dParallel is a learnable parallel decoding method for large language models that achieves fast sampling by exploiting the intrinsic parallelism of the model. This method can significantly reduce the decoding steps while maintaining the model performance, achieving 8.5 - 10.5 times acceleration in benchmarks such as GSM8K and MBPP.

Natural Language ProcessingTransformers

dParallel is a learnable parallel decoding method for large language models. It achieves fast sampling by exploring the intrinsic parallelism of the model, significantly reducing the decoding steps while maintaining performance, and achieving 8.5 - 10.5 times acceleration in benchmark tests such as GSM8K and MBPP.

Natural Language ProcessingTransformers

A single-image intrinsic image decomposition model fine-tuned based on Stable Diffusion, capable of decomposing images into albedo, diffuse shading, and non-diffuse residual components

Computer VisionDiffusersEnglish

By fine-tuning text-to-image models (such as FLUX), it can generate groups of images with custom intrinsic associations and supports cross-group conditional generation based on SDEdit. This technology can adapt to various task scenarios.

Computer VisionDiffusers

Generates customizable image sets with intrinsic relationships through fine-tuned text-to-image models, supporting multiple task scenarios

Computer VisionDiffusers

IC-LoRA is a framework for generating image groups with custom intrinsic associations by fine-tuning text-to-image models (such as FLUX), and it supports conditional generation through SDEdit.

MultimodalDiffusers

A single-image intrinsic decomposition model fine-tuned based on Stable Diffusion 2, used to decompose images into albedo, roughness, and metallic properties

Computer VisionDiffusersEnglish

Zero-shot monocular ranging depth estimation foundation model capable of synthesizing high-resolution depth maps with absolute scale prediction independent of camera intrinsics

SpeechGPT is a large language model with intrinsic cross-modal dialogue capabilities, capable of perceiving and generating multimodal content, supporting interaction via speech and text.

MultimodalTransformers

SpeechGPT is a large language model with intrinsic cross-modal dialogue capabilities, capable of perceiving and generating multimodal content based on human instructions.

MultimodalTransformers

A Transformer model pre-trained on English corpus using ELECTRA-like objective functions, learning intrinsic representations of English language through self-supervised methods.

Natural Language ProcessingTransformersEnglish