Related papers: BEAST: Efficient Tokenization of B-Splines Encoded…

Masked image modeling (MIM) has demonstrated impressive results in self-supervised representation learning by recovering corrupted image patches. However, most existing studies operate on low-level image pixels, which hinders the…

Scene Text Recognition (STR) is the problem of recognizing the correct word or character sequence in a cropped word image. To obtain more robust output sequences, the notion of bidirectional STR has been introduced. So far, bidirectional…

Image tokenization plays a critical role in reducing the computational demands of modeling high-resolution images, significantly improving the efficiency of image and multimodal understanding and generation. Recent advances in 1D latent…

Named entity recognition (NER) is a widely studied task in natural language processing. Recently, a growing number of studies have focused on the nested NER. The span-based methods, considering the entity recognition as a span…

Invasive cortical brain-machine interfaces (BMIs) can significantly improve the life quality of motor-impaired patients. Nonetheless, externally mounted pedestals pose an infection risk, which calls for fully implanted systems. Such…

Although large vision-language models (LVLMs) leverage rich visual token representations to achieve strong performance on multimodal tasks, these tokens also introduce significant computational overhead during inference. Existing…

Masked image modeling (MIM) has become a leading self-supervised learning strategy. MIMs such as Masked Autoencoder (MAE) learn strong representations by randomly masking input tokens for the encoder to process, with the decoder…

Under the multispecies coalescent model of molecular evolution, gene trees have independent evolutionary histories within a shared species tree. In comparison, supermatrix concatenation methods assume that gene trees share a single common…

Fine-grained sparsity promises higher parametric capacity without proportional per-token compute, but often suffers from training instability, load balancing, and communication overhead. We introduce STEM (Scaling Transformers with…

This paper presents an efficient multi-scale vision Transformer, called ResT, that capably served as a general-purpose backbone for image recognition. Unlike existing Transformer methods, which employ standard Transformer blocks to tackle…

Discrete audio tokenizers are fundamental to empowering large language models with native audio processing and generation capabilities. Despite recent progress, existing approaches often rely on pretrained encoders, semantic distillation,…

Animal behavior analysis plays a crucial role in various fields, such as life science and biomedical research. However, the scarcity of available data and the high cost associated with obtaining a large number of labeled datasets pose…

We present a unified, promptable model capable of simultaneously segmenting, recognizing, and captioning anything. Unlike SAM, we aim to build a versatile region representation in the wild via visual prompting. To achieve this, we train a…

Subword tokenization methods, such as Byte-Pair Encoding (BPE), significantly impact the performance and efficiency of large language models (LLMs). The standard approach involves training a general-purpose tokenizer that uniformly…

World models provide a powerful framework for simulating environment dynamics conditioned on actions or instructions, enabling downstream tasks such as action planning or policy learning. Recent approaches leverage world models as learned…

Tokenization is a critical preprocessing step in large language model pipelines, yet widely-used implementations remain CPU-bound and suboptimal for batch inference workflows on GPU. We present BlockBPE, a parallel GPU implementation of…

We introduce a new tokenizer for language models that minimizes the average tokens per character, thereby reducing the number of tokens needed to represent text during training and to generate text during inference. Our method, which we…

Achieving robust, human-like whole-body control on humanoid robots for agile, contact-rich behaviors remains a central challenge, demanding heavy per-skill engineering and a brittle process of tuning controllers. We introduce ZEST…

Quantization is the key method for reducing inference latency, power and memory footprint of generative AI models. However, accuracy often degrades sharply when activations are quantized below eight bits. Recent work suggests that…

This paper presents LongCat-Audio-Codec, an audio tokenizer and detokenizer solution designed for industrial grade end-to-end speech large language models. By leveraging a decoupled model architecture and a multistage training strategy,…