paper-with-me

홈 › Papers

A Study of Autoregressive Decoders for Multi-Tasking in Computer Vision

2023-03-30 · Lucas Beyer, Bo Wan, Gagan Madan, Filip Pavetic, Andreas Steiner, Alexander Kolesnikov, André Susano Pinto, Emanuele Bugliarello, Xiao Wang, Qihang Yu, Liang-Chieh Chen, Xiaohua Zhai

There has been a recent explosion of computer vision models which perform many tasks and are composed of an image encoder (usually a ViT) and an autoregressive decoder (usually a Transformer). However, most of this work simply presents one system and its results, leaving many questions regarding design decisions and trade-offs of such systems unanswered. In this work, we aim to provide such answers. We take a close look at autoregressive decoders for multi-task learning in multimodal computer vision, including classification, captioning, visual question answering, and optical character recognition. Through extensive systematic experiments, we study the effects of task and data mixture, training and regularization hyperparameters, conditioning type and specificity, modality combination, and more. Importantly, we compare these to well-tuned single-task baselines to highlight the cost incurred by multi-tasking. A key finding is that a small decoder learned on top of a frozen pretrained encoder works surprisingly well. We call this setup locked-image tuning with decoder (LiT-decoder). It can be seen as teaching a decoder to interact with a pretrained vision model via natural language.

📄 PDF Abstract BibTeX arXiv:2303.17376

Code (1)

google-research/big_vision 공식 구현 jax

Tasks

DecoderMulti-Task LearningOptical Character RecognitionQuestion AnsweringSpecificityVisual Question Answering

Similar Papers 제목 키워드 기반

MMTM: Multi-Tasking Multi-Decoder Transformer for Math Word Problems

2022-06-02 · Keyur Faldu, Amit Sheth, Prashant Kikani, Darshan Patel

Recently, quite a few novel neural architectures were derived to solve math word problems by predicting expression trees. These architectures varied from seq2seq models, including encoders leveraging graph relationships …

DecoderMathMathematical Reasoning

Hyperdecoders: Instance-specific decoders for multi-task NLP

2022-03-15 · Hamish Ivison, Matthew E. Peters

We investigate input-conditioned hypernetworks for multi-tasking in NLP, generating parameter-efficient adaptations for a decoder using a hypernetwork conditioned on the output of an encoder. This approach produces a uni…

Decoderparameter-efficient fine-tuning

Natural Language Multitasking: Analyzing and Improving Syntactic Saliency of Hidden Representations

2018-01-18 · Gino Brunner, Yuyi Wang, Roger Wattenhofer, Michael Weigelt

We train multi-task autoencoders on linguistic tasks and analyze the learned hidden sentence representations. The representations change significantly when translation and part-of-speech decoders are added. The more deco…

SentenceTranslation

SkexGen: Autoregressive Generation of CAD Construction Sequences with Disentangled Codebooks

2022-07-11 · Xiang Xu, Karl D. D. Willis, Joseph G. Lambourne, Chin-Yi Cheng 외

We present SkexGen, a novel autoregressive generative model for computer-aided design (CAD) construction sequences containing sketch-and-extrude modeling operations. Our model utilizes distinct Transformer architectures …

Efficient Exploration

BrainFLORA: Uncovering Brain Concept Representation via Multimodal Neural Embeddings

2025-07-13 · Dongyang Li, Haoyang Qin, Mingyang Wu, Chen Wei 외 arxiv

Understanding how the brain represents visual information is a fundamental challenge in neuroscience and artificial intelligence. While AI-driven decoding of neural data has provided insights into the human visual system…