Mattia Opper, Merve Basoz, Andrew Horne
We lifted 5 functions out of this paper's own repositories and ran 0 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| mbasoz/xllora-embedding | canonical | 0 of 5 |
| Function | Status | Where it lives |
|---|---|---|
| cl_or_sentemb_forward | Not yet run | mbasoz/xllora-embedding/src/simcse_train.py code served (permissive licence) · get_code("c30690135148911a") |
| generate_pairs_df | Not yet run | mbasoz/xllora-embedding/src/generate_answers_multigpu.py code served (permissive licence) · get_code("213cd4addc2af25d") |
| last_dir | Not yet run | mbasoz/xllora-embedding/src/evaluation_mteb.py code served (permissive licence) · get_code("c299ba74525ae352") |
| preprocess_sts | Not yet run | mbasoz/xllora-embedding/src/simcse_train.py code served (permissive licence) · get_code("1a1bd28fcf95f9eb") |
| to_chat_format | Not yet run | mbasoz/xllora-embedding/src/generate_answers_multigpu.py code served (permissive licence) · get_code("65c47b0359649891") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Embedding models are crucial to modern NLP. However, the creation of the most effective models relies on carefully constructed supervised finetuning data. For high resource languages, such as English, such datasets are readily available. However, for hundreds of other languages, they are simply non-existent. We investigate whether the advent of large language models can help to bridge this gap. We test three different strategies for generating synthetic triplet data used to optimise embedding models. These include in-context learning as well as two novel approaches, leveraging adapter composition and cross lingual finetuning of the LLM generator (XL-LoRA) respectively. We find that while in-context learning still falls short of strong non-synthetic baselines, adapter composition and XL-LoRA yield strong performance gains across a wide array of tasks and languages, offering a clear, scalable pathway to producing performant embedding models for a wide variety of languages.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2603.01732")
get_code_for_paper("2603.01732")
have("2603.01732")
Connect an agent — have() is free.