CORAG: A Cost-Constrained Retrieval Optimization System for Retrieval-Augmented Generation

Large Language Models (LLMs) have demonstrated remarkable generation capabilities but often struggle to access up-to-date information, which can lead to hallucinations. Retrieval-Augmented Generation (RAG) addresses this issue by incorporating knowledge from external databases, enabling more accurate and relevant responses. Due to the context window constraints of LLMs, it is impractical to input the entire external database context directly into the model. Instead, only the most relevant information, referred to as chunks, is selectively retrieved. However, current RAG research faces three key challenges. First, existing solutions often select each chunk independently, overlooking potential correlations among them. Second, in practice the utility of chunks is non-monotonic, meaning that adding more chunks can decrease overall utility. Traditional methods emphasize maximizing the number of included chunks, which can inadvertently compromise performance. Third, each type of user query possesses unique characteristics that require tailored handling, an aspect that current approaches do not fully consider. To overcome these challenges, we propose a cost constrained retrieval optimization system CORAG for retrieval-augmented generation. We employ a Monte Carlo Tree Search (MCTS) based policy framework to find optimal chunk combinations sequentially, allowing for a comprehensive consideration of correlations among chunks. Additionally, rather than viewing budget exhaustion as a termination condition, we integrate budget constraints into the optimization of chunk combinations, effectively addressing the non-monotonicity of chunk utility.

Efficient Adaptation of Pre-trained Vision Transformer via Householder Transformation

A common strategy for Parameter-Efficient Fine-Tuning (PEFT) of pre-trained Vision Transformers (ViTs) involves adapting the model to downstream tasks by learning a low-rank adaptation matrix. This matrix is decomposed into a product of down-projection and up-projection matrices, with the bottleneck dimensionality being crucial for reducing the number of learnable parameters, as exemplified by prevalent methods like LoRA and Adapter. However, these low-rank strategies typically employ a fixed bottleneck dimensionality, which limits their flexibility in handling layer-wise variations. To address this limitation, we propose a novel PEFT approach inspired by Singular Value Decomposition (SVD) for representing the adaptation matrix. SVD decomposes a matrix into the product of a left unitary matrix, a diagonal matrix of scaling values, and a right unitary matrix. We utilize Householder transformations to construct orthogonal matrices that efficiently mimic the unitary matrices, requiring only a vector. The diagonal values are learned in a layer-wise manner, allowing them to flexibly capture the unique properties of each layer. This approach enables the generation of adaptation matrices with varying ranks across different layers, providing greater flexibility in adapting pre-trained models. Experiments on standard downstream vision tasks demonstrate that our method achieves promising fine-tuning performance.

ECMamba: Consolidating Selective State Space Model with Retinex Guidance for Efficient Multiple Exposure Correction

Exposure Correction (EC) aims to recover proper exposure conditions for images captured under over-exposure or under-exposure scenarios. While existing deep learning models have shown promising results, few have fully embedded Retinex theory into their architecture, highlighting a gap in current methodologies. Additionally, the balance between high performance and efficiency remains an under-explored problem for exposure correction task. Inspired by Mamba which demonstrates powerful and highly efficient sequence modeling, we introduce a novel framework based on Mamba for Exposure Correction (ECMamba) with dual pathways, each dedicated to the restoration of reflectance and illumination map, respectively. Specifically, we firstly derive the Retinex theory and we train a Retinex estimator capable of mapping inputs into two intermediary spaces, each approximating the target reflectance and illumination map, respectively. This setup facilitates the refined restoration process of the subsequent Exposure Correction Mamba Module (ECMM). Moreover, we develop a novel 2D Selective State-space layer guided by Retinex information (Retinex-SS2D) as the core operator of ECMM. This architecture incorporates an innovative 2D scanning strategy based on deformable feature aggregation, thereby enhancing both efficiency and effectiveness. Extensive experiment results and comprehensive ablation studies demonstrate the outstanding performance and the importance of each component of our proposed ECMamba. Code is available at https://github.com/LowlevelAI/ECMamba.

PMM-Net: Single-stage Multi-agent Trajectory Prediction with Patching-based Embedding and Explicit Modal Modulation

Analyzing and forecasting trajectories of agents like pedestrians plays a pivotal role for embodied intelligent applications. The inherent indeterminacy of human behavior and complex social interaction among a rich variety of agents make this task more challenging than common time-series forecasting. In this letter, we aim to explore a distinct formulation for multi-agent trajectory prediction framework. Specifically, we proposed a patching-based temporal feature extraction module and a graph-based social feature extraction module, enabling effective feature extraction and cross-scenario generalization. Moreover, we reassess the role of social interaction and present a novel method based on explicit modality modulation to integrate temporal and social features, thereby constructing an efficient single-stage inference pipeline. Results on public benchmark datasets demonstrate the superior performance of our model compared with the state-of-the-art methods. The code is available at: github.com/TIB-K330/pmm-net.

Particle-based Instance-aware Semantic Occupancy Mapping in Dynamic Environments

Representing the 3D environment with instance-aware semantic and geometric information is crucial for interaction-aware robots in dynamic environments. Nonetheless, creating such a representation poses challenges due to sensor noise, instance segmentation and tracking errors, and the objects' dynamic motion. This paper introduces a novel particle-based instance-aware semantic occupancy map to tackle these challenges. Particles with an augmented instance state are used to estimate the Probability Hypothesis Density (PHD) of the objects and implicitly model the environment. Utilizing a State-augmented Sequential Monte Carlo PHD (S$^2$MC-PHD) filter, these particles are updated to jointly estimate occupancy status, semantic, and instance IDs, mitigating noise. Additionally, a memory module is adopted to enhance the map's responsiveness to previously observed objects. Experimental results on the Virtual KITTI 2 dataset demonstrate that the proposed approach surpasses state-of-the-art methods across multiple metrics under different noise conditions. Subsequent tests using real-world data further validate the effectiveness of the proposed approach.

TG-LLaVA: Text Guided LLaVA via Learnable Latent Embeddings

Currently, inspired by the success of vision-language models (VLMs), an increasing number of researchers are focusing on improving VLMs and have achieved promising results. However, most existing methods concentrate on optimizing the connector and enhancing the language model component, while neglecting improvements to the vision encoder itself. In contrast, we propose Text Guided LLaVA (TG-LLaVA) in this paper, which optimizes VLMs by guiding the vision encoder with text, offering a new and orthogonal optimization direction. Specifically, inspired by the purpose-driven logic inherent in human behavior, we use learnable latent embeddings as a bridge to analyze textual instruction and add the analysis results to the vision encoder as guidance, refining it. Subsequently, another set of latent embeddings extracts additional detailed text-guided information from high-resolution local patches as auxiliary information. Finally, with the guidance of text, the vision encoder can extract text-related features, similar to how humans focus on the most relevant parts of an image when considering a question. This results in generating better answers. Experiments on various datasets validate the effectiveness of the proposed method. Remarkably, without the need for additional training data, our propsoed method can bring more benefits to the baseline (LLaVA-1.5) compared with other concurrent methods. Furthermore, the proposed method consistently brings improvement in different settings.

A Novel Ternary Evolving Estimator for Positioning Unmanned Aerial Vehicle in Harsh Environments

Obtaining reliable position estimation is fundamental for unmanned aerial vehicles during mission execution, especially in harsh environments. But environmental interference and abrupt changes usually degrade measurement reliability, leading to estimation divergence. To address this, existing works explore adaptive adjustment of sensor confidence. Unfortunately, existing methods typically lack synchronous evaluation of estimation precision, thereby rendering adjustments sensitive to abnormal data and susceptible to divergence. To tackle this issue, we propose a novel ternary-channel adaptive evolving estimator equipped with an online error monitor, where the ternary channels, states, noise covariance matrices and especially aerial drag, evolve simultaneously with environment. Firstly, an augmented filter is employed to pre-processes multidimensional data, followed by an inverse-Wishart smoother utilized to obtain posterior states and covariance matrices. Error propagation relation during estimation is analysed and hence an indicator is devised for online monitoring estimation errors. Under this premise, several restrictions are applied to suppress potential divergence led by interference. Additionally, considering motion dynamics, aerial drag matrix is reformulated based on updated states and covariance matrices. Finally, the observability, numerical sensitivity and arithmetic complexity of the proposed estimator are mathematically analyzed. Extensive experiments are conducted in both common and harsh environments (with average RMSE 0.17m and 0.39m respectively) to verify adaptability of algorithm and effectiveness of restriction design, which shows our method significantly outperforms the state-of-the-art.

Start from Video-Music Retrieval: An Inter-Intra Modal Loss for Cross Modal Retrieval

The burgeoning short video industry has accelerated the advancement of video-music retrieval technology, assisting content creators in selecting appropriate music for their videos. In self-supervised training for video-to-music retrieval, the video and music samples in the dataset are separated from the same video work, so they are all one-to-one matches. This does not match the real situation. In reality, a video can use different music as background music, and a music can be used as background music for different videos. Many videos and music that are not in a pair may be compatible, leading to false negative noise in the dataset. A novel inter-intra modal (II) loss is proposed as a solution. By reducing the variation of feature distribution within the two modalities before and after the encoder, II loss can reduce the model's overfitting to such noise without removing it in a costly and laborious way. The video-music retrieval framework, II-CLVM (Contrastive Learning for Video-Music Retrieval), incorporating the II Loss, achieves state-of-the-art performance on the YouTube8M dataset. The framework II-CLVTM shows better performance when retrieving music using multi-modal video information (such as text in videos). Experiments are designed to show that II loss can effectively alleviate the problem of false negative noise in retrieval tasks. Experiments also show that II loss improves various self-supervised and supervised uni-modal and cross-modal retrieval tasks, and can obtain good retrieval models with a small amount of training samples.

GLARE: Low Light Image Enhancement via Generative Latent Feature based Codebook Retrieval

Most existing Low-light Image Enhancement (LLIE) methods either directly map Low-Light (LL) to Normal-Light (NL) images or use semantic or illumination maps as guides. However, the ill-posed nature of LLIE and the difficulty of semantic retrieval from impaired inputs limit these methods, especially in extremely low-light conditions. To address this issue, we present a new LLIE network via Generative LAtent feature based codebook REtrieval (GLARE), in which the codebook prior is derived from undegraded NL images using a Vector Quantization (VQ) strategy. More importantly, we develop a generative Invertible Latent Normalizing Flow (I-LNF) module to align the LL feature distribution to NL latent representations, guaranteeing the correct code retrieval in the codebook. In addition, a novel Adaptive Feature Transformation (AFT) module, featuring an adjustable function for users and comprising an Adaptive Mix-up Block (AMB) along with a dual-decoder architecture, is devised to further enhance fidelity while preserving the realistic details provided by codebook prior. Extensive experiments confirm the superior performance of GLARE on various benchmark datasets and real-world data. Its effectiveness as a preprocessing tool in low-light object detection tasks further validates GLARE for high-level vision applications. Code is released at https://github.com/LowLevelAI/GLARE.

A Swift and Omnidirectional Formation Approach based on Hierarchical Reorganization

Current formations commonly rely on invariant hierarchical structures, such as predetermined leaders or enumerated formation shapes. These structures could be unidirectional and sluggish, constraining their adaptability and agility when encountering cluttered environments. To surmount these constraints, this work proposes an omnidirectional affine formation approach with hierarchical reorganizations. We first delineate the critical conditions requisite for facilitating hierarchical reorganizations within formations, which informs the development of the omnidirectional affine criterion. Central to our approach is the fluid leadership and authority redistribution, for which we develop a minimum time-driven leadership evaluation algorithm and a power transition control algorithm. These algorithms facilitate autonomous leader selection and ensure smooth power transitions, enabling the swarm to adapt hierarchically in alignment with the external environment. Furthermore, we deploy a power-centric topology switching mechanism tailored for the dynamic reorganization of in-team connections. Finally, simulations and experiments demonstrate the performance of the proposed method. The formation successfully performs several hierarchical reorganizations, with the longest reorganization taking only 0.047s. This swift adaptability allows five aerial robots to carry out complex tasks, including executing swerving movements and navigating through hoops at velocities up to 1.9m/s.