With the rapid development of the warehousing and logistics industries, the packing of goods has gradually attracted the attention of academia and industry. The packing of footwear products is a typical representative paired-item packing task involving irregular shapes and deformable objects. Although studies on shoe packing have been conducted, different initial states due to the irregular shapes of shoes and standard packing placement poses have not been considered. This study proposes a robotic manipulation framework, including a perception module, reorientation planners, and a packing planner, that can complete the packing of pairs of shoes in any initial state. First, to adapt to the large intraclass variations due to the state, shape, and deformation of the shoe, we propose a vision module based on semantic keypoints, which can also infer more information such as size, state, pose, and manipulation points by combining geometric features. Subsequently, we not only proposed primitive-based reorientation methods for different states of a single deformable shoe but also proposed a fast reorientation method for the top state using box edge contact and gravity, which further improved the efficiency of reorientation. Finally, based on the perception module and reorientation methods, we propose a task planner for shoe pair packing in any initial state to provide an optimal packing strategy. Real-world experiments were conducted to verify the robustness of the reorientation methods and the effectiveness of the packing strategy for various types of shoes. In this study, we highlight the potential of semantic keypoint representation methods, introduce new perspectives on the reorientation of 3D deformable objects and multi-object manipulation, and provide a reference for paired object packing.

Abstract--With the development of large-scale diffusion-based and language-modeling-based generative models, impressive progress has been achieved in text-to-audio generation. Despite producing high-quality outputs, existing text-to-audio models mainly aim to generate semantically aligned sound and fall short on precisely controlling fine-grained acoustic characteristics of specific sounds. As a result, users that need specific sound content may find it challenging to generate the desired audio clips. In this paper, we present DreamAudio for customized text-to-audio generation (CTT A). Specifically, we introduce a new framework that is designed to enable the model to identify auditory information from user-provided reference concepts for audio generation. Given a few reference audio samples containing personalized audio events, our system can generate new audio samples that include these specific events. In addition, two types of datasets are developed for training and testing the customized systems. The experiments show that the proposed model, DreamAudio, generates audio samples that are highly consistent with the customized audio features and aligned well with the input text prompts. Furthermore, DreamAudio offers comparable performance in general text-to-audio tasks. We also provide a human-involved dataset containing audio events from real-world CTT A cases as the benchmark for customized generation tasks. Udio generation, as a crucial technology for enabling artificial intelligence generated content (AIGC) [1], has gained significant interest from the research community.

While Retrieval-Augmented Generation (RAG) effectively reduces hallucinations by integrating external knowledge bases, it introduces vulnerabilities to membership inference attacks (MIAs), particularly in systems handling sensitive data. Existing MIAs targeting RAG's external databases often rely on model responses but ignore the interference of non-member-retrieved documents on RAG outputs, limiting their effectiveness. To address this, we propose DCMI, a differential calibration MIA that mitigates the negative impact of non-member-retrieved documents. Specifically, DCMI leverages the sensitivity gap between member and non-member retrieved documents under query perturbation. It generates perturbed queries for calibration to isolate the contribution of member-retrieved documents while minimizing the interference from non-member-retrieved documents. Experiments under progressively relaxed assumptions show that DCMI consistently outperforms baselines--for example, achieving 97.42% AUC and 94.35% Accuracy against the RAG system with Flan-T5, exceeding the MBA baseline by over 40%. Furthermore, on real-world RAG platforms such as Dify and MaxKB, DCMI maintains a 10%-20% advantage over the baseline. These results highlight significant privacy risks in RAG systems and emphasize the need for stronger protection mechanisms. We appeal to the community's consideration of deeper investigations, like ours, against the data leakage risks in rapidly evolving RAG systems.

A central goal of artificial intelligence is to build systems that can understand and predict complex, evolving sequences of events. However, current foundation models, designed for natural language, fail to grasp the holistic nature of structured interactions found in domains like telecommunications, e-commerce and finance. By serializing events into text, they disassemble them into semantically fragmented parts, losing critical context. In this work, we introduce the Unified Interaction Foundation Model (UIFM), a foundation model engineered for genuine behavioral understanding. At its core is the principle of composite tokenization, where each multi-attribute event is treated as a single, semantically coherent unit. This allows UIFM to learn the underlying "grammar" of user behavior, perceiving entire interactions rather than a disconnected stream of data points. We demonstrate that this architecture is not just more accurate, but represents a fundamental step towards creating more adaptable and intelligent predictive systems.

Reinforcement learning (RL) has recently become the dominant paradigm for strengthening the reasoning abilities of large language models (LLMs). Yet the rule-based reward functions commonly used on mathematical or programming benchmarks assess only answer format and correctness, providing no signal as to whether the induced Chain-of-Thought (CoT) actually improves the answer. Furthermore, such task-specific training offers limited control over logical depth and therefore may fail to reveal a model's genuine reasoning capacity. We propose Dynamic Reasoning Efficiency Reward (DRER) -- a plug-and-play RL reward framework that reshapes both reward and advantage signals. (i) A Reasoning Quality Reward assigns fine-grained credit to those reasoning chains that demonstrably raise the likelihood of the correct answer, directly incentivising the trajectories with beneficial CoT tokens. (ii) A Dynamic Length Advantage decays the advantage of responses whose length deviates from a validation-derived threshold, stabilising training. To facilitate rigorous assessment, we also release Logictree, a dynamically constructed deductive reasoning dataset that functions both as RL training data and as a comprehensive benchmark. Experiments confirm the effectiveness of DRER: our 7B model attains GPT-o3-mini level performance on Logictree with 400 trianing steps, while the average confidence of CoT-augmented answers rises by 30%. The model further exhibits generalisation across diverse logical-reasoning datasets, and the mathematical benchmark AIME24. These results illuminate how RL shapes CoT behaviour and chart a practical path toward enhancing formal-reasoning skills in large language models. All code and data are available in repository https://github.com/Henryhe09/DRER.

As global interest in diverse culinary experiences grows, food image models are essential for improving food-related applications by enabling accurate food recognition, recipe suggestions, dietary tracking, and automated meal planning. Despite the abundance of food datasets, a noticeable gap remains in capturing the nuances of Indian cuisine due to its vast regional diversity, complex preparations, and the lack of comprehensive labeled datasets that cover its full breadth. Through this exploration, we uncover Khana, a new benchmark dataset for food image classification, segmentation, and retrieval of dishes from Indian cuisine. Khana fills the gap by establishing a taxonomy of Indian cuisine and offering around 131K images in the dataset spread across 80 labels, each with a resolution of 500x500 pixels. This paper describes the dataset creation process and evaluates state-of-the-art models on classification, segmentation, and retrieval as baselines. Khana bridges the gap between research and development by providing a comprehensive and challenging benchmark for researchers while also serving as a valuable resource for developers creating real-world applications that leverage the rich tapestry of Indian cuisine. Webpage: https://khana.omkar.xyz

Abstract--Monocular 3D Object Detection represents a challenging Computer Vision task due to the nature of the input used, which is a single 2D image, lacking in any depth cues and placing the depth estimation problem as an ill-posed one. Existing solutions leverage the information extracted from the input by using Convolutional Neural Networks or Transformer architectures as feature extraction backbones, followed by specific detection heads for 3D parameters prediction. In this paper, we introduce a decoupled strategy based on injecting precomputed segmentation information priors and fusing them directly into the feature space for guiding the detection, without expanding the detection model or jointly learning the priors. The focus is on evaluating the impact of additional segmentation information on existing detection pipelines without adding additional prediction branches. The proposed method is evaluated on the KITTI 3D Object Detection Benchmark, outperforming the equivalent architecture that relies only on RGB image features for small objects in the scene: pedestrians and cyclists, and proving that understanding the input data can balance the need for additional sensors or training data. The difference between 2D and 3D object detection is the target environment - 2D or 3D space, and the predicted parameters which define the bounding boxes.

Diffractive neural networks leverage the high-dimensional characteristics of electromagnetic (EM) fields for high-throughput computing. However, the existing architectures face challenges in integrating large-scale multidimensional metasurfaces with precise network training and haven't utilized multidimensional EM field coding scheme for super-resolution sensing. Here, we propose diffractive meta-neural networks (DMNNs) for accurate EM field modulation through metasurfaces, which enable multidimensional multiplexing and coding for multi-task learning and high-throughput super-resolution direction of arrival estimation. DMNN integrates pre-trained mini-metanets to characterize the amplitude and phase responses of meta-atoms across different polarizations and frequencies, with structure parameters inversely designed using the gradient-based meta-training. For wide-field super-resolution angle estimation, the system simultaneously resolves azimuthal and elevational angles through x and y-polarization channels, while the interleaving of frequency-multiplexed angular intervals generates spectral-encoded optical super-oscillations to achieve full-angle high-resolution estimation. Post-processing lightweight electronic neural networks further enhance the performance. Experimental results validate that a three-layer DMNN operating at 27 GHz, 29 GHz, and 31 GHz achieves $\sim7\times$ Rayleigh diffraction-limited angular resolution (0.5$^\circ$), a mean absolute error of 0.048$^\circ$ for two incoherent targets within a $\pm 11.5^\circ$ field of view, and an angular estimation throughput an order of magnitude higher (1917) than that of existing methods. The proposed architecture advances high-dimensional photonic computing systems by utilizing inherent high-parallelism and all-optical coding methods for ultra-high-resolution, high-throughput applications.

Abstract--Recently, cross-attention-based contextual automatic speech recognition (ASR) models have made notable advancements in recognizing personalized biasing phrases. However, the effectiveness of cross-attention is affected by variations in biasing information volume, especially when the length of the biasing list increases significantly. We find that, regardless of the length of the biasing list, only a limited amount of biasing information is most relevant to a specific ASR intermediate representation. Therefore, by identifying and integrating the most relevant biasing information rather than the entire biasing list, we can alleviate the effects of variations in biasing information volume for contextual ASR. T o this end, we propose a purified semantic correlation joint modeling (PSC-Joint) approach. In PSC-Joint, we define and calculate three semantic correlations between the ASR intermediate representations and biasing information from coarse to fine: list-level, phrase-level, and token-level. Then, the three correlations are jointly modeled to produce their intersection, so that the most relevant biasing information across various granularities is highlighted and integrated for contextual recognition. In addition, to reduce the computational cost introduced by the joint modeling of three semantic correlations, we also propose a purification mechanism based on a grouped-and-competitive strategy to filter out irrelevant biasing phrases. Compared with baselines, our PSC-Joint approach achieves average relative F1 score improvements of up to 21.34% on AISHELL-1 and 28.46% on KeSpeech, across biasing lists of varying lengths. N recent years, remarkable advancements have been made on end-to-end automatic speech recognition (E2E ASR), such as connectionist temporal classification [1], recurrent neural network transducer [2], [3], and attention encoder-decoder [4]-[7].

With Large Language Models' (LLMs) emergent abilities on code generation tasks, Text-to-SQL has become one of the most popular downstream applications. Despite the strong results of multiple recent LLM-based Text-to-SQL frameworks, the research community often overlooks the importance of database schema information for generating high-quality SQL queries. We find that such schema information plays a significant or even dominant role in the Text-to-SQL task. To tackle this challenge, we propose a novel database schema expert with two components. We first introduce X-Linking, an LLM Supervised Finetuning (SFT)- based method that achieves superior Schema Linking results compared to existing open-source Text-to-SQL methods. In addition, we innovatively propose an X-Admin component that focuses on Schema Understanding by bridging the gap between abstract schema information and the user's natural language question. Aside from better learning with schema information, we experiment with Multi-LLMs for different components within the system to further boost its performance. By incorporating these techniques into our end-to-end framework, X-SQL, we have achieved Execution Accuracies of 84.9% on the Spider-Dev dataset and 82.5% on the Spider-Test dataset. This outstanding performance establishes X-SQL as the leading Text-to-SQL framework based on open-source models.