Enabling Autoregressive Models to Fill In Masked Tokens

• URL: http://arxiv.org/abs/2502.06901v1
• Date: Sun, 09 Feb 2025 20:02:05 GMT
• Title: Enabling Autoregressive Models to Fill In Masked Tokens
• Authors: Daniel Israel, Aditya Grover, Guy Van den Broeck,
• Abstract summary: This work introduces MARIA (Masked and Autoregressive Infilling Architecture), a novel approach that achieves state-of-the-art masked infilling performance. MARIA combines a pre-trained and AR model by training a linear decoder that takes their hidden states as input. Our results demonstrate that MARIA significantly outperforms existing methods, namely discrete diffusion models, on masked infilling tasks.
• Score: 50.9948753314669
• License:
• Abstract: Historically, LLMs have been trained using either autoregressive (AR) or masked language modeling (MLM) objectives, with AR models gaining dominance in recent years. However, AR models are inherently incapable of masked infilling, which is the ability to predict masked tokens between past and future context. In contrast, MLM models suffer from intrinsic computational inefficiencies during both training and inference that hinder their scalability. This work introduces MARIA (Masked and Autoregressive Infilling Architecture), a novel approach that leverages the strengths of both paradigms to achieve state-of-the-art masked infilling performance. MARIA combines a pre-trained MLM and AR model by training a linear decoder that takes their concatenated hidden states as input. This minimal modification enables the AR model to perform infilling while retaining its inherent advantages in terms of faster inference with KV caching. Our results demonstrate that MARIA significantly outperforms existing methods, namely discrete diffusion models, on masked infilling tasks.

Related papers

• Large Language Diffusion Models [77.02553707673418]
  Autoregressive models (ARMs) are widely regarded as the cornerstone of large language models (LLMs) We introduce LLaDA, a diffusion model trained from scratch under the pre-training and supervised fine-tuning paradigm. Across extensive benchmarks, LLaDA demonstrates strong scalability, outperforming our self-constructed ARM baselines.
  arXiv  Detail & Related papers  (2025-02-14T08:23:51Z)
• Beyond Autoregression: Fast LLMs via Self-Distillation Through Time [1.5858234832499587]
  We show that diffusion language models are capable of generating at least 32 tokens simultaneously, while exceeding the performance of AR models in text quality and on the LAMBADA natural language understanding benchmark. Practically, at the 1.3B parameters scale, diffusion models, even without caching, can generate tokens at a rate that is up to 8 times faster than AR models employing KV-caching.
  arXiv  Detail & Related papers  (2024-10-28T13:56:30Z)
• Scaling Diffusion Language Models via Adaptation from Autoregressive Models [105.70889434492143]
  Diffusion Language Models (DLMs) have emerged as a promising new paradigm for text generative modeling. We show that we can convert AR models ranging from 127M to 7B parameters into diffusion models DiffuGPT and DiffuLLaMA, using less than 200B tokens for training. Our experimental results reveal that these models outperform earlier DLMs and are competitive with their AR counterparts.
  arXiv  Detail & Related papers  (2024-10-23T14:04:22Z)
• LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding [30.630803933771865]
  Experimental results demonstrate the efficacy of our method in providing a substantial speed-up over speculative decoding. LANTERN increases speed-ups by $mathbf1.75times$ and $mathbf1.82times$, as compared to greedy decoding and random sampling.
  arXiv  Detail & Related papers  (2024-10-04T12:21:03Z)
• Emerging Property of Masked Token for Effective Pre-training [15.846621577804791]
  Masked Image Modeling (MIM) has been instrumental in driving recent breakthroughs in computer vision. MIM's overall efficiency is occasionally hampered by the lengthy duration of the pre-training phase. We propose a novel approach termed masked token optimization (MTO), specifically designed to improve model efficiency through weight recalibration and the enhancement of the key property of masked tokens.
  arXiv  Detail & Related papers  (2024-04-12T08:46:53Z)
• Masked Autoencoding for Scalable and Generalizable Decision Making [93.84855114717062]
  MaskDP is a simple and scalable self-supervised pretraining method for reinforcement learning and behavioral cloning. We find that a MaskDP model gains the capability of zero-shot transfer to new BC tasks, such as single and multiple goal reaching.
  arXiv  Detail & Related papers  (2022-11-23T07:04:41Z)
• Scaling Hidden Markov Language Models [118.55908381553056]
  This work revisits the challenge of scaling HMMs to language modeling datasets. We propose methods for scaling HMMs to massive state spaces while maintaining efficient exact inference, a compact parameterization, and effective regularization.
  arXiv  Detail & Related papers  (2020-11-09T18:51:55Z)
• Semi-Autoregressive Training Improves Mask-Predict Decoding [119.8412758943192]
  We introduce a new training method for conditional masked language models, SMART, which mimics the semi-autoregressive behavior of mask-predict. Models trained with SMART produce higher-quality translations when using mask-predict decoding, effectively closing the remaining performance gap with fully autoregressive models.
  arXiv  Detail & Related papers  (2020-01-23T19:56:35Z)

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.