Related papers: iBOT: Image BERT Pre-Training with Online Tokenize…

In recent times, the standard practice for developing MLLMs is to feed features from vision encoder(s) into the LLM and train with natural language supervision. This approach often causes models to lean towards language comprehension and…

Masked language models (MLMs) such as BERT and RoBERTa have revolutionized the field of Natural Language Understanding in the past few years. However, existing pre-trained MLMs often output an anisotropic distribution of token…

Image segmentation is often ambiguous at the level of individual image patches and requires contextual information to reach label consensus. In this paper we introduce Segmenter, a transformer model for semantic segmentation. In contrast to…

We introduce a vision-language foundation model called VL-BEiT, which is a bidirectional multimodal Transformer learned by generative pretraining. Our minimalist solution conducts masked prediction on both monomodal and multimodal data with…

Masked language modeling, widely used in discriminative language model (e.g., BERT) pretraining, commonly adopts a random masking strategy. However, random masking does not consider the importance of the different words in the sentence…

The rapid success of Vision Large Language Models (VLLMs) often depends on the high-resolution images with abundant visual tokens, which hinders training and deployment efficiency. Current training-free visual token compression methods…

In recent years, general visual foundation models (VFMs) have witnessed increasing adoption, particularly as image encoders for popular multi-modal large language models (MLLMs). However, without semantically fine-grained supervision, these…

Despite interpretability work analyzing VIT encoders and transformer activations, we don't yet understand why Multimodal Language Models (MLMs) struggle on perception-heavy tasks. We offer an under-studied perspective by examining how…

Pre-training has been a useful method for learning implicit transferable knowledge and it shows the benefit of offering complementary features across different modalities. Recent work mainly focuses on the modalities such as image and text,…

This paper represents a neat yet effective framework, named SemanticMIM, to integrate the advantages of masked image modeling (MIM) and contrastive learning (CL) for general visual representation. We conduct a thorough comparative analysis…

Choosing an appropriate tokenization scheme is often a bottleneck in low-resource cross-lingual transfer. To understand the downstream implications of text representation choices, we perform a comparative analysis on language models having…

Masked Language Models (MLMs) have achieved remarkable success in many self-supervised representation learning tasks. MLMs are trained by randomly masking portions of the input sequences with [MASK] tokens and learning to reconstruct the…

Visual captioning aims to generate textual descriptions given images or videos. Traditionally, image captioning models are trained on human annotated datasets such as Flickr30k and MS-COCO, which are limited in size and diversity. This…

In this work, we present a novel method, named AV2vec, for learning audio-visual speech representations by multimodal self-distillation. AV2vec has a student and a teacher module, in which the student performs a masked latent feature…

Advancements in language foundation models have primarily fueled the recent surge in artificial intelligence. In contrast, generative learning of non-textual modalities, especially videos, significantly trails behind language modeling. This…

Non-overlapping patch-wise convolution is the default image tokenizer for all state-of-the-art vision Transformer (ViT) models. Even though many ViT variants have been proposed to improve its efficiency and accuracy, little research on…

Tokenization significantly influences language models(LMs)' performance. This paper traces the evolution of tokenizers from word-level to subword-level, analyzing how they balance tokens and types to enhance model adaptability while…

Masked language modeling (MLM) plays a key role in pretraining large language models. But the MLM objective is often dominated by high-frequency words that are sub-optimal for learning factual knowledge. In this work, we propose an approach…

SimMIM is a widely used method for pretraining vision transformers using masked image modeling. However, despite its success in fine-tuning performance, it has been shown to perform sub-optimally when used for linear probing. We propose an…

In this paper, we introduce ILLUME, a unified multimodal large language model (MLLM) that seamlessly integrates multimodal understanding and generation capabilities within a single large language model through a unified next-token…