Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference

• URL: http://arxiv.org/abs/2412.13663v2
• Date: Thu, 19 Dec 2024 06:32:26 GMT
• Title: Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference
• Authors: Benjamin Warner, Antoine Chaffin, Benjamin Clavié, Orion Weller, Oskar Hallström, Said Taghadouini, Alexis Gallagher, Raja Biswas, Faisal Ladhak, Tom Aarsen, Nathan Cooper, Griffin Adams, Jeremy Howard, Iacopo Poli,
• Abstract summary: We introduce ModernBERT, bringing modern model optimizations to encoder-only models.<n>ModernBERT models exhibit state-of-the-art results on a large pool of evaluations.<n>ModernBERT is also the most speed and memory efficient encoder.
• Score: 15.921261060193416
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Encoder-only transformer models such as BERT offer a great performance-size tradeoff for retrieval and classification tasks with respect to larger decoder-only models. Despite being the workhorse of numerous production pipelines, there have been limited Pareto improvements to BERT since its release. In this paper, we introduce ModernBERT, bringing modern model optimizations to encoder-only models and representing a major Pareto improvement over older encoders. Trained on 2 trillion tokens with a native 8192 sequence length, ModernBERT models exhibit state-of-the-art results on a large pool of evaluations encompassing diverse classification tasks and both single and multi-vector retrieval on different domains (including code). In addition to strong downstream performance, ModernBERT is also the most speed and memory efficient encoder and is designed for inference on common GPUs.

Related papers

• Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation [129.45368843861917]
  We introduce the Gated Memory Unit (GMU), a simple yet effective mechanism for efficient memory sharing across layers.<n>We apply it to create SambaY, a decoder-hybrid-decoder architecture that incorporates GMUs to share memory readout states from a Samba-based self-decoder.
  arXiv  Detail & Related papers  (2025-07-09T07:27:00Z)
• ModernGBERT: German-only 1B Encoder Model Trained from Scratch [3.193989599110687]
  We introduce ModernGBERT (134M, 1B), a fully transparent family of German encoder models trained from scratch.<n>We also present LL"aMmlein2Vec (120M, 1B, 7B), a family of encoders derived from German decoder-only models via LLM2Vec.
  arXiv  Detail & Related papers  (2025-05-19T14:07:20Z)
• Encoder-Decoder Gemma: Improving the Quality-Efficiency Trade-Off via Adaptation [52.19855651708349]
  We study a novel problem: adapting decoder-only large language models to encoder-decoder models. We argue that adaptation not only enables inheriting the capability of decoder-only LLMs but also reduces the demand for computation. Under similar inference budget, encoder-decoder LLMs achieve comparable (often better) pretraining performance but substantially better finetuning performance than their decoder-only counterpart.
  arXiv  Detail & Related papers  (2025-04-08T17:13:41Z)
• Efficient Encoder-Decoder Transformer Decoding for Decomposable Tasks [53.550782959908524]
  We introduce a new configuration for encoder-decoder models that improves efficiency on structured output and decomposable tasks. Our method, prompt-in-decoder (PiD), encodes the input once and decodes the output in parallel, boosting both training and inference efficiency.
  arXiv  Detail & Related papers  (2024-03-19T19:27:23Z)
• BPDec: Unveiling the Potential of Masked Language Modeling Decoder in BERT pretraining [0.5919433278490629]
  BERT (Bidirectional Representations from Transformers) has revolutionized the field of natural language processing through its exceptional performance on numerous tasks.<n>DeBERTa introduced an enhanced decoder adapted for BERT's encoder model for pretraining, proving to be highly effective.<n>We argue that the design and research around enhanced masked language modeling decoders have been underappreciated.
  arXiv  Detail & Related papers  (2024-01-29T03:25:11Z)
• MosaicBERT: A Bidirectional Encoder Optimized for Fast Pretraining [10.421048804389343]
  We introduce MosaicBERT, a BERT-style encoder architecture and training recipe that is empirically optimized for fast pretraining. When pretrained from scratch on the C4 dataset, this base model achieves a downstream average GLUE (dev) score of 79.6 in 1.13 hours on 8 A100 80 GB GPUs at a cost of roughly $20. This empirical speed up in pretraining enables researchers and engineers to pretrain custom BERT-style models at low cost instead of finetune on existing generic models.
  arXiv  Detail & Related papers  (2023-12-29T06:05:19Z)
• Stateful Conformer with Cache-based Inference for Streaming Automatic Speech Recognition [20.052245837954175]
  We propose an efficient and accurate streaming speech recognition model based on the FastConformer architecture. We introduce an activation caching mechanism to enable the non-autoregressive encoder to operate autoregressively during inference. A hybrid CTC/RNNT architecture which utilizes a shared encoder with both a CTC and RNNT decoder to boost the accuracy and save computation.
  arXiv  Detail & Related papers  (2023-12-27T21:04:26Z)
• Practical Conformer: Optimizing size, speed and flops of Conformer for on-Device and cloud ASR [67.63332492134332]
  We design an optimized conformer that is small enough to meet on-device restrictions and has fast inference on TPUs. Our proposed encoder can double as a strong standalone encoder in on device, and as the first part of a high-performance ASR pipeline.
  arXiv  Detail & Related papers  (2023-03-31T23:30:48Z)
• NarrowBERT: Accelerating Masked Language Model Pretraining and Inference [50.59811343945605]
  We propose NarrowBERT, a modified transformer encoder that increases the throughput for masked language model pretraining by more than $2times$. NarrowBERT sparsifies the transformer model such that the self-attention queries and feedforward layers only operate on the masked tokens of each sentence during pretraining. We show that NarrowBERT increases the throughput at inference time by as much as $3.5times$ with minimal (or no) performance degradation on sentence encoding tasks like MNLI.
  arXiv  Detail & Related papers  (2023-01-11T23:45:50Z)
• An Exploration of Hierarchical Attention Transformers for Efficient Long Document Classification [37.069127262896764]
  Non-hierarchical sparse attention Transformer-based models, such as Longformer and Big Bird, are popular approaches to working with long documents. We develop and release fully pre-trained HAT models that use segment-wise followed by cross-segment encoders. Our best HAT model outperforms equally-sized Longformer models while using 10-20% less GPU memory and processing documents 40-45% faster.
  arXiv  Detail & Related papers  (2022-10-11T15:17:56Z)
• ED2LM: Encoder-Decoder to Language Model for Faster Document Re-ranking Inference [70.36083572306839]
  This paper proposes a new training and inference paradigm for re-ranking. We finetune a pretrained encoder-decoder model using in the form of document to query generation. We show that this encoder-decoder architecture can be decomposed into a decoder-only language model during inference.
  arXiv  Detail & Related papers  (2022-04-25T06:26:29Z)
• BinaryBERT: Pushing the Limit of BERT Quantization [74.65543496761553]
  We propose BinaryBERT, which pushes BERT quantization to the limit with weight binarization. We find that a binary BERT is hard to be trained directly than a ternary counterpart due to its complex and irregular loss landscapes. Empirical results show that BinaryBERT has negligible performance drop compared to the full-precision BERT-base.
  arXiv  Detail & Related papers  (2020-12-31T16:34:54Z)
• HyperSeg: Patch-wise Hypernetwork for Real-time Semantic Segmentation [95.47168925127089]
  We present a novel, real-time, semantic segmentation network in which the encoder both encodes and generates the parameters (weights) of the decoder. We design a new type of hypernetwork, composed of a nested U-Net for drawing higher level context features.
  arXiv  Detail & Related papers  (2020-12-21T18:58:18Z)

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.