datalyes/problem2solution

Problem to Solution Retrieval

Dataset Description

This dataset is part of PatenTEB, a comprehensive benchmark for evaluating text embedding models on patent-specific tasks. PatenTEB comprises 15 tasks across retrieval, classification, paraphrase detection, and clustering, with 2.06 million examples designed to reflect real-world patent analysis workflows.

Paper: PatenTEB: A Comprehensive Benchmark and Model Family for Patent Text Embedding

Task Details

• Task Name: problem2solution
• Task Type: Retrieval
• Test Samples: 18,729

Asymmetric retrieval task connecting problem descriptions to their corresponding solutions. Evaluates whether models can connect a described technical need to a potential solution, testing role-complementary understanding.

Dataset Structure

This is a retrieval task where models find relevant patents given a query.

Splits:

• test: Query-document pairs for retrieval evaluation

Columns:

• first_ipcr3
• q
• problem
• solution
• first_ipcr3_count

Data Sample

Below is a 5-row preview of the test set:

first_ipcr3,q,problem,solution,first_ipcr3_count
A23,167-323-371-660-167,"obtain the subject food with moderate hardness and hetero-feel, by adding a coagulant to an aqueous sol consisting mainly of konjak (devil's tongue...","an aqueous sol made up of (a) 1pt.wt. of konjak (devil's tongue) mannan, (b) 2-10pts.wt. of a second gell-forming material such as potato starch or...",1
A23,015-945-594-805-145,"provide soluble particles with encapsulated aroma, releasing a burst of aroma upon dissolution in hot water, and containing aromatized vegetable oi...","the soluble particles with encapsulated aroma are prepared from 100% coffee-derived material or other hydrocarbon materials, comprise capsules rele...",1
A23,129-509-092-122-224,"obtain a blend which consists of a mixture of calcium salt of specified volatile fatty acids and palm oil pr palm oil and animal fat, which does no...","this blend contains (a) a mixture of (i) isobutyric acid and (ii) calcium salt of at least one acid selected from among isovaleric acid, valeric ac...",1
A23,095-431-918-294-016,provide a medicine for treatment of oxalate-related diseases in human,this invention provides materials and procedures for delivery of selected strains of bacteria and/or oxalate-degrading enzymes to intestinal tracts...,1
A23,036-386-316-122-372,obtain the subject paste containing kappa carrageenan and potassium ion and having body feeling and smooth texture by specifying mesh passability i...,this paste contains (a) 5-10wt.% of carrageenan consisting essentially of kappa carrageenan and (b) 2-30wt.% (based on the component a) of potassiu...,1

Evaluation Metrics

This task uses NDCG@10 (Normalized Discounted Cumulative Gain at rank 10) as the primary metric. NDCG measures ranking quality by discounting relevance scores by logarithmic position, normalized by the ideal ranking.

Usage

Load Dataset

from datasets import load_dataset

# Load the dataset
dataset = load_dataset("datalyes/{task_name}")

# Access test split
test_data = dataset['test']

Use with Sentence Transformers

from sentence_transformers import SentenceTransformer

# Load a patent-specialized model
model = SentenceTransformer("datalyes/patembed-base")

# Encode patent texts
embeddings = model.encode(test_data['text'])

Integrate with MTEB

This dataset is designed to be integrated with the MTEB (Massive Text Embedding Benchmark) framework. Integration with MTEB is in progress and will be available once the corresponding pull requests are accepted.

Benchmark Context

This dataset is part of a larger benchmark suite:

Benchmark Component	Description
PatenTEB	15 tasks covering retrieval, classification, paraphrase, clustering
Test Data (Released)	319,320 examples across all 15 tasks
Training/Validation Data	1.74 million examples (planned for future release)
Total Dataset Size	2.06 million annotated instances

Note: Currently, only the test split is publicly available. Training and validation data release is planned for a future date.

All 15 Tasks (NEW to MTEB):

• 3 classification tasks: Bloom timing, NLI directionality, IPC3 classification
• 2 clustering tasks: IPC-based, Inventor-based
• 8 retrieval tasks: 3 symmetric (IN/MIXED/OUT domain) + 5 asymmetric (fragment-to-full)
• 2 paraphrase tasks: Problem and solution paraphrase detection

MTEB Integration: Upcoming (PR in progress)

Citation

If you use this dataset, please cite our paper:

@misc{ayaou2025patentebcomprehensivebenchmarkmodel,
      title={PatenTEB: A Comprehensive Benchmark and Model Family for Patent Text Embedding}, 
      author={Iliass Ayaou and Denis Cavallucci},
      year={2025},
      eprint={2510.22264},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2510.22264}
}

License

This dataset is released under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) license.

• You are free to share and adapt the material
• You must give appropriate credit
• You may not use the material for commercial purposes
• If you remix, transform, or build upon the material, you must distribute your contributions under the same license

For full license details, see: https://creativecommons.org/licenses/by-nc-sa/4.0/

Contact

• Authors: Iliass Ayaou, Denis Cavallucci
• Institution: ICUBE Laboratory, INSA Strasbourg
• GitHub: github.com/iliass-y/patenteb
• HuggingFace: huggingface.co/datalyes