Generic instance search models can dramatically reduce the manual effort required to analyze vast surveillance footage during criminal investigations by retrieving specific objects of interest to law enforcement. However, our research reveals an unintended emergent capability: through overlearning, these models can single out specific individuals even when trained on datasets without human subjects. This capability raises concerns regarding identification and profiling of individuals based on their personal data, while there is currently no clear standard on how de-identification can be achieved. We evaluate two technical safeguards to curtail a model's person re-identification capacity: index exclusion and confusion loss. Our experiments demonstrate that combining these approaches can reduce person re-identification accuracy to below 2% while maintaining 82% of retrieval performance for non-person objects. However, we identify critical vulnerabilities in these mitigations, including potential circumvention using partial person images. These findings highlight urgent regulatory questions at the intersection of AI governance and data protection: How should we classify and regulate systems with emergent identification capabilities? And what technical standards should be required to prevent identification capabilities from developing in seemingly benign applications?

Referring expression comprehension (REC) aims to localize the target object described by a natural language expression. Recent advances in vision-language learning have led to significant performance improvements in REC tasks. However, localizing extremely small objects remains a considerable challenge despite its importance in real-world applications such as autonomous driving. T o address this issue, we introduce a novel dataset and method for REC targeting small objects. First, we present the small object REC (SOREC) dataset, which consists of 100,000 pairs of referring expressions and corresponding bounding boxes for small objects in driving scenarios. Second, we propose the progressive-iterative zooming adapter (PIZA), an adapter module for parameter-efficient fine-tuning that enables models to progressively zoom in and localize small objects. In a series of experiments, we apply PIZA to GroundingDINO and demonstrate a significant improvement in accuracy on the SOREC dataset. Our dataset, codes and pre-trained models are publicly available on the project page.

Neural Information Processing Systems http://nips.cc/

Semantic Simultaneous Localization and Mapping (SLAM) is a critical area of research within robotics and computer vision, focusing on the simultaneous localization of robotic systems and associating semantic information to construct the most accurate and complete comprehensive model of the surrounding environment. Since the first foundational work in Semantic SLAM appeared more than two decades ago, this field has received increasing attention across various scientific communities. Despite its significance, the field lacks comprehensive surveys encompassing recent advances and persistent challenges. In response, this study provides a thorough examination of the state-of-the-art of Semantic SLAM techniques, with the aim of illuminating current trends and key obstacles. Beginning with an in-depth exploration of the evolution of visual SLAM, this study outlines its strengths and unique characteristics, while also critically assessing previous survey literature. Subsequently, a unified problem formulation and evaluation of the modular solution framework is proposed, which divides the problem into discrete stages, including visual localization, semantic feature extraction, mapping, data association, and loop closure optimization. Moreover, this study investigates alternative methodologies such as deep learning and the utilization of large language models, alongside a review of relevant research about contemporary SLAM datasets. Concluding with a discussion on potential future research directions, this study serves as a comprehensive resource for researchers seeking to navigate the complex landscape of Semantic SLAM.

Noname manuscript No. (will be inserted by the editor) Abstract Understanding human instructions and accomplishing Vision-Language Navigation tasks in unknown environments is essential for robots. However, existing modular approaches heavily rely on the quality of training data and often exhibit poor generalization. Vision-Language Model based methods, while demonstrating strong generalization capabilities, tend to perform unsatisfactorily when semantic cues are weak. To address these issues, this paper proposes SONAR, an aggregated reasoning approach through a cross modal paradigm. The proposed method integrates a semantic map based target prediction module with a Vision-Language Model based value map module, enabling more robust navigation in unknown environments with varying levels of semantic cues, and effectively balancing generalization ability with scene adaptability. In terms of target localization, we propose a strategy that integrates multi-scale semantic maps with confidence maps, aiming to mitigate false detections of target objects. We conducted an evaluation of the SONAR within the Gazebo simulator, leveraging the most challenging Mat-null Jiankun Wang E-mail: wangjk@sustech.edu.cn Experimental results demonstrate that SONAR achieves a success rate of 38.4% and an SPL of 17.7%. Keywords Object Goal Navigation Vision-Language Model Aggregated Reasoning 1 Introduction In an unknown environment, for a robot to accurately understand human instructions and complete vision language navigation tasks, it needs to rely on limited visual and linguistic cues to develop efficient exploration strategies while achieving precise identification of target objects[1].

Achieving generalizable bimanual manipulation requires systems that can learn efficiently from minimal human input while adapting to real-world uncertainties and diverse embodiments. Existing approaches face a dilemma: imitation policy learning demands extensive demonstrations to cover task variations, while modular methods often lack flexibility in dynamic scenes. We introduce VLBiMan, a framework that derives reusable skills from a single human example through task-aware decomposition, preserving invariant primitives as anchors while dynamically adapting adjustable components via vision-language grounding. This adaptation mechanism resolves scene ambiguities caused by background changes, object repositioning, or visual clutter without policy retraining, leveraging semantic parsing and geometric feasibility constraints. Moreover, the system inherits human-like hybrid control capabilities, enabling mixed synchronous and asynchronous use of both arms. Extensive experiments validate VLBiMan across tool-use and multi-object tasks, demonstrating: (1) a drastic reduction in demonstration requirements compared to imitation baselines, (2) compositional generalization through atomic skill splicing for long-horizon tasks, (3) robustness to novel but semantically similar objects and external disturbances, and (4) strong cross-embodiment transfer, showing that skills learned from human demonstrations can be instantiated on different robotic platforms without retraining. By bridging human priors with vision-language anchored adaptation, our work takes a step toward practical and versatile dual-arm manipulation in unstructured settings.

Abstract--Rearranging objects in cluttered tabletop environments remains a long-standing challenge in robotics. Classical planners often generate inefficient, high-cost plans by shuffling objects individually and using fixed buffers--temporary spaces such as empty table regions or static stacks--to resolve conflicts. When only free table locations are used as buffers, dense scenes become inefficient, since placing an object can restrict others from reaching their goals and complicate planning. Allowing stacking provides extra buffer capacity, but conventional stacking is static: once an object supports another, the base cannot be moved, which limits efficiency. T o overcome these issues, a novel planning primitive called the Dynamic Buffer is introduced. Inspired by human grouping strategies, it enables robots to form temporary, movable stacks that can be transported as a unit. This improves both feasibility and efficiency in dense layouts, and it also reduces travel in large-scale settings where space is abundant. Compared with a state-of-the-art rearrangement planner, the approach reduces manipulator travel cost by 11.89% in dense scenarios with a stationary robot and by 5.69% in large, low-density settings with a mobile manipulator . Practicality is validated through experiments on a Delta parallel robot with a two-finger gripper . These findings establish dynamic buffering as a key primitive for cost-efficient and robust rearrangement planning. The growing field of Embodied AI is focused on creating autonomous systems that can physically interact with and modify their environments to achieve goals. A crucial aspect of this interaction is the ability to rearrange objects, a task identified as a canonical challenge for embodied agents [1].

Abstract-- Object-level Simultaneous Localization and Mapping (SLAM), which incorporates semantic information for high-level scene understanding, faces challenges of under-constrained optimization due to sparse observations. Prior work has introduced additional constraints using commonsense knowledge, but obtaining such priors has traditionally been labor-intensive and lacks generalizability across diverse object categories. We address this limitation by leveraging large language models (LLMs) to provide commonsense knowledge of object geometric attributes, specifically size and orientation, as prior factors in a graph-based SLAM framework. These priors are particularly beneficial during the initial phase when object observations are limited. We implement a complete pipeline integrating these priors, achieving robust data association on sparse object-level features and enabling real-time object SLAM. Our system, evaluated on the TUM RGB-D and 3RScan datasets, improves mapping accuracy by 36.8% over the latest baseline. Object Simultaneous Localization and Mapping (SLAM) builds environment maps by identifying and localizing objects, and using this information to infer the robot's position. Unlike traditional feature-based SLAM, object-level representations are sparse, focusing on semantic object data. Comparing to semantic segmentation on dense representations, such sparsity improves computational efficiency and reduces storage requirements.

Generating realistic 3D point clouds is a fundamental problem in computer vision with applications in remote sensing, robotics, and digital object modeling. Existing generative approaches primarily capture geometry, and when semantics are considered, they are typically imposed post hoc through external segmentation or clustering rather than integrated into the generative process itself. We propose a diffusion-based framework that embeds per-point semantic conditioning directly within generation. Each point is associated with a conditional variable corresponding to its semantic label, which guides the diffusion dynamics and enables the joint synthesis of geometry and semantics. This design produces point clouds that are both structurally coherent and segmentation-aware, with object parts explicitly represented during synthesis. Through a comparative analysis of guided and unguided diffusion processes, we demonstrate the significant impact of conditional variables on diffusion dynamics and generation quality. Extensive experiments validate the efficacy of our approach, producing detailed and accurate 3D point clouds tailored to specific parts and features.

Recently, query-based methods have achieved remarkable performance in Referring Video Object Segmentation (RVOS) by using textual static object queries to drive cross-modal alignment. However, these static queries are easily misled by distractors with similar appearance or motion, resulting in \emph{query selection bias}. To address this issue, we propose Triple Query Former (TQF), which factorizes the referring query into three specialized components: an appearance query for static attributes, an intra-frame interaction query for spatial relations, and an inter-frame motion query for temporal association. Instead of relying solely on textual embeddings, our queries are dynamically constructed by integrating both linguistic cues and visual guidance. Furthermore, we introduce two motion-aware aggregation modules that enhance object token representations: Intra-frame Interaction Aggregation incorporates position-aware interactions among objects within a single frame, while Inter-frame Motion Aggregation leverages trajectory-guided alignment across frames to ensure temporal coherence. Extensive experiments on multiple RVOS benchmarks demonstrate the advantages of TQF and the effectiveness of our structured query design and motion-aware aggregation modules.