Low-Rank Adaptation with Task-Relevant Feature Enhancement for Fine-tuning Language Models

Fine-tuning pre-trained large language models in a parameter-efficient manner is widely studied for its effectiveness and efficiency. LoRA is one of the most widely used methods, which assumes that the optimization process is essentially low dimensional. Although LoRA has demonstrated commendable performance, there remains a significant performance gap between LoRA and full fine-tuning when learning new tasks. In this work, we propose Low-Rank Adaptation with Task-Relevant Feature Enhancement(LoRATRF) for enhancing task-relevant features from the perspective of editing neural network representations. To prioritize task-relevant features, a task-aware filter that selectively extracts valuable knowledge from hidden representations for the target or current task is designed. As the experiments on a vareity of datasets including NLU, commonsense reasoning and mathematical reasoning tasks demonstrates, our method reduces 33.71% parameters and achieves better performance on a variety of datasets in comparison with SOTA low-rank methods.

YingSound: Video-Guided Sound Effects Generation with Multi-modal Chain-of-Thought Controls

Generating sound effects for product-level videos, where only a small amount of labeled data is available for diverse scenes, requires the production of high-quality sounds in few-shot settings. To tackle the challenge of limited labeled data in real-world scenes, we introduce YingSound, a foundation model designed for video-guided sound generation that supports high-quality audio generation in few-shot settings. Specifically, YingSound consists of two major modules. The first module uses a conditional flow matching transformer to achieve effective semantic alignment in sound generation across audio and visual modalities. This module aims to build a learnable audio-visual aggregator (AVA) that integrates high-resolution visual features with corresponding audio features at multiple stages. The second module is developed with a proposed multi-modal visual-audio chain-of-thought (CoT) approach to generate finer sound effects in few-shot settings. Finally, an industry-standard video-to-audio (V2A) dataset that encompasses various real-world scenarios is presented. We show that YingSound effectively generates high-quality synchronized sounds across diverse conditional inputs through automated evaluations and human studies. Project Page: \url{https://giantailab.github.io/yingsound/}

Multi-Stage Graph Learning for fMRI Analysis to Diagnose Neuro-Developmental Disorders

The insufficient supervision limit the performance of the deep supervised models for brain disease diagnosis. It is important to develop a learning framework that can capture more information in limited data and insufficient supervision. To address these issues at some extend, we propose a multi-stage graph learning framework which incorporates 1) pretrain stage : self-supervised graph learning on insufficient supervision of the fmri data 2) fine-tune stage : supervised graph learning for brain disorder diagnosis. Experiment results on three datasets, Autism Brain Imaging Data Exchange ABIDE I, ABIDE II and ADHD with AAL1,demonstrating the superiority and generalizability of the proposed framework compared to the state of art of models.(ranging from 0.7330 to 0.9321,0.7209 to 0.9021,0.6338 to 0.6699)

OCC-MLLM-Alpha:Empowering Multi-modal Large Language Model for the Understanding of Occluded Objects with Self-Supervised Test-Time Learning

There is a gap in the understanding of occluded objects in existing large-scale visual language multi-modal models. Current state-of-the-art multi-modal models fail to provide satisfactory results in describing occluded objects through universal visual encoders and supervised learning strategies. Therefore, we introduce a multi-modal large language framework and corresponding self-supervised learning strategy with support of 3D generation. We start our experiments comparing with the state-of-the-art models in the evaluation of a large-scale dataset SOMVideo [18]. The initial results demonstrate the improvement of 16.92% in comparison with the state-of-the-art VLM models.

OCC-MLLM:Empowering Multimodal Large Language Model For the Understanding of Occluded Objects

There is a gap in the understanding of occluded objects in existing large-scale visual language multi-modal models. Current state-of-the-art multimodal models fail to provide satisfactory results in describing occluded objects for visual-language multimodal models through universal visual encoders. Another challenge is the limited number of datasets containing image-text pairs with a large number of occluded objects. Therefore, we introduce a novel multimodal model that applies a newly designed visual encoder to understand occluded objects in RGB images. We also introduce a large-scale visual-language pair dataset for training large-scale visual-language multimodal models and understanding occluded objects. We start our experiments comparing with the state-of-the-art models.

Towards Full-parameter and Parameter-efficient Self-learning For Endoscopic Camera Depth Estimation

Adaptation methods are developed to adapt depth foundation models to endoscopic depth estimation recently. However, such approaches typically under-perform training since they limit the parameter search to a low-rank subspace and alter the training dynamics. Therefore, we propose a full-parameter and parameter-efficient learning framework for endoscopic depth estimation. At the first stage, the subspace of attention, convolution and multi-layer perception are adapted simultaneously within different sub-spaces. At the second stage, a memory-efficient optimization is proposed for subspace composition and the performance is further improved in the united sub-space. Initial experiments on the SCARED dataset demonstrate that results at the first stage improves the performance from 10.2% to 4.1% for Sq Rel, Abs Rel, RMSE and RMSE log in the comparison with the state-of-the-art models.

Self-Supervised Learning of Deviation in Latent Representation for Co-speech Gesture Video Generation

Gestures are pivotal in enhancing co-speech communication. While recent works have mostly focused on point-level motion transformation or fully supervised motion representations through data-driven approaches, we explore the representation of gestures in co-speech, with a focus on self-supervised representation and pixel-level motion deviation, utilizing a diffusion model which incorporates latent motion features. Our approach leverages self-supervised deviation in latent representation to facilitate hand gestures generation, which are crucial for generating realistic gesture videos. Results of our first experiment demonstrate that our method enhances the quality of generated videos, with an improvement from 2.7 to 4.5% for FGD, DIV, and FVD, and 8.1% for PSNR, 2.5% for SSIM over the current state-of-the-art methods.

Bailing-TTS: Chinese Dialectal Speech Synthesis Towards Human-like Spontaneous Representation

Large-scale text-to-speech (TTS) models have made significant progress recently.However, they still fall short in the generation of Chinese dialectal speech. Toaddress this, we propose Bailing-TTS, a family of large-scale TTS models capable of generating high-quality Chinese dialectal speech. Bailing-TTS serves as a foundation model for Chinese dialectal speech generation. First, continual semi-supervised learning is proposed to facilitate the alignment of text tokens and speech tokens. Second, the Chinese dialectal representation learning is developed using a specific transformer architecture and multi-stage training processes. With the proposed design of novel network architecture and corresponding strategy, Bailing-TTS is able to generate Chinese dialectal speech from text effectively and efficiently. Experiments demonstrate that Bailing-TTS generates Chinese dialectal speech towards human-like spontaneous representation. Readers are encouraged to listen to demos at \url{https://c9412600.github.io/bltts_tech_report/index.html}.

Hierarchical Reinforcement Learning for Furniture Layout in Virtual Indoor Scenes

In real life, the decoration of 3D indoor scenes through designing furniture layout provides a rich experience for people. In this paper, we explore the furniture layout task as a Markov decision process (MDP) in virtual reality, which is solved by hierarchical reinforcement learning (HRL). The goal is to produce a proper two-furniture layout in the virtual reality of the indoor scenes. In particular, we first design a simulation environment and introduce the HRL formulation for a two-furniture layout. We then apply a hierarchical actor-critic algorithm with curriculum learning to solve the MDP. We conduct our experiments on a large-scale real-world interior layout dataset that contains industrial designs from professional designers. Our numerical results demonstrate that the proposed model yields higher-quality layouts as compared with the state-of-art models.

LWA-HAND: Lightweight Attention Hand for Interacting Hand Reconstruction

Recent years have witnessed great success for hand reconstruction in real-time applications such as visual reality and augmented reality while interacting with two-hand reconstruction through efficient transformers is left unexplored. In this paper, we propose a method called lightweight attention hand (LWA-HAND) to reconstruct hands in low flops from a single RGB image. To solve the occlusion and interaction problem in efficient attention architectures, we propose three mobile attention modules in this paper. The first module is a lightweight feature attention module that extracts both local occlusion representation and global image patch representation in a coarse-to-fine manner. The second module is a cross image and graph bridge module which fuses image context and hand vertex. The third module is a lightweight cross-attention mechanism that uses element-wise operation for the cross-attention of two hands in linear complexity. The resulting model achieves comparable performance on the InterHand2.6M benchmark in comparison with the state-of-the-art models. Simultaneously, it reduces the flops to $0.47GFlops$ while the state-of-the-art models have heavy computations between $10GFlops$ and $20GFlops$.