AAAI 2026 论文列表

Sequential-Horizon Vision-and-Language Navigation (SH-VLN) presents a challenging scenario where agents should sequentially execute multi-task trajectory navigation guided by complex, long-horizon natural language instructions. Current vision-and-language navigation models exhibit significant performance degradation with such instructions, as information overload impairs the agent's ability to attend to observationally relevant details. To address this problem, we propose SeqWalker, a novel navigation model built on a hierarchical planning framework. Our SeqWalker features: (1) A High-Level Planner that dynamically selects global instructions into contextually relevant sub-instructions based on the agent's current visual observations, thus reducing cognitive load; (2) A Low-Level Planner incorporating an Exploration-Verification strategy that leverages the inherent logical structure of instructions for trajectory error correction. To evaluate SH-VLN performance, we also extend the IVLN dataset and establish a new benchmark. Extensive experiments are performed to demonstrate the effectiveness and superiority of SeqWalker.

While classical control theory assumes that the controller has access to measurements of the entire state (or output) at every time instant, this paper investigates a setting where the feedback controller can only access a randomly selected subset of the state vector at each time step. Due to the random sparsification that selects only a subset of the state components at each step, we analyze the stability of the closed-loop system in terms of Asymptotic Mean-Square Stability (AMSS), which ensures that the system state converges to zero in the mean-square sense. We consider the problem of designing both a feedback gain matrix and a measurement sparsification strategy that minimizes the number of state components required for feedback, while ensuring AMSS of the closed-loop system. Interestingly, (1) we provide conditions on the dynamics of the system under which it is possible to find a sparsification strategy, and (2) we propose a Linear Matrix Inequality (LMI) based algorithm that jointly computes a stabilizing gain matrix, and a randomized sparsification strategy that minimizes the expected number of measured state coordinates while preserving the AMSS. Our approach is then extended to the case where the sparsification probabilities vary across the state components. Based on these theoretical findings, we propose an algorithmic procedure to compute the vector of sparsification parameters, along with the corresponding feedback gain matrix. To the best of our knowledge, this is the first study to investigate the stability properties of control systems that rely solely on randomly selected state measurements. Numerical simulations demonstrate that, in some settings, the system achieves comparable performance to full-state feedback while requiring measurements from only 0.3 percent of the state coordinates.

Diffusion policies have recently shown great promise for generating actions in robotic manipulation. However, existing approaches often rely on global instructions to produce short-term control signals, which can result in misalignment in action generation. We conjecture that the primitive skills, referred to as fine-grained, short-horizon manipulations, such as "move up" and "open the gripper", provide a more intuitive and effective interface for robot learning. To bridge this gap, we propose SDP, a skill-conditioned diffusion policy that integrates interpretable skill learning with conditional action planning. SDP abstracts eight reusable primitive skills across tasks and employs a vision-language model to extract discrete representations from visual observations and language instructions. Based on the representations, a lightweight router network is designed to assign a desired primitive skill for each state, which helps construct a single-skill policy to generate skill-aligned actions. By decomposing complex tasks into a sequence of primitive skills and selecting a single-skill policy, the proposed SDP ensures skill-consistent behavior across diverse tasks. Extensive experiments on two challenging simulation benchmarks and real-world robot deployments demonstrate that SDP consistently outperforms state-of-the-art methods, providing a new paradigm for skill-based robot learning with diffusion policies.

While natural language is commonly used to guide embodied agents, the inherent ambiguity and verbosity of language often hinder the effectiveness of language-guided navigation in complex environments. To this end, we propose Visual Prompt Navigation (VPN), a novel paradigm that guides agents to navigate using only user-provided visual prompts within 2D top-view maps. This visual prompt primarily focuses on marking the visual navigation trajectory on a top-down view of a scene, offering intuitive and spatially grounded guidance without relying on language instructions. It is more friendly for non-expert users and reduces interpretive ambiguity. We build VPN tasks in both discrete and continuous navigation settings, constructing two new datasets, R2R-VP and R2R-CE-VP, by extending existing R2R and R2R-CE episodes with corresponding visual prompts. Furthermore, we introduce VPNet, a dedicated baseline network to handle the VPN tasks, with two data augmentation strategies: view-level augmentation (altering initial headings and prompt orientations) and trajectory-level augmentation (incorporating diverse trajectories from large-scale 3D scenes), to enhance navigation performance. Extensive experiments evaluate how visual prompt forms, top-view map formats, and data augmentation strategies affect the performance of visual prompt navigation.

Neural Radiance Fields (NeRF)-based Visual Simultaneous Localization and Mapping (SLAM) achieve superior scene geometric modeling and robust camera tracking by leveraging neural representations. Existing methods typically relied on multi-resolution hash encoding with truncated signed distance fields (TSDF) to achieve high frame rates. However, unavoidable hash collisions can lead to artifacts, and multi-view color inconsistencies in indoor scenes can result in shape-radiance ambiguity, adversely affecting geometric quality and tracking accuracy. To address these issues, we propose a novel Multi-scale Hybrid Encoding-based Decoupled SLAM (MHED-SLAM). First, to mitigate the adverse effects of hash collisions and reduce the number of learnable parameters, we innovatively fuse a coarse-scale hash tri-plane with a fine-scale hash grid within a single latent volume. Second, to enable precise geometric reconstruction and camera tracking, we decouple the reconstruction and rendering processes, independently learning a TSDF field for reconstruction and a density field for rendering. Third, we devise a Symmetric Kullback-Leibler (SKL) strategy based on ray termination distributions to align the probability distributions derived from the TSDF and density fields for their synchronous convergence. Extensive experimental evaluations demonstrate that our approach surpasses the state-of-the-art (SOTA) methods by utilizing a faster frame rate of 20 Hz and fewer parameters, while achieving higher tracking and reconstruction accuracy.

Embodied navigation is a fundamental capability for intelligent agents, yet remains challenging in partially observable environments where navigation instructions can be difficult to interpret. However, existing tasks only provide unimodal instructions, which are ambiguous in complex multimodal environments with multiple similar objects, and may result in misinterpretation and navigation failure. To overcome these limitations, we introduce MINav, a novel task where the navigation path is precisely described by a multimodal instruction. The instruction provides multimodal cues, including object categories, RGB images, language descriptions, and auditory descriptions, which help the agent to disambiguate and ground objects in the environment and navigate effectively. We further construct a large-scale dataset of 43.9K navigation episodes using a two-stage pipeline that first annotates multimodal references of objects and then synthesizes diverse multimodal instructions. We find that existing methods struggle on MINav task, indicating substantial room for improvement in agents' multimodal grounding. To address this, we propose NaVLA^2, a vision-language-audio-action model that additionally integrates spatial audio and employs a CoThinkAct module to jointly generate high-level reasoning and consistent low-level actions. Experimental results demonstrate that NaVLA^2 significantly outperforms competitive baselines on MINav benchmark. We hope that our proposed MINav and NaVLA^2 will facilitate future research toward agents with stronger multimodal understanding and grounding capabilities for navigation.

Aerial Vision-and-Language Navigation (AVLN) requires Unmanned Aerial Vehicle (UAV) agents to localize targets in large-scale urban environments based on linguistic instructions. While successful navigation demands both global environmental reasoning and local scene comprehension, existing UAV agents typically adopt mono-granularity frameworks that struggle to balance these two aspects. To address this limitation, this work proposes a History-Enhanced Two-Stage Transformer (HETT) framework, which integrates the two aspects through a coarse-to-fine navigation pipeline. Specifically, HETT first predicts coarse-grained target positions by fusing spatial landmarks and historical context, then refines actions via fine-grained visual analysis. In addition, a historical grid map is designed to dynamically aggregate visual features into a structured spatial memory, enhancing comprehensive scene awareness. Additionally, the CityNav dataset annotations are manually refined to enhance data quality. Experiments on the refined CityNav dataset show that HETT delivers significant performance gains, while extensive ablation studies further verify the effectiveness of each component.

Ultra-low altitude UAVs (below 120 meters) are gaining importance in the booming low-altitude economy, where GNSS signals are often unreliable or unavailable. Vision-based localization emerges as a promising alternative; however, existing benchmarks are not designed for ultra-low flight and typically adopt pinhole cameras with limited field of view, making them less effective in handling occlusions and repetitive textures near the ground. To address these limitations, we introduce the first panoramic UAV localization dataset tailored for ultra-low altitude scenarios. Built on a four-fisheye-camera system in the high-fidelity RflySim platform, our dataset captures diverse conditions — including day/night cycles, extreme weather, and dynamic obstacles — and contains over hundreds of thousands of frames. It is further enhanced with real-world UAV panoramic data to narrow the sim-to-real gap and will be continuously updated for broader applicability. Comprehensive experiments confirm the effectiveness and transferability of our dataset, establishing it as a robust benchmark for future research in vision-based UAV localization.

Robust object detection for challenging scenarios increasingly relies on event cameras, yet existing Event-RGB datasets remain constrained by sparse coverage of extreme conditions and low spatial resolution (≤ 640 × 480), which prevents comprehensive evaluation of detectors under challenging scenarios. To address these limitations, we propose PEOD, the first large-scale, pixel-aligned and hign-resolution (1280 × 720) Event-RGB dataset for object detection under challenge conditions. PEOD contains 130+ spatiotemporal-aligned sequences and 340k manual bounding boxes, with 57% of data captured under low-light, overexposure, and high-speed motion. Furthermore, we benchmark 14 methods across three input configurations (Event-based, RGB-based, and Event-RGB fusion) on PEOD. On the full test set and normal subset, fusion-based models achieve the excellent performance. However, in illumination challenge subset, the top event-based model outperforms all fusion models, while fusion models still outperform their RGB-based counterparts, indicating limits of existing fusion methods when the frame modality is severely degraded. PEOD establishes a realistic, high-quality benchmark for multimodal perception and will be publicly released later to facilitate future research.

This paper explores the challenges of integrating tactile sensing into intelligent systems for multimodal reasoning, particularly in enabling commonsense reasoning about the open-ended physical world. We identify two key challenges: modality discrepancy, where existing touch-language models often treat touch as a mere sub-modality of language without further addressing the semantic differences, and open-ended tactile data scarcity, where current datasets lack the diversity, open-endedness, and complexity needed for reasoning. To overcome these challenges, we introduce SToLa, a Self-Adaptive Touch-Language framework. SToLa utilizes Mixture of Experts (MoE) to dynamically process, unify, and manage tactile and language modalities, capturing their unique characteristics. Crucially, we also present a comprehensive tactile commonsense reasoning dataset and benchmark featuring free-form questions and responses, 8 physical properties, 4 interactive characteristics, and diverse commonsense knowledge. Experiments show SToLa exhibits competitive performance compared to existing models on the PHYSICLEAR benchmark and self-constructed datasets, proving the effectiveness of the Mixture of Experts architecture in multimodal management and the performance advantages for open-scenario tactile commonsense reasoning tasks.

One-shot imitation learning (OSIL) offers a promising way to teach robots new skills without large-scale data collection. However, current OSIL methods are primarily limited to short-horizon tasks, thus limiting their applicability to complex, long-horizon manipulations. To address this limitation, we propose ManiLong-Shot, a novel framework that enables effective OSIL for long-horizon prehensile manipulation tasks. ManiLong-Shot structures long-horizon tasks around physical interaction events, reframing the problem as sequencing interaction-aware primitives instead of directly imitating continuous trajectories. This primitive decomposition can be driven by high-level reasoning from a vision-language model (VLM) or by rule-based heuristics derived from robot state changes. For each primitive, ManiLong-Shot predicts invariant regions critical to the interaction, establishes correspondences between the demonstration and the current observation, and computes the target end-effector pose, enabling effective task execution. Extensive simulation experiments show that ManiLong-Shot, trained on only 10 short-horizon tasks, generalizes to 20 unseen long-horizon tasks across three difficulty levels via one-shot imitation, achieving a 22.8% relative improvement over the SOTA. Additionally, real-robot experiments validate ManiLong-Shot’s ability to robustly execute three long-horizon manipulation tasks via OSIL, confirming its practical applicability.

The powerful generalization of Vision-Language-Action (VLA) models is bottlenecked by their heavy reliance on massive, redundant, and unevenly valued datasets, hindering their widespread application. Existing model-centric optimization paths, such as model compression (which often leads to performance degradation) or policy distillation (whose products are model-dependent and lack generality), fail to fundamentally address this data-level challenge. To this end, this paper introduces FT-NCFM, a fundamentally different, data-centric generative data distillation framework. Our framework employs a self-contained Fact-Tracing (FT) engine that combines causal attribution with programmatic contrastive verification to assess the intrinsic value of samples. Guided by these assessments, an adversarial NCFM process synthesizes a model-agnostic, information-dense, and reusable data asset. Experimental results on several mainstream VLA benchmarks show that models trained on just 5\% of our distilled coreset achieve a success rate of 85-90\% compared with training on the full dataset, while reducing training time by over 80\%. Our work demonstrates that intelligent data distillation is a highly promising new path for building efficient, high-performance VLA models.

Multi-robot systems in complex physical collaborations face a "shared brain dilemma": transmitting high-dimensional multimedia data (e.g., video streams at ~30MB/s) creates severe bandwidth bottlenecks and decision-making latency. To address this, we propose PIPHEN, an innovative distributed physical cognition-control framework. Its core idea is to replace "raw data communication" with "semantic communication" by performing "semantic distillation" at the robot edge, reconstructing high-dimensional perceptual data into compact, structured physical representations. This idea is primarily realized through two key components: (1) a novel Physical Interaction Prediction Network (PIPN), derived from large model knowledge distillation, to generate this representation; and (2) a Hamiltonian Energy Network (HEN) controller, based on energy conservation, to precisely translate this representation into coordinated actions. Experiments show that, compared to baseline methods, PIPHEN can compress the information representation to less than 5\% of the original data volume and reduce collaborative decision-making latency from 315ms to 76ms, while significantly improving task success rates. This work provides a fundamentally efficient paradigm for resolving the "shared brain dilemma" in resource-constrained multi-robot systems.

Visual Dialogue Navigation (VDN) aims to enable agents to reach target locations through dialogue with humans. The integration of VDN into Unmanned Aerial Vehicle (UAV) systems enhances human-machine interaction by enabling intuitive, hands-free operation, thereby unlocking vast applications. However, existing VDN models for UAVs can only perform navigation based on dialogue history, lacking proactive interaction capabilities to correct trajectories. Moreover, their sequential observation history recording mechanism struggles to accurately localize landmarks observed in the historical context, leading to ineffective utilization of referential information in new user instructions.To address these, we present AerialVLA, an end-to-end UAV navigation framework integrating dialogue comprehension, action decision-making, and navigational question generation. AerialVLA comprises three core components: i) we propose the Progress-Driven Navigation-Query Alternation mechanism to determine optimal questioning timing through navigation progress estimation autonomously. ii) To effectively model long-horizon history observation sequences, we develop the History Spatial-Temporal Fusion module that extracts discriminative spatial-temporal representations from historical observations. iii) Furthermore, to overcome data scarcity in training, we devise the Online Task-Driven Augmentation strategy that enhances learning through action-conditioned data augmentation. Experimental results demonstrate that AerialVLA achieves state-of-the-art navigation performance while exhibiting effective dialogue capabilities.Moreover, to better evaluate the agent's proactive dialogue and navigation abilities, our evaluation benchmark, named UAV Navigation with Online Dialogue (UNOD), incorporates an online dialogue interaction module. The UNOD assesses UAV agents' real-time questioning capabilities by leveraging an Air Commander Large Language Model to simulate human-UAV interactions during testing.

Controlling nonlinear stochastic systems with parametric uncertainty is a fundamental challenge in modern control theory. This paper presents a comprehensive theoretical framework for a natural-gradient method applied to polynomial chaos theory. We focus on quadratic regulator problems characterized by both parametric uncertainty and additive stochastic disturbances. We extend existing polynomial chaos approaches from linear systems to general nonlinear dynamics. To achieve this, we develop new mathematical tools to handle the complex interactions between nonlinearity, parameter uncertainty, and noise. The framework provides local convergence guarantees for the proposed natural gradient algorithm. Furthermore, it offers practical computational strategies while carefully characterizing the theoretical limitations in the nonlinear setting.

We aim to develop a goal specification method that is semantically clear, spatially sensitive, domain-agnostic, and intuitive for human users to guide agent interactions in 3D environments. Specifically, we propose a novel cross-view goal alignment framework that allows users to specify target objects using segmentation masks from their camera views rather than the agent’s observations. We highlight that behavior cloning alone fails to align the agent’s behavior with human intent when the human and agent camera views differ significantly. To address this, we introduce two auxiliary objectives: cross-view consistency loss and target visibility loss, which explicitly enhance the agent's spatial reasoning ability. According to this, we develop ROCKET-2, a state-of-the-art agent trained in Minecraft, achieving an improvement in the efficiency of inference 3x to 6x. We demonstrate that ROCKET-2 can directly interpret goals from human camera views, enabling better human-agent interaction. Remarkably, ROCKET-2 demonstrates zero-shot generalization capabilities: despite being trained exclusively on the Minecraft dataset, it can adapt and generalize to other 3D environments like Doom, DMLab, and Unreal through a simple action space mapping.

Imitation learning for robotic manipulation faces a fundamental challenge: the scarcity of large-scale, high-quality robot demonstration data. Recent robotic foundation models often pre-train on cross-embodiment robot datasets to increase data scale, while they face significant limitations as the diverse morphologies and action spaces across different robot embodiments make unified training challenging. In this paper, we present H-RDT (Human to Robotics Diffusion Transformer), a novel approach that leverages human manipulation data to enhance robot manipulation capabilities. Our key insight is that large-scale egocentric human manipulation videos with paired 3D hand pose annotations provide rich behavioral priors that capture natural manipulation strategies and can benefit robotic policy learning. We introduce a two-stage training paradigm: (1) pre-training on large-scale egocentric human manipulation data, and (2) cross-embodiment fine-tuning on robot-specific data with modular action encoders and decoders. Built on a diffusion transformer architecture with 2B parameters, H-RDT uses flow matching to model complex action distributions. The modular design of action encoder and decoder components enables effective knowledge transfer from the unified human embodiment to diverse robot platforms through efficient fine-tuning. Extensive evaluations encompassing both simulation and real-world experiments, single-task and multitask scenarios, as well as few-shot learning and robustness assessments, demonstrate that H-RDT outperforms training from scratch and existing state-of-the-art methods, including π0 and RDT, achieving significant improvements of 13.9% and 40.5% over training from scratch in simulation and real-world experiments, respectively. The results validate our core hypothesis that human manipulation data can serve as a powerful foundation for learning bimanual robotic manipulation policies.

The inherent difficulty and limited scalability of collecting manipulation data using multi-fingered robot hand hardware platforms have resulted in severe data scarcity, impeding research on data-driven dexterous manipulation policy learning. To address this challenge, we present a hand-agnostic manipulation transfer system. It efficiently converts human hand manipulation sequences from demonstration videos into high-quality dexterous manipulation trajectories without requirements of massive training data. To tackle the multi-dimensional disparities between human hands and dexterous hands, as well as the challenges posed by high-degree-of-freedom coordinated control of dexterous hands, we design a progressive transfer framework: first, we establish primary control signals for dexterous hands based on kinematic matching; subsequently, we train residual policies with action space rescaling and thumb-guided initialization to dynamically optimize contact interactions under unified rewards; finally, we compute wrist control trajectories with the objective of preserving operational semantics. Using only human hand manipulation videos, our system automatically configures system parameters for different tasks, balancing kinematic matching and dynamic optimization across dexterous hands, object categories, and tasks. Extensive experimental results demonstrate that our framework can automatically generate smooth and semantically correct dexterous hand manipulation that faithfully reproduces human intentions, achieving high efficiency and strong generalizability with an average transfer success rate of 73%, providing an easily implementable and scalable method for collecting robot dexterous manipulation data. Refer to the arXiv version for the appendix.

Task-Oriented Grasping (TOG) presents a significant challenge, requiring a nuanced understanding of task semantics, object affordances, and the functional constraints dictating how an object should be grasped for a specific task. To address these challenges, we introduce GRIM (Grasp Re-alignment via Iterative Matching), a novel training-free framework for task-oriented grasping. Initially, a coarse alignment strategy is developed using a combination of geometric cues and principal component analysis (PCA)-reduced DINO features for similarity scoring. Subsequently, the full grasp pose associated with the retrieved memory instance is transferred to the aligned scene object and further refined against a set of task-agnostic, geometrically stable grasps generated for the scene object, prioritizing task compatibility. In contrast to existing learning-based methods, GRIM demonstrates strong generalization capabilities, achieving robust performance with only a small number of conditioning examples.