ReGATE: Learning Faster and Better with Fewer Tokens in MLLMs

• URL: http://arxiv.org/abs/2507.21420v1
• Date: Tue, 29 Jul 2025 01:07:09 GMT
• Title: ReGATE: Learning Faster and Better with Fewer Tokens in MLLMs
• Authors: Chaoyu Li, Yogesh Kulkarni, Pooyan Fazli,
• Abstract summary: ReGATE (Reference$-$Guided Adaptive Token Elision) is an adaptive token pruning method for accelerating MLLM training.<n>It matches the peak accuracy of standard training on MVBench up to 2$times$ faster, using only 35% of the tokens.
• Score: 1.1834200163382398
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The computational cost of training multimodal large language models (MLLMs) rapidly increases with the number of tokens involved. Existing efficiency methods primarily target inference and rely on token reduction or merging, offering limited benefit during training. In this paper, we propose ReGATE (Reference$-$Guided Adaptive Token Elision), an adaptive token pruning method for accelerating MLLM training. Specifically, ReGATE adopts a teacher-student framework in which the MLLM being trained serves as the student, and a frozen reference large language model (LLM) acts as the teacher. The teacher computes per-token reference losses, which are combined with an exponential moving average (EMA) of the student's own difficulty scores. This adaptive difficulty-based scoring enables the selective processing of crucial tokens while bypassing less informative ones in the forward pass, significantly reducing computational overhead. Experiments demonstrate that ReGATE, when applied to VideoLLaMA2, matches the peak accuracy of standard training on MVBench up to 2$\times$ faster, using only 35% of the tokens. With additional training, it even surpasses the baseline on several multimodal benchmarks, all while reducing the total token count by over 41%. Code and models will be released soon.

Related papers

• ESLM: Risk-Averse Selective Language Modeling for Efficient Pretraining [53.893792844055106]
  Large language model pretraining is compute-intensive, yet many tokens contribute marginally to learning, resulting in inefficiency.<n>We introduce Selective Efficient Language Modeling, a risk-aware algorithm that improves training efficiency and distributional robustness by performing online token-level batch selection.<n> Experiments on GPT-2 pretraining show that ESLM significantly reduces training FLOPs while maintaining or improving both perplexity and downstream performance compared to baselines.
  arXiv  Detail & Related papers  (2025-05-26T12:23:26Z)
• freePruner: A Training-free Approach for Large Multimodal Model Acceleration [23.561529800086454]
  freePruner is a training-free token reduction approach that can be directly applied to any open-source LMM without additional training. Experiments demonstrate that freePruner achieves 2x acceleration while maintaining comparable performance across mainstream visual question-answering benchmarks.
  arXiv  Detail & Related papers  (2024-11-23T04:25:16Z)
• Context-aware Prompt Tuning: Advancing In-Context Learning with Adversarial Methods [69.36397993451742]
  This work introduces Context-aware Prompt Tuning (CPT), a method inspired by ICL, PT, and adversarial attacks. We modify specific context tokens, considering the unique structure of input and output formats. Inspired by adversarial attacks, we adjust the input based on the labels present in the context, focusing on minimizing, rather than maximizing, the loss.
  arXiv  Detail & Related papers  (2024-10-22T17:45:47Z)
• Beyond Next Token Prediction: Patch-Level Training for Large Language Models [69.67438563485887]
  We introduce patch-level training for Large Language Models (LLMs)<n>During patch-level training, we feed the language model shorter sequences of patches and train it to predict the next patch.<n>We show that patch-level training can reduce the overall training costs to 0.5$times$, without compromising the model performance.
  arXiv  Detail & Related papers  (2024-07-17T15:48:39Z)
• Getting the most out of your tokenizer for pre-training and domain adaptation [26.427537023771844]
  We show that the size, pre-tokenization regular expression, and training data of a tokenizer can significantly impact the model's generation speed. We specialize the tokenizer of a pre-trained LLM to obtain large gains in generation speed and effective context size.
  arXiv  Detail & Related papers  (2024-02-01T21:49:34Z)
• Accelerating Vision-Language Pretraining with Free Language Modeling [62.30042851111692]
  Free language modeling (FLM) enables a 100% prediction rate with arbitrary corruption rates. FLM successfully frees the prediction rate from the tie-up with the corruption rate. Experiments show FLM could achieve an impressive 2.5x pretraining time reduction.
  arXiv  Detail & Related papers  (2023-03-24T14:49:22Z)
• MC-BERT: Efficient Language Pre-Training via a Meta Controller [96.68140474547602]
  Large-scale pre-training is computationally expensive. ELECTRA, an early attempt to accelerate pre-training, trains a discriminative model that predicts whether each input token was replaced by a generator. We propose a novel meta-learning framework, MC-BERT, to achieve better efficiency and effectiveness.
  arXiv  Detail & Related papers  (2020-06-10T09:22:19Z)
• ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators [108.3381301768299]
  Masked language modeling (MLM) pre-training methods such as BERT corrupt the input by replacing some tokens with [MASK] and then train a model to reconstruct the original tokens. We propose a more sample-efficient pre-training task called replaced token detection.
  arXiv  Detail & Related papers  (2020-03-23T21:17:42Z)

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.