README.md

metadata

library_name: transformers
license: mit
language:
  - multilingual
pipeline_tag: sentence-similarity
tags:
  - text-embedding
  - embeddings
  - information-retrieval
  - beir
  - text-classification
  - language-model
  - text-clustering
  - text-semantic-similarity
  - text-evaluation
  - text-reranking
  - feature-extraction
  - sentence-similarity
  - Sentence Similarity
  - natural_questions
  - ms_marco
  - fever
  - hotpot_qa
  - mteb

`NLLB-LLM2Vec': Self-Distillation for Model Stacking Unlocks Cross-Lingual NLU in 200+ Languages

• Repository: https://github.com/fdschmidt93/trident-nllb-llm2vec
• Paper: https://arxiv.org/abs/2406.12739

NLLB-LLM2Vec multilingually extends LLM2Vec via efficient self-supervised distillation. We train the up-projection and LoRA adapters of the NLLB-LLM2Vec by forcing its mean-pooled token embeddings to match (via mean-squared error) the output of the original LLM2Vec.

This model has only been trained on self-supervised data and not yet been fine-tuned on any downstream task! This version is expected to perform better than self-supervised adaptation in the original paper, as LoRAs are merged into the model prior to task fine-tuning. The backbone of this model is LLM2Vec-Meta-Llama-31-8B-Instruct-mntp-unsup-simcse. We use the encoder of NLLB-600M.

Usage

import torch
import torch.nn.functional as F
from transformers import AutoTokenizer, AutoModel, AutoConfig

# Loading base Mistral model, along with custom code that enables bidirectional connections in decoder-only LLMs.
tokenizer = AutoTokenizer.from_pretrained(
    "facebook/nllb-200-distilled-600M"
)

model = AutoModel.from_pretrained(
    "fdschmidt93/NLLB-LLM2Vec-Meta-Llama-31-8B-Instruct-mntp-unsup-simcse",
    trust_remote_code=True,
    torch_dtype=torch.bfloat16,
    device_map="cuda" if torch.cuda.is_available() else "cpu",
)

# Encoding queries using instructions
instruction = (
    "Given a web search query, retrieve relevant passages that answer the query:"
)
queries = [
    [instruction, "how much protein should a female eat"],
    [instruction, "summit define"],
]
q_reps = l2v.encode(queries)

# Encoding documents. Instruction are not required for documents
documents = [
    "As a general guideline, the CDC's average requirement of protein for women ages 19 to 70 is 46 grams per day. But, as you can see from this chart, you'll need to increase that if you're expecting or training for a marathon. Check out the chart below to see how much protein you should be eating each day.",
    "Definition of summit for English Language Learners. : 1  the highest point of a mountain : the top of a mountain. : 2  the highest level. : 3  a meeting or series of meetings between the leaders of two or more governments.",
]
d_reps = l2v.encode(documents)

# Compute cosine similarity
q_reps_norm = F.normalize(q_reps, p=2, dim=1)
d_reps_norm = F.normalize(d_reps, p=2, dim=1)
cos_sim = q_reps_norm @ d_reps_norm.T

print(cos_sim)
"""
tensor([[0.7740, 0.5580],
        [0.4845, 0.4993]])
"""

Fine-tuning

You should fine-tune the model on labelled data unless you are using the model for unsupervised retrieval-style tasks. NLLB-LLM2Vec supports both AutoModelForSequenceClassification and AutoModelForTokenClassification.

import torch
from transformers import AutoModelForSequenceClassification, AutoModelForTokenClassification
from peft import get_peft_model
from peft.tuners.lora.config import LoraConfig

# Only attach LoRAs to the linear layers of LLM2Vec inside NLLB-LLM2Vec
lora_config = LoraConfig(
    lora_alpha = 32,
    target_modules = r".*llm2vec.*(self_attn\.(q|k|v|o)_proj|mlp\.(gate|up|down)_proj).*",
    bias = "none",
    task_type = "SEQ_CLS"
  )
model = AutoModel.from_pretrained(
    "fdschmidt93/NLLB-LLM2Vec-Meta-Llama-31-8B-Instruct-mntp-unsup-simcse",
    trust_remote_code=True,
    torch_dtype=torch.bfloat16,
)
model = get_peft_model(model, lora_config)

Questions

If you have any question about the code, feel free to email Fabian David Schmidt ([email protected]).

Citation

If you are using NLLB-LLM2Vec in your work, please cite

@misc{schmidt2024selfdistillationmodelstackingunlocks,
      title={Self-Distillation for Model Stacking Unlocks Cross-Lingual NLU in 200+ Languages}, 
      author={Fabian David Schmidt and Philipp Borchert and Ivan Vulić and Goran Glavaš},
      year={2024},
      eprint={2406.12739},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2406.12739}, 
}

The work has been accepted to EMNLP findings. The Bibtex will therefore be updated when the paper will be released on ACLAnthology.