embed-eulaw-multi

A cross-lingual retriever for European Union law. It is a LoRA adapter over intfloat/multilingual-e5-small that finds the same provision of EU law across five languages: German, English, Spanish, French and Italian.

Give it a provision in one language and it retrieves the matching provision in another. It is a retriever, not a reranker, and not a question-answering model.

The number

Measured on a document-isolated held out split, 5,509 queries against a 5,509-passage pool:

	recall@1	95% CI	hits
base intfloat/multilingual-e5-small	0.2919	[0.2800, 0.3040]	1,608 / 5,509
+ this adapter	0.6480	[0.6353, 0.6605]	3,570 / 5,509
delta	+0.3561		+122.0%

The two intervals do not overlap.

Per language after adaptation: de 0.6534, en 0.6480, es 0.6400, fr 0.6480, it 0.6510. The spread is flat, so no single language is carrying the average. The largest gain is German, which starts weakest on the base at 0.1352.

What this number does and does not establish

It is in distribution. The adapter was trained and scored on the same corpus and the same task. It measures how much adaptation helps where it was trained. It is not evidence of transfer to other legal text, other domains, or other languages, and we have not measured that. A sibling model of ours gained substantially in distribution and then transferred flat out of origin, so please do not assume this one will not do the same.

It is one training run. We report the draw count, which is one. The training loop is only partly seeded: the data shuffle and dropout draw from a configured seed, but the LoRA down projection is initialised at attach time, before that seed is set. A re-run would start from a different adapter and we cannot promise it lands on the same number. That is not a claim it would move. We have not measured this adapter's run to run spread.

Passages were scored under a 256 token cap. We have not measured how often a relevant passage was truncated at that cap. The number above is the number under that setting.

Two earlier figures are retracted. Development runs reported +126.3% and +128.79% on a split that held out texts rather than whole documents. Those measured a different quantity and are withdrawn. They are not larger versions of +122.0%, the training volumes are not matched, and no difference between them may be read as the cost of anything.

Using it

Install the package, then download the adapter and load it from the local directory:

pip install 'quanfire-multilingual-embedding[neural]'
hf download quanfire-ai/embed-eulaw-multi --revision v1.0.0 --local-dir embed-eulaw-multi

from multilingual_embedding.embedding.neural.adapter import load_adapter

encoder, meta = load_adapter("embed-eulaw-multi")

texts = [
    "The processing of personal data shall be lawful only if the data subject has given consent.",
    "Die Verarbeitung personenbezogener Daten ist nur rechtmaessig, wenn die betroffene Person ihre Einwilligung erteilt hat.",
    "Member States shall ensure that fishing vessels exceeding 12 metres carry a satellite tracking device.",
]
vectors = encoder.encode_batch(texts)

load_adapter returns the encoder together with its metadata, because the metadata carries the prefixes needed to use the model correctly. encode_batch takes a list of strings; encode takes a single string.

On the three sentences above, the English and German expressions of the same provision sit at cosine 0.835, while the unrelated fisheries article sits at 0.379.

Base revision

This adapter's manifest predates our base pinning convention, so meta.checkpoint_revision reads back as None. The base revision it was built against is 614241f622f53c4eeff9890bdc4f31cfecc418b3, which we establish from the training host's model cache rather than from a field the run recorded: that snapshot is the only one present there and it predates the run by a month. Pass it explicitly if you need the base held still, since an upstream repository can change what sits behind a name:

encoder, meta = load_adapter(
    "embed-eulaw-multi",
    revision="614241f622f53c4eeff9890bdc4f31cfecc418b3",
)

How it was trained

LoRA rank 32, alpha 64, applied to the query and value projections. 589,824 adapter parameters. One epoch over 19,966 aligned pairs, learning rate 1e-4, batch size 256, mean pooling, 384 dimensions, max sequence length 256, normalised outputs.

The training pairs are not mined and not machine translated. The Union publishes the same act as professionally aligned expressions in many languages, so the German text and the French text of one regulation are two authentic expressions of the same law, aligned by the Union's own structure: the same CELEX identifier, the same article position. The alignment was done by the people who wrote the law.

The held out split isolates whole documents. No held out provision shares a document with anything in training.

The exact training tree cannot be handed over as a commit. The working copy of the training module on the build host was written after this adapter was, so the file on disk is provably not the file that ran. What we can hand over is the pair of commits that introduced the document level split, both verifiably in effect before this run started. The evaluation provenance is intact and independently checkable; it is the training tree that is not recoverable. No claim here asserts reproducible training.

Provenance and licence

The corpus is EU legal text from EUR-Lex, the Union's official legal database, taken through its official channels only. No mirrors and no third party re-hosts.

EUR-Lex legal texts are reusable under Creative Commons Attribution 4.0, granted by Commission Decision 2011/833/EU. Commercial reuse is named in the grant, not inferred. Database metadata is CC0.

Only the Official Journal is treated as the authentic source. Material that travels alongside it is out of scope and is not in the corpus: no third party inserts, no publisher marks, and no international accounting standards, which carry their own rights and would need separate clearance.

The weights are Apache-2.0. The corpus is not redistributed. The model is non-reconstructive: it is a 589,824 parameter adapter that emits vectors, not text, and it cannot reproduce its training corpus. Attribution for the underlying legal text is to the European Union via EUR-Lex.

Scope

Five languages, EU legal provisions, in distribution. It has not been evaluated on national law, on case law, on the other nineteen EU languages, or on any non legal text.