tokens	min: 16 tokens, mean: 111.45 tokens, max: 256 tokens	min: 9 tokens, mean: 110.57 tokens, max: 256 tokens	min: 12 tokens, mean: 107.72 tokens, max: 256 tokens	min: 14 tokens, mean: 110.13 tokens, max: 256 tokens	size: 100 elements

• Samples:

  query	document	negative_0
  Returns the value of the 'go_package' option of the first .proto file found in the same directory as projectFile	func detectGoPackageForProject(projectFile string) (string, error) {<br> var goPkg string<br> projectDir := filepath.Dir(projectFile)<br> if err := filepath.Walk(projectDir, func(protoFile string, info os.FileInfo, err error) error {<br> // already set<br> if goPkg != "" {<br> return nil<br> }<br> if !strings.HasSuffix(protoFile, ".proto") {<br> return nil<br> }<br> // search for go_package on protos in the same dir as the project.json<br> if projectDir != filepath.Dir(protoFile) {<br> return nil<br> }<br> content, err := ioutil.ReadFile(protoFile)<br> if err != nil {<br> return err<br> }<br> lines := strings.Split(string(content), "\n")<br> for _, line := range lines {<br> goPackage := goPackageStatementRegex.FindStringSubmatch(line)<br> if len(goPackage) == 0 {<br> continue<br> }<br> if len(goPackage) != 2 {<br> return errors.Errorf("parsing go_package error: from %v found %v", line, goPackage)<br> }<br> goPkg = goPackage[1]<br> break<br> }<br> return nil<br> }); err != nil {<br> return "", err<br> }<br> if goPkg == "" {<br> return "", errors.Er...	func (g *Generator) GoFilePackage(depfile *fdep.DepFile) string {<br> return fproto_wrap.BaseName(g.GoWrapPackage(depfile))<br>}

• Loss: pylate.losses.cached_contrastive.CachedContrastive

ruby

• Dataset: ruby at d821c55
• Size: 631,161 training samples
• Approximate statistics based on the first 1000 samples:

  	query	document	negative_0	negative_1	negative_2	negative_3	negative_4	negative_5	negative_6	negative_7	negative_8	negative_9	negative_10	negative_11	negative_12	negative_13	negative_14	negative_15	negative_16	negative_17	negative_18	negative_19	negative_20	negative_21	negative_22	negative_23	negative_24	negative_25	negative_26	negative_27	negative_28	negative_29	negative_30	negative_31	negative_32	negative_33	negative_34	negative_35	negative_36	negative_37	negative_38	negative_39	negative_40	negative_41	negative_42	negative_43	negative_44	negative_45	negative_46	negative_47	negative_48	negative_49	negative_50	negative_51	negative_52	negative_53	negative_54	negative_55	negative_56	negative_57	negative_58	negative_59	negative_60	negative_61	negative_62	negative_63	negative_64	negative_65	negative_66	negative_67	negative_68	negative_69	negative_70	negative_71	negative_72	negative_73	negative_74	negative_75	negative_76	negative_77	negative_78	negative_79	negative_80	negative_81	negative_82	negative_83	negative_84	negative_85	negative_86	negative_87	negative_88	negative_89	negative_90	negative_91	negative_92	negative_93	negative_94	negative_95	negative_96	negative_97	negative_98	negative_99	negative_scores
  type	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	list
  details	min: 7 tokens, mean: 27.46 tokens, max: 256 tokens	min: 8 tokens, mean: 81.82 tokens, max: 256 tokens	min: 7 tokens, mean: 69.28 tokens, max: 256 tokens	min: 7 tokens, mean: 70.87 tokens, max: 256 tokens	min: 7 tokens, mean: 68.13 tokens, max: 256 tokens	min: 7 tokens, mean: 70.04 tokens, max: 256 tokens	min: 7 tokens, mean: 65.91 tokens, max: 256 tokens	min: 7 tokens, mean: 69.42 tokens, max: 256 tokens	min: 7 tokens, mean: 67.64 tokens, max: 256 tokens	min: 7 tokens, mean: 68.89 tokens, max: 256 tokens	min: 7 tokens, mean: 70.62 tokens, max: 256 tokens	min: 7 tokens, mean: 70.05 tokens, max: 256 tokens	min: 7 tokens, mean: 73.0 tokens, max: 256 tokens	min: 7 tokens, mean: 68.58 tokens, max: 256 tokens	min: 7 tokens, mean: 71.39 tokens, max: 256 tokens	min: 7 tokens, mean: 72.31 tokens, max: 256 tokens	min: 7 tokens, mean: 69.39 tokens, max: 256 tokens	min: 7 tokens, mean: 71.65 tokens, max: 256 tokens	min: 7 tokens, mean: 70.94 tokens, max: 256 tokens	min: 7 tokens, mean: 68.46 tokens, max: 256 tokens	min: 7 tokens, mean: 67.78 tokens, max: 256 tokens	min: 7 tokens, mean: 70.1 tokens, max: 256 tokens	min: 7 tokens, mean: 71.75 tokens, max: 256 tokens	min: 7 tokens, mean: 72.49 tokens, max: 256 tokens	min: 7 tokens, mean: 69.72 tokens, max: 256 tokens	min: 7 tokens, mean: 70.09 tokens, max: 256 tokens	min: 7 tokens, mean: 70.19 tokens, max: 256 tokens	min: 7 tokens, mean: 72.2 tokens, max: 256 tokens	min: 7 tokens, mean: 72.02 tokens, max: 256 tokens	min: 7 tokens, mean: 70.91 tokens, max: 256 tokens	min: 7 tokens, mean: 73.2 tokens, max: 256 tokens	min: 7 tokens, mean: 71.11 tokens, max: 256 tokens	min: 7 tokens, mean: 70.94 tokens, max: 256 tokens	min: 7 tokens, mean: 74.89 tokens, max: 256 tokens	min: 7 tokens, mean: 69.67 tokens, max: 256 tokens	min: 7 tokens, mean: 71.91 tokens, max: 256 tokens	min: 7 tokens, mean: 71.25 tokens, max: 256 tokens	min: 7 tokens, mean: 71.58 tokens, max: 256 tokens	min: 7 tokens, mean: 72.9 tokens, max: 256 tokens	min: 7 tokens, mean: 75.1 tokens, max: 256 tokens	min: 7 tokens, mean: 74.55 tokens, max: 256 tokens	min: 7 tokens, mean: 77.13 tokens, max: 256 tokens	min: 7 tokens, mean: 73.25 tokens, max: 256 tokens	min: 7 tokens, mean: 68.97 tokens, max: 256 tokens	min: 7 tokens, mean: 72.48 tokens, max: 256 tokens	min: 7 tokens, mean: 72.67 tokens, max: 256 tokens	min: 7 tokens, mean: 74.04 tokens, max: 256 tokens	min: 7 tokens, mean: 70.5 tokens, max: 256 tokens	min: 7 tokens, mean: 72.2 tokens, max: 256 tokens	min: 7 tokens, mean: 73.39 tokens, max: 256 tokens	min: 7 tokens, mean: 73.69 tokens, max: 256 tokens	min: 7 tokens, mean: 71.32 tokens, max: 256 tokens	min: 7 tokens, mean: 74.51 tokens, max: 256 tokens	min: 7 tokens, mean: 72.13 tokens, max: 256 tokens	min: 7 tokens, mean: 75.34 tokens, max: 256 tokens	min: 7 tokens, mean: 75.59 tokens, max: 256 tokens	min: 7 tokens, mean: 72.12 tokens, max: 256 tokens	min: 7 tokens, mean: 73.14 tokens, max: 256 tokens	min: 7 tokens, mean: 76.15 tokens, max: 256 tokens	min: 7 tokens, mean: 73.08 tokens, max: 256 tokens	min: 7 tokens, mean: 75.75 tokens, max: 256 tokens	min: 7 tokens, mean: 72.52 tokens, max: 256 tokens	min: 7 tokens, mean: 70.75 tokens, max: 256 tokens	min: 7 tokens, mean: 69.18 tokens, max: 256 tokens	min: 7 tokens, mean: 70.06 tokens, max: 256 tokens	min: 7 tokens, mean: 72.35 tokens, max: 256 tokens	min: 7 tokens, mean: 73.01 tokens, max: 256 tokens	min: 7 tokens, mean: 72.39 tokens, max: 256 tokens	min: 7 tokens, mean: 73.27 tokens, max: 256 tokens	min: 7 tokens, mean: 72.95 tokens, max: 256 tokens	min: 7 tokens, mean: 72.0 tokens, max: 256 tokens	min: 7 tokens, mean: 71.09 tokens, max: 256 tokens	min: 7 tokens, mean: 71.23 tokens, max: 256 tokens	min: 7 tokens, mean: 72.0 tokens, max: 256 tokens	min: 7 tokens, mean: 72.24 tokens, max: 256 tokens	min: 7 tokens, mean: 73.3 tokens, max: 256 tokens	min: 7 tokens, mean: 74.85 tokens, max: 256 tokens	min: 7 tokens, mean: 72.45 tokens, max: 256 tokens	min: 7 tokens, mean: 75.66 tokens, max: 256 tokens	min: 7 tokens, mean: 75.36 tokens, max: 256 tokens	min: 7 tokens, mean: 71.31 tokens, max: 256 tokens	min: 7 tokens, mean: 72.53 tokens, max: 256 tokens	min: 7 tokens, mean: 70.6 tokens, max: 256 tokens	min: 7 tokens, mean: 72.82 tokens, max: 256 tokens	min: 7 tokens, mean: 72.79 tokens, max: 256 tokens	min: 7 tokens, mean: 72.75 tokens, max: 256 tokens	min: 7 tokens, mean: 72.92 tokens, max: 256 tokens	min: 7 tokens, mean: 74.62 tokens, max: 256 tokens	min: 7 tokens, mean: 73.26 tokens, max: 256 tokens	min: 7 tokens, mean: 72.5 tokens, max: 256 tokens	min: 7 tokens, mean: 72.96 tokens, max: 256 tokens	min: 7 tokens, mean: 69.5 tokens, max: 256 tokens	min: 7 tokens, mean: 71.73 tokens, max: 256 tokens	min: 7 tokens, mean: 71.43 tokens, max: 256 tokens	min: 7 tokens, mean: 72.52 tokens, max: 256 tokens	min: 7 tokens, mean: 70.29 tokens, max: 256 tokens	min: 7 tokens, mean: 73.48 tokens, max: 256 tokens	min: 7 tokens, mean: 73.07 tokens, max: 256 tokens	min: 7 tokens, mean: 73.89 tokens, max: 256 tokens	min: 7 tokens, mean: 73.68 tokens, max: 256 tokens	min: 7 tokens, mean: 74.27 tokens, max: 256 tokens	min: 7 tokens, mean: 75.13 tokens, max: 256 tokens	size: 100 elements

• Samples:

  query	document	negative_0
  GET /property_between_floor_slaps GET /property_between_floor_slaps.json	def index<br> @property_between_floor_slaps = PropertyBetweenFloorSlap.all<br> end	def set_property_between_floor_slap<br> @property_between_floor_slap = PropertyBetweenFloorSlap.find(params[:id])<br> end

• Loss: pylate.losses.cached_contrastive.CachedContrastive

javascript

• Dataset: javascript at d821c55
• Size: 1,386,353 training samples
• Approximate statistics based on the first 1000 samples:

  	query	document	negative_0	negative_1	negative_2	negative_3	negative_4	negative_5	negative_6	negative_7	negative_8	negative_9	negative_10	negative_11	negative_12	negative_13	negative_14	negative_15	negative_16	negative_17	negative_18	negative_19	negative_20	negative_21	negative_22	negative_23	negative_24	negative_25	negative_26	negative_27	negative_28	negative_29	negative_30	negative_31	negative_32	negative_33	negative_34	negative_35	negative_36	negative_37	negative_38	negative_39	negative_40	negative_41	negative_42	negative_43	negative_44	negative_45	negative_46	negative_47	negative_48	negative_49	negative_50	negative_51	negative_52	negative_53	negative_54	negative_55	negative_56	negative_57	negative_58	negative_59	negative_60	negative_61	negative_62	negative_63	negative_64	negative_65	negative_66	negative_67	negative_68	negative_69	negative_70	negative_71	negative_72	negative_73	negative_74	negative_75	negative_76	negative_77	negative_78	negative_79	negative_80	negative_81	negative_82	negative_83	negative_84	negative_85	negative_86	negative_87	negative_88	negative_89	negative_90	negative_91	negative_92	negative_93	negative_94	negative_95	negative_96	negative_97	negative_98	negative_99	negative_scores
  type	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	list
  details	min: 7 tokens, mean: 23.61 tokens, max: 256 tokens	min: 9 tokens, mean: 122.61 tokens, max: 256 tokens	min: 7 tokens, mean: 108.12 tokens, max: 256 tokens	min: 6 tokens, mean: 112.04 tokens, max: 256 tokens	min: 6 tokens, mean: 109.22 tokens, max: 256 tokens	min: 7 tokens, mean: 110.62 tokens, max: 256 tokens	min: 7 tokens, mean: 109.77 tokens, max: 256 tokens	min: 6 tokens, mean: 112.58 tokens, max: 256 tokens	min: 6 tokens, mean: 108.78 tokens, max: 256 tokens	min: 6 tokens, mean: 108.21 tokens, max: 256 tokens	min: 7 tokens, mean: 111.04 tokens, max: 256 tokens	min: 6 tokens, mean: 108.05 tokens, max: 256 tokens	min: 7 tokens, mean: 108.58 tokens, max: 256 tokens	min: 6 tokens, mean: 110.81 tokens, max: 256 tokens	min: 6 tokens, mean: 110.29 tokens, max: 256 tokens	min: 7 tokens, mean: 108.2 tokens, max: 256 tokens	min: 7 tokens, mean: 109.19 tokens, max: 256 tokens	min: 6 tokens, mean: 111.03 tokens, max: 256 tokens	min: 6 tokens, mean: 111.45 tokens, max: 256 tokens	min: 6 tokens, mean: 110.48 tokens, max: 256 tokens	min: 6 tokens, mean: 110.37 tokens, max: 256 tokens	min: 7 tokens, mean: 114.22 tokens, max: 256 tokens	min: 7 tokens, mean: 112.62 tokens, max: 256 tokens	min: 7 tokens, mean: 113.8 tokens, max: 256 tokens	min: 7 tokens, mean: 110.23 tokens, max: 256 tokens	min: 7 tokens, mean: 112.49 tokens, max: 256 tokens	min: 7 tokens, mean: 109.46 tokens, max: 256 tokens	min: 7 tokens, mean: 113.62 tokens, max: 256 tokens	min: 7 tokens, mean: 108.73 tokens, max: 256 tokens	min: 7 tokens, mean: 107.68 tokens, max: 256 tokens	min: 7 tokens, mean: 112.89 tokens, max: 256 tokens	min: 6 tokens, mean: 110.91 tokens, max: 256 tokens	min: 6 tokens, mean: 107.13 tokens, max: 256 tokens	min: 6 tokens, mean: 110.15 tokens, max: 256 tokens	min: 6 tokens, mean: 111.73 tokens, max: 256 tokens	min: 6 tokens, mean: 113.99 tokens, max: 256 tokens	min: 7 tokens, mean: 110.67 tokens, max: 256 tokens	min: 7 tokens, mean: 115.34 tokens, max: 256 tokens	min: 7 tokens, mean: 111.74 tokens, max: 256 tokens	min: 6 tokens, mean: 115.7 tokens, max: 256 tokens	min: 6 tokens, mean: 116.1 tokens, max: 256 tokens	min: 6 tokens, mean: 114.03 tokens, max: 256 tokens	min: 6 tokens, mean: 114.13 tokens, max: 256 tokens	min: 6 tokens, mean: 115.99 tokens, max: 256 tokens	min: 6 tokens, mean: 113.55 tokens, max: 256 tokens	min: 7 tokens, mean: 116.25 tokens, max: 256 tokens	min: 7 tokens, mean: 114.8 tokens, max: 256 tokens	min: 8 tokens, mean: 114.66 tokens, max: 256 tokens	min: 6 tokens, mean: 112.9 tokens, max: 256 tokens	min: 7 tokens, mean: 112.67 tokens, max: 256 tokens	min: 8 tokens, mean: 112.66 tokens, max: 256 tokens	min: 6 tokens, mean: 112.93 tokens, max: 256 tokens	min: 6 tokens, mean: 112.36 tokens, max: 256 tokens	min: 7 tokens, mean: 115.37 tokens, max: 256 tokens	min: 7 tokens, mean: 116.0 tokens, max: 256 tokens	min: 7 tokens, mean: 117.86 tokens, max: 256 tokens	min: 6 tokens, mean: 112.58 tokens, max: 256 tokens	min: 6 tokens, mean: 112.56 tokens, max: 256 tokens	min: 6 tokens, mean: 110.88 tokens, max: 256 tokens	min: 6 tokens, mean: 111.73 tokens, max: 256 tokens	min: 7 tokens, mean: 112.62 tokens, max: 256 tokens	min: 7 tokens, mean: 117.56 tokens, max: 256 tokens	min: 6 tokens, mean: 110.65 tokens, max: 256 tokens	min: 6 tokens, mean: 116.67 tokens, max: 256 tokens	min: 6 tokens, mean: 120.18 tokens, max: 256 tokens	min: 6 tokens, mean: 113.18 tokens, max: 256 tokens	min: 6 tokens, mean: 111.28 tokens, max: 256 tokens	min: 6 tokens, mean: 112.35 tokens, max: 256 tokens	min: 6 tokens, mean: 115.84 tokens, max: 256 tokens	min: 6 tokens, mean: 107.41 tokens, max: 256 tokens	min: 6 tokens, mean: 112.68 tokens, max: 256 tokens	min: 6 tokens, mean: 113.94 tokens, max: 256 tokens	min: 6 tokens, mean: 115.98 tokens, max: 256 tokens	min: 6 tokens, mean: 115.12 tokens, max: 256 tokens	min: 6 tokens, mean: 117.5 tokens, max: 256 tokens	min: 7 tokens, mean: 110.15 tokens, max: 256 tokens	min: 7 tokens, mean: 111.66 tokens, max: 256 tokens	min: 6 tokens, mean: 114.64 tokens, max: 256 tokens	min: 7 tokens, mean: 115.12 tokens, max: 256 tokens	min: 7 tokens, mean: 114.63 tokens, max: 256 tokens	min: 6 tokens, mean: 114.87 tokens, max: 256 tokens	min: 6 tokens, mean: 113.57 tokens, max: 256 tokens	min: 6 tokens, mean: 112.34 tokens, max: 256 tokens	min: 7 tokens, mean: 114.15 tokens, max: 256 tokens	min: 7 tokens, mean: 110.8 tokens, max: 256 tokens	min: 7 tokens, mean: 115.0 tokens, max: 256 tokens	min: 7 tokens, mean: 115.64 tokens, max: 256 tokens	min: 7 tokens, mean: 113.33 tokens, max: 256 tokens	min: 7 tokens, mean: 114.12 tokens, max: 256 tokens	min: 7 tokens, mean: 116.79 tokens, max: 256 tokens	min: 7 tokens, mean: 113.86 tokens, max: 256 tokens	min: 7 tokens, mean: 114.26 tokens, max: 256 tokens	min: 6 tokens, mean: 112.31 tokens, max: 256 tokens	min: 6 tokens, mean: 114.21 tokens, max: 256 tokens	min: 7 tokens, mean: 117.53 tokens, max: 256 tokens	min: 6 tokens, mean: 115.43 tokens, max: 256 tokens	min: 6 tokens, mean: 116.28 tokens, max: 256 tokens	min: 6 tokens, mean: 113.86 tokens, max: 256 tokens	min: 6 tokens, mean: 114.42 tokens, max: 256 tokens	min: 6 tokens, mean: 112.08 tokens, max: 256 tokens	min: 6 tokens, mean: 115.86 tokens, max: 256 tokens	min: 7 tokens, mean: 115.5 tokens, max: 256 tokens	size: 100 elements

• Samples:

  query	document	negative_0
  Example binToHex(["0111110", "1000000", "1000000", "1111110", "1000001", "1000001", "0111110"])	function binToHex(bins) {<br> return bins.map(bin => ("00" + (parseInt(bin, 2).toString(16))).substr(-2).toUpperCase()).join("");<br>}	function binToHex(a) {<br> var newVal = "";<br> for (i = 0; i < a.length/8; i++)<br> newVal += ("00" + parseInt(a.slice(8i, 8i+8),2).toString(16)).slice(-2);<br> return newVal;<br>}

• Loss: pylate.losses.cached_contrastive.CachedContrastive

java

• Dataset: java at d821c55
• Size: 4,103,086 training samples
• Approximate statistics based on the first 1000 samples:

  	query	document	negative_0	negative_1	negative_2	negative_3	negative_4	negative_5	negative_6	negative_7	negative_8	negative_9	negative_10	negative_11	negative_12	negative_13	negative_14	negative_15	negative_16	negative_17	negative_18	negative_19	negative_20	negative_21	negative_22	negative_23	negative_24	negative_25	negative_26	negative_27	negative_28	negative_29	negative_30	negative_31	negative_32	negative_33	negative_34	negative_35	negative_36	negative_37	negative_38	negative_39	negative_40	negative_41	negative_42	negative_43	negative_44	negative_45	negative_46	negative_47	negative_48	negative_49	negative_50	negative_51	negative_52	negative_53	negative_54	negative_55	negative_56	negative_57	negative_58	negative_59	negative_60	negative_61	negative_62	negative_63	negative_64	negative_65	negative_66	negative_67	negative_68	negative_69	negative_70	negative_71	negative_72	negative_73	negative_74	negative_75	negative_76	negative_77	negative_78	negative_79	negative_80	negative_81	negative_82	negative_83	negative_84	negative_85	negative_86	negative_87	negative_88	negative_89	negative_90	negative_91	negative_92	negative_93	negative_94	negative_95	negative_96	negative_97	negative_98	negative_99	negative_scores
  type	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	string	list
  details	min: 7 tokens, mean: 21.93 tokens, max: 256 tokens	min: 7 tokens, mean: 77.23 tokens, max: 256 tokens	min: 6 tokens, mean: 71.27 tokens, max: 256 tokens	min: 6 tokens, mean: 69.49 tokens, max: 256 tokens	min: 6 tokens, mean: 72.21 tokens, max: 256 tokens	min: 6 tokens, mean: 68.86 tokens, max: 256 tokens	min: 6 tokens, mean: 70.93 tokens, max: 256 tokens	min: 6 tokens, mean: 67.48 tokens, max: 256 tokens	min: 6 tokens, mean: 69.5 tokens, max: 256 tokens	min: 6 tokens, mean: 71.93 tokens, max: 256 tokens	min: 7 tokens, mean: 69.65 tokens, max: 256 tokens	min: 6 tokens, mean: 74.01 tokens, max: 256 tokens	min: 6 tokens, mean: 69.52 tokens, max: 256 tokens	min: 6 tokens, mean: 76.13 tokens, max: 256 tokens	min: 6 tokens, mean: 69.41 tokens, max: 256 tokens	min: 6 tokens, mean: 74.8 tokens, max: 256 tokens	min: 6 tokens, mean: 74.64 tokens, max: 256 tokens	min: 6 tokens, mean: 69.87 tokens, max: 256 tokens	min: 6 tokens, mean: 73.56 tokens, max: 256 tokens	min: 6 tokens, mean: 75.1 tokens, max: 256 tokens	min: 6 tokens, mean: 72.11 tokens, max: 256 tokens	min: 6 tokens, mean: 73.47 tokens, max: 256 tokens	min: 6 tokens, mean: 73.88 tokens, max: 256 tokens	min: 6 tokens, mean: 73.89 tokens, max: 256 tokens	min: 7 tokens, mean: 74.63 tokens, max: 256 tokens	min: 6 tokens, mean: 73.94 tokens, max: 256 tokens	min: 6 tokens, mean: 74.29 tokens, max: 256 tokens	min: 6 tokens, mean: 73.66 tokens, max: 256 tokens	min: 6 tokens, mean: 73.94 tokens, max: 256 tokens	min: 6 tokens, mean: 73.71 tokens, max: 256 tokens	min: 6 tokens, mean: 76.59 tokens, max: 256 tokens	min: 6 tokens, mean: 73.44 tokens, max: 256 tokens	min: 6 tokens, mean: 73.41 tokens, max: 256 tokens	min: 6 tokens, mean: 73.11 tokens, max: 256 tokens	min: 6 tokens, mean: 73.49 tokens, max: 256 tokens	min: 6 tokens, mean: 74.49 tokens, max: 256 tokens	min: 6 tokens, mean: 75.75 tokens, max: 256 tokens	min: 6 tokens, mean: 71.11 tokens, max: 256 tokens	min: 6 tokens, mean: 72.25 tokens, max: 256 tokens	min: 6 tokens, mean: 74.47 tokens, max: 256 tokens	min: 7 tokens, mean: 75.86 tokens, max: 256 tokens	min: 7 tokens, mean: 73.47 tokens, max: 256 tokens	min: 7 tokens, mean: 76.36 tokens, max: 256 tokens	min: 6 tokens, mean: 79.31 tokens, max: 256 tokens	min: 6 tokens, mean: 74.5 tokens, max: 256 tokens	min: 6 tokens, mean: 75.54 tokens, max: 256 tokens	min: 6 tokens, mean: 77.99 tokens, max: 256 tokens	min: 6 tokens, mean: 76.56 tokens, max: 256 tokens	min: 6 tokens, mean: 74.27 tokens, max: 256 tokens	min: 6 tokens, mean: 77.28 tokens, max: 256 tokens	min: 6 tokens, mean: 76.97 tokens, max: 256 tokens	min: 6 tokens, mean: 76.73 tokens, max: 256 tokens	min: 6 tokens, mean: 70.69 tokens, max: 256 tokens	min: 6 tokens, mean: 75.53 tokens, max: 256 tokens	min: 6 tokens, mean: 73.91 tokens, max: 256 tokens	min: 6 tokens, mean: 76.89 tokens, max: 256 tokens	min: 6 tokens, mean: 73.97 tokens, max: 256 tokens	min: 6 tokens, mean: 74.69 tokens, max: 256 tokens	min: 6 tokens, mean: 75.5 tokens, max: 256 tokens	min: 6 tokens, mean: 72.88 tokens, max: 256 tokens	min: 6 tokens, mean: 76.94 tokens, max: 256 tokens	min: 6 tokens, mean: 77.67 tokens, max: 256 tokens	min: 6 tokens, mean: 76.24 tokens, max: 256 tokens	min: 6 tokens, mean: 77.79 tokens, max: 256 tokens	min: 6 tokens, mean: 75.04 tokens, max: 256 tokens	min: 6 tokens, mean: 75.43 tokens, max: 256 tokens	min: 6 tokens, mean: 74.78 tokens, max: 256 tokens	min: 6 tokens, mean: 77.16 tokens, max: 256 tokens	min: 6 tokens, mean: 75.1 tokens, max: 256 tokens	min: 6 tokens, mean: 77.79 tokens, max: 256 tokens	min: 6 tokens, mean: 72.59 tokens, max: 256 tokens	min: 6 tokens, mean: 77.14 tokens, max: 256 tokens	min: 6 tokens, mean: 73.62 tokens, max: 256 tokens	min: 6 tokens, mean: 82.23 tokens, max: 256 tokens	min: 6 tokens, mean: 75.65 tokens, max: 256 tokens	min: 6 tokens, mean: 76.31 tokens, max: 256 tokens	min: 6 tokens, mean: 76.04 tokens, max: 256 tokens	min: 6 tokens, mean: 74.85 tokens, max: 256 tokens	min: 6 tokens, mean: 78.05 tokens, max: 256 tokens	min: 6 tokens, mean: 76.59 tokens, max: 256 tokens	min: 6 tokens, mean: 78.1 tokens, max: 256 tokens	min: 6 tokens, mean: 76.14 tokens, max: 256 tokens	min: 6 tokens, mean: 73.1 tokens, max: 256 tokens	min: 6 tokens, mean: 75.61 tokens, max: 256 tokens	min: 6 tokens, mean: 75.79 tokens, max: 256 tokens	min: 6 tokens, mean: 77.7 tokens, max: 256 tokens	min: 6 tokens, mean: 75.6 tokens, max: 256 tokens	min: 6 tokens, mean: 77.71 tokens, max: 256 tokens	min: 6 tokens, mean: 75.1 tokens, max: 256 tokens	min: 6 tokens, mean: 75.92 tokens, max: 256 tokens	min: 6 tokens, mean: 76.13 tokens, max: 256 tokens	min: 6 tokens, mean: 79.2 tokens, max: 256 tokens	min: 6 tokens, mean: 76.79 tokens, max: 256 tokens	min: 6 tokens, mean: 73.95 tokens, max: 256 tokens	min: 6 tokens, mean: 76.74 tokens, max: 256 tokens	min: 6 tokens, mean: 76.28 tokens, max: 256 tokens	min: 6 tokens, mean: 75.48 tokens, max: 256 tokens	min: 6 tokens, mean: 80.97 tokens, max: 256 tokens	min: 6 tokens, mean: 73.05 tokens, max: 256 tokens	min: 6 tokens, mean: 78.6 tokens, max: 256 tokens	min: 6 tokens, mean: 79.92 tokens, max: 256 tokens	min: 6 tokens, mean: 74.14 tokens, max: 256 tokens	size: 100 elements

• Samples:

  query	document	negative_0
  private void signSetter(String[] lines, Player p, Block s)	private void signSetter(Block b, Player p, String[] lines) <br> { <br> //TODO: virer debug<br> //p.sendMessage("dbg1");<br> <br> <br> if(b==null) <br> return;<br> <br> BoutiqueSign bs = new BoutiqueSign();<br> <br> bs.setOwner(p);<br> bs.setLocation(b.getLocation());<br> bs.setLines(lines);<br><br> //TODO: virer debug<br> /*<br> p.sendMessage("dbg1 : line1 = " + bs.getLine1());<br> p.sendMessage("dbg1 : line2 = " + bs.getLine2());<br> p.sendMessage("dbg1 : line3 = " + bs.getLine3());<br> p.sendMessage("dbg1 : line4 = " + bs.getLine4()); <br> p.sendMessage("dbg2 : type = " + bs.getType());<br> */<br> <br> if(bs.isSignServer())<br> {<br> <br> if (!PermissionsHandler.canSetGlobalSign(p))<br> {<br> p.sendMessage(PermissionsHandler.permissionErr);<br> return;<br> }<br> <br> if(!bs.checkLines(p))<br> {<br> return;<br> }<br> <br> p.sendMessage(plugin.chatPrefix + Messages.getString("Sign.SERVERSIGNADDED")); //$NON-NLS-1$<br> }<br> <br> else if(bs.isSignChest())<br> {<br> if (!PermissionsHandler.canSetPersonalSign(p))<br> {<br> p.sendMessage(plugin.chatPrefix +...	void updateSignToPlayer(Player player, Location location, String[] lines);

• Loss: pylate.losses.cached_contrastive.CachedContrastive

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 128
• per_device_eval_batch_size: 128
• learning_rate: 8e-06
• num_train_epochs: 1
• bf16: True
• dataloader_num_workers: 8
• accelerator_config: {'split_batches': True, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}

All Hyperparameters

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 128
• per_device_eval_batch_size: 128
• 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: 8e-06
• 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: 0.0
• 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: 42
• 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: 2
• ddp_backend: None
• tpu_num_cores: None
• tpu_metrics_debug: False
• debug: []
• dataloader_drop_last: True
• dataloader_num_workers: 8
• dataloader_prefetch_factor: None
• past_index: -1
• disable_tqdm: False
• remove_unused_columns: True
• label_names: None
• load_best_model_at_end: False
• 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': True, '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

Framework Versions

• Python: 3.12.9
• Sentence Transformers: 5.1.1
• PyLate: 1.3.4
• Transformers: 4.52.3
• PyTorch: 2.7.0+cu126
• Accelerate: 1.10.1
• Datasets: 4.4.1
• Tokenizers: 0.21.1

Citation

BibTeX

LateOn-Code

@misc{LateOn-Code,
  title  = {LateOn-Code: a Family of State-Of-The-Art Late Interaction Code Retrieval Models},
  author = {Chaffin, Antoine},
  url    = {https://huggingface.co/collections/lightonai/lateon-code},
  year   = {2026}
}

ColGrep

@software{next-plaid,
  title  = {NextPlaid, ColGREP: Multi-vector search, from database to coding agents.},
  url    = {https://github.com/lightonai/next-plaid},
  author = {Raphaël Sourty},
  year   = {2026},
}

CoRNStack

@inproceedings{DBLP:conf/iclr/SureshRXNMDJ25,
  author       = {Tarun Suresh and
                  Revanth Gangi Reddy and
                  Yifei Xu and
                  Zach Nussbaum and
                  Andriy Mulyar and
                  Brandon Duderstadt and
                  Heng Ji},
  title        = {CoRNStack: High-Quality Contrastive Data for Better Code Retrieval
                  and Reranking},
  booktitle    = {The Thirteenth International Conference on Learning Representations,
                  {ICLR} 2025, Singapore, April 24-28, 2025},
  publisher    = {OpenReview.net},
  year         = {2025},
  url          = {https://openreview.net/forum?id=iyJOUELYir},
  timestamp    = {Sun, 25 May 2025 21:25:19 +0200},
  biburl       = {https://dblp.org/rec/conf/iclr/SureshRXNMDJ25.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}

CoIR

@inproceedings{li2025coir,
  title     = {Coir: A comprehensive benchmark for code information retrieval models},
  author    = {Li, Xiangyang and Dong, Kuicai and Lee, Yi Quan and Xia, Wei and Zhang, Hao and Dai, Xinyi and Wang, Yasheng and Tang, Ruiming},
  booktitle = {Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  pages     = {22074--22091},
  year      = {2025}
}

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

@inproceedings{DBLP:conf/cikm/ChaffinS25,
  author       = {Antoine Chaffin and
                  Rapha{"{e}}l Sourty},
  editor       = {Meeyoung Cha and
                  Chanyoung Park and
                  Noseong Park and
                  Carl Yang and
                  Senjuti Basu Roy and
                  Jessie Li and
                  Jaap Kamps and
                  Kijung Shin and
                  Bryan Hooi and
                  Lifang He},
  title        = {PyLate: Flexible Training and Retrieval for Late Interaction Models},
  booktitle    = {Proceedings of the 34th {ACM} International Conference on Information
                  and Knowledge Management, {CIKM} 2025, Seoul, Republic of Korea, November
                  10-14, 2025},
  pages        = {6334--6339},
  publisher    = {{ACM}},
  year         = {2025},
  url          = {https://github.com/lightonai/pylate},
  doi          = {10.1145/3746252.3761608},
}

CachedContrastive

@misc{gao2021scaling,
  title         = {Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
  author        = {Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
  year          = {2021},
  eprint        = {2101.06983},
  archivePrefix = {arXiv},
  primaryClass  = {cs.LG}
}