NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models

• URL: http://arxiv.org/abs/2405.17428v2
• Date: Thu, 09 Jan 2025 22:27:06 GMT
• Title: NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models
• Authors: Chankyu Lee, Rajarshi Roy, Mengyao Xu, Jonathan Raiman, Mohammad Shoeybi, Bryan Catanzaro, Wei Ping,
• Abstract summary: We introduce the NV-Embed model, incorporating architectural designs, training procedures, and curated datasets. For model architecture, we propose a latent attention layer to obtain pooled embeddings, which consistently improves retrieval and downstream task accuracy. For training data, we utilize the hard-negative mining, synthetic data generation and existing public available datasets to boost the performance of embedding model.
• Score: 38.41524186248607
• License:
• Abstract: Decoder-only large language model (LLM)-based embedding models are beginning to outperform BERT or T5-based embedding models in general-purpose text embedding tasks, including dense vector-based retrieval. In this work, we introduce the NV-Embed model, incorporating architectural designs, training procedures, and curated datasets to significantly enhance the performance of LLM as a versatile embedding model, while maintaining its simplicity and reproducibility. For model architecture, we propose a latent attention layer to obtain pooled embeddings, which consistently improves retrieval and downstream task accuracy compared to mean pooling or using the last <EOS> token embedding from LLMs. To enhance representation learning, we remove the causal attention mask of LLMs during contrastive training. For training algorithm, we introduce a two-stage contrastive instruction-tuning method. It first applies contrastive training with instructions on retrieval datasets, utilizing in-batch negatives and curated hard negative examples. At stage-2, it blends various non-retrieval into instruction tuning, which not only enhances non-retrieval task accuracy but also improves retrieval performance. For training data, we utilize the hard-negative mining, synthetic data generation and existing public available datasets to boost the performance of embedding model. By combining these techniques, our NV-Embed-v1 and NV-Embed-v2 models obtained the No.1 position on the Massive Text Embedding Benchmark (MTEB) (as of May 24, 2024 and August 30, 2024, respectively) across 56 embedding tasks, demonstrating the sustained effectiveness of the proposed methods over time. Additionally, it achieved the highest scores in the Long Doc section and the second-highest scores in the QA section of the AIR Benchmark, which covers a range of out-of-domain information retrieval topics beyond those in MTEB.

Related papers

• Rethinking Data Synthesis: A Teacher Model Training Recipe with Interpretation [12.736045604858738]
  Recent advances in large language model (LLM) training have highlighted the need for diverse, high-quality instruction data. We propose a paradigm shift named textbfNOMAD by investigating how to specifically train models for data generation.
  arXiv  Detail & Related papers  (2024-10-27T07:38:39Z)
• Is Tokenization Needed for Masked Particle Modelling? [8.79008927474707]
  Masked particle modeling (MPM) is a self-supervised learning scheme for constructing expressive representations of unordered sets. We improve MPM by addressing inefficiencies in the implementation and incorporating a more powerful decoder. We show that these new methods outperform the tokenized learning objective from the original MPM on a new test bed for foundation models for jets.
  arXiv  Detail & Related papers  (2024-09-19T09:12:29Z)
• NVLM: Open Frontier-Class Multimodal LLMs [64.00053046838225]
  We introduce NVLM 1.0, a family of frontier-class multimodal large language models (LLMs) that achieve state-of-the-art results on vision-language tasks. We propose a novel architecture that enhances both training efficiency and multimodal reasoning capabilities. We develop production-grade multimodality for the NVLM-1.0 models, enabling them to excel in vision-language tasks.
  arXiv  Detail & Related papers  (2024-09-17T17:59:06Z)
• Accelerating Large Language Model Pretraining via LFR Pedagogy: Learn, Focus, and Review [50.78587571704713]
  Learn-Focus-Review (LFR) is a dynamic training approach that adapts to the model's learning progress. LFR tracks the model's learning performance across data blocks (sequences of tokens) and prioritizes revisiting challenging regions of the dataset. Compared to baseline models trained on the full datasets, LFR consistently achieved lower perplexity and higher accuracy.
  arXiv  Detail & Related papers  (2024-09-10T00:59:18Z)
• Pooling And Attention: What Are Effective Designs For LLM-Based Embedding Models? [18.990655668481075]
  We propose a new pooling strategy, Multi-Layers Trainable Pooling, which transforms the outputs of all hidden layers, rather than just the last layer, using a cross-attention network. This paper sheds light on effective training strategies for LLM-based embedding models.
  arXiv  Detail & Related papers  (2024-09-04T14:01:48Z)
• Alpaca against Vicuna: Using LLMs to Uncover Memorization of LLMs [61.04246774006429]
  We introduce a black-box prompt optimization method that uses an attacker LLM agent to uncover higher levels of memorization in a victim agent. We observe that our instruction-based prompts generate outputs with 23.7% higher overlap with training data compared to the baseline prefix-suffix measurements. Our findings show that instruction-tuned models can expose pre-training data as much as their base-models, if not more so, and using instructions proposed by other LLMs can open a new avenue of automated attacks.
  arXiv  Detail & Related papers  (2024-03-05T19:32:01Z)
• Towards Efficient Active Learning in NLP via Pretrained Representations [1.90365714903665]
  Fine-tuning Large Language Models (LLMs) is now a common approach for text classification in a wide range of applications. We drastically expedite this process by using pretrained representations of LLMs within the active learning loop. Our strategy yields similar performance to fine-tuning all the way through the active learning loop but is orders of magnitude less computationally expensive.
  arXiv  Detail & Related papers  (2024-02-23T21:28:59Z)
• Efficient Grammatical Error Correction Via Multi-Task Training and Optimized Training Schedule [55.08778142798106]
  We propose auxiliary tasks that exploit the alignment between the original and corrected sentences. We formulate each task as a sequence-to-sequence problem and perform multi-task training. We find that the order of datasets used for training and even individual instances within a dataset may have important effects on the final performance.
  arXiv  Detail & Related papers  (2023-11-20T14:50:12Z)
• Model-Agnostic Multitask Fine-tuning for Few-shot Vision-Language Transfer Learning [59.38343286807997]
  We propose Model-Agnostic Multitask Fine-tuning (MAMF) for vision-language models on unseen tasks. Compared with model-agnostic meta-learning (MAML), MAMF discards the bi-level optimization and uses only first-order gradients. We show that MAMF consistently outperforms the classical fine-tuning method for few-shot transfer learning on five benchmark datasets.
  arXiv  Detail & Related papers  (2022-03-09T17:26:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.