paper-with-me

Papers

Faster Depth-Adaptive Transformers

2020-04-27 · Yijin Liu, Fandong Meng, Jie zhou, Yufeng Chen, Jinan Xu

Depth-adaptive neural networks can dynamically adjust depths according to the hardness of input words, and thus improve efficiency. The main challenge is how to measure such hardness and decide the required depths (i.e., layers) to conduct. Previous works generally build a halting unit to decide whether the computation should continue or stop at each layer. As there is no specific supervision of depth selection, the halting unit may be under-optimized and inaccurate, which results in suboptimal and unstable performance when modeling sentences. In this paper, we get rid of the halting unit and estimate the required depths in advance, which yields a faster depth-adaptive model. Specifically, two approaches are proposed to explicitly measure the hardness of input words and estimate corresponding adaptive depth, namely 1) mutual information (MI) based estimation and 2) reconstruction loss based estimation. We conduct experiments on the text classification task with 24 datasets in various sizes and domains. Results confirm that our approaches can speed up the vanilla Transformer (up to 7x) while preserving high accuracy. Moreover, efficiency and robustness are significantly improved when compared with other depth-adaptive approaches.

📄 PDF Abstract BibTeX arXiv:2004.13542

Code (0)

등록된 구현이 없습니다.

Tasks

Sentence Embeddingstext-classificationText Classification

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
SPEED The monocular depth estimation (MDE) is the task of estimating depth from a single frame. This information is an essential knowledge in many computer vision tasks such as scene…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음
Interpretability 설명 없음
Residual Connection 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…

Similar Papers 제목 키워드 기반

Image Complexity-Aware Adaptive Retrieval for Efficient Vision-Language Models

2025-12-17 · Mikel Williams-Lekuona, Georgina Cosma arxiv

Vision transformers in vision-language models typically use the same amount of compute for every image, regardless of whether it is simple or complex. We propose ICAR (Image Complexity-Aware Retrieval), an adaptive compu…

Image-text matching

TeamFormer: Shallow Parallel Transformers with Progressive Approximation

2025-10-17 · Wei Wang, Xiao-Yong Wei, Qing Li arxiv

The widespread 'deeper is better' philosophy has driven the creation of architectures like ResNet and Transformer, which achieve high performance by stacking numerous layers. However, increasing model depth comes with ch…

Model Compression

Do Transformers Use their Depth Adaptively? Evidence from a Relational Reasoning Task

2026-04-14 · Alicia Curth, Rachel Lawrence, Sushrut Karmalkar, Niranjani Prasad arxiv

We investigate whether transformers use their depth adaptively across tasks of increasing difficulty. Using a controlled multi-hop relational reasoning task based on family stories, where difficulty is determined by the …

Relational Reasoning

Adaptive inference and function vectors in deep transformers

2026-06-15 · Ravin Raj, Gautam Reddy arxiv

Transformers are widely used as a general-purpose substrate for learning complex correlations between a large collection of coupled variables, but their internal mechanisms have remained mysterious. We introduce a theory…

Spectrum-Adaptive Generalization Bounds for Trained Deep Transformers

2026-05-08 · Mana Sakai, Masaaki Imaizumi arxiv

Understanding why trained Transformers generalize well is a fundamental problem in modern machine learning theory, and complexity-based generalization bounds provide a principled way to study this question. While existin…