Daily News - Daily News

Generated at 2025-12-03 05:02:01

We have 210 news from different sources.

2feed¶

2.1悟空杀进NeurIPS！把七十二变写成数学公式，这才是我们想看的“东方魔法”¶

src root src

2025/12/02 05:05 GTM

这波反向文化输出我给满分

2.2差分隐私失效！揭秘GNN拓扑推断攻击：PGR用「零重叠」合成图破解危局¶

src root src

2025/12/02 05:05 GTM

GNN拓扑隐私全面失守

Modeling dynamic 3D environments from LiDAR sequences is central to building reliable 4D worlds for autonomous driving and embodied AI. Existing generative frameworks, however, often treat all spatial regions uniformly, overlooking the varying uncertainty across real-world scenes. This uniform generation leads to artifacts in complex or ambiguous regions, limiting realism and temporal stability. In this work, we present U4D, an uncertainty-aware framework for 4D LiDAR world modeling. Our approach first estimates spatial uncertainty maps from a pretrained segmentation model to localize semantically challenging regions. It then performs generation in a “hard-to-easy” manner through two sequential stages: (1) uncertainty-region modeling, which reconstructs high-entropy regions with fine geometric fidelity, and (2) uncertainty-conditioned completion, which synthesizes the remaining areas under learned structural priors. To further ensure temporal coherence, U4D incorporates a mixture of spatio-temporal (MoST) block that adaptively fuses spatial and temporal representations during diffusion. Extensive experiments show that U4D produces geometrically faithful and temporally consistent LiDAR sequences, advancing the reliability of 4D world modeling for autonomous perception and simulation.

arxiv html pdf kimi

2025/12/03 05:00 GTM

Hallucination remains a critical challenge in large language models (LLMs), hindering the development of reliable multimodal LLMs (MLLMs). Existing solutions often rely on human intervention or underutilize the agent’s ability to autonomously mitigate hallucination. To address these limitations, we draw inspiration from how humans make reliable decisions in the real world. They begin with introspective reasoning to reduce uncertainty and form an initial judgment, then rely on external verification from diverse perspectives to reach a final decision. Motivated by this cognitive paradigm, we propose InEx, a training-free, multi-agent framework designed to autonomously mitigate hallucination. InEx introduces internal introspective reasoning, guided by entropy-based uncertainty estimation, to improve the reliability of the decision agent’s reasoning process. The agent first generates a response, which is then iteratively verified and refined through external cross-modal multi-agent collaboration with the editing agent and self-reflection agents, further enhancing reliability and mitigating hallucination. Extensive experiments show that InEx consistently outperforms existing methods, achieving 4%-27% gains on general and hallucination benchmarks, and demonstrating strong robustness.

3.5Contextual Image Attack: How Visual Context Exposes Multimodal Safety Vulnerabilities¶

3.85Generalizing Vision-Language Models with Dedicated Prompt Guidance¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

Fine-tuning large pretrained vision-language models (VLMs) has emerged as a prevalent paradigm for downstream adaptation, yet it faces a critical trade-off between domain specificity and domain generalization (DG) ability. Current methods typically fine-tune a universal model on the entire dataset, which potentially compromises the ability to generalize to unseen domains. To fill this gap, we provide a theoretical understanding of the generalization ability for VLM fine-tuning, which reveals that training multiple parameter-efficient expert models on partitioned source domains leads to better generalization than fine-tuning a universal model. Inspired by this finding, we propose a two-step domain-expert-Guided DG (GuiDG) framework. GuiDG first employs prompt tuning to obtain source domain experts, then introduces a Cross-Modal Attention module to guide the fine-tuning of the vision encoder via adaptive expert integration. To better evaluate few-shot DG, we construct ImageNet-DG from ImageNet and its variants. Extensive experiments on standard DG benchmarks and ImageNet-DG demonstrate that GuiDG improves upon state-of-the-art fine-tuning methods while maintaining efficiency.

3.86MitUNet: Enhancing Floor Plan Recognition using a Hybrid Mix-Transformer and U-Net Architecture¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

Automatic 3D reconstruction of indoor spaces from 2D floor plans requires high-precision semantic segmentation of structural elements, particularly walls. However, existing methods optimized for standard metrics often struggle to detect thin structural components and yield masks with irregular boundaries, lacking the geometric precision required for subsequent vectorization. To address this issue, we introduce MitUNet, a hybrid neural network architecture specifically designed for wall segmentation tasks in the context of 3D modeling. In MitUNet, we utilize a hierarchical Mix-Transformer encoder to capture global context and a U-Net decoder enhanced with scSE attention blocks for precise boundary recovery. Furthermore, we propose an optimization strategy based on the Tversky loss function to effectively balance precision and recall. By fine-tuning the hyperparameters of the loss function, we prioritize the suppression of false positive noise along wall boundaries while maintaining high sensitivity to thin structures. Our experiments on the public CubiCasa5k dataset and a proprietary regional dataset demonstrate that the proposed approach ensures the generation of structurally correct masks with high boundary accuracy, outperforming standard single-task models. MitUNet provides a robust tool for data preparation in automated 3D reconstruction pipelines.

3.87WISE: Weighted Iterative Society-of-Experts for Robust Multimodal Multi-Agent Debate¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

3.147CREST: Universal Safety Guardrails Through Cluster-Guided Cross-Lingual Transfer¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

Ensuring content safety in large language models (LLMs) is essential for their deployment in real-world applications. However, existing safety guardrails are predominantly tailored for high-resource languages, leaving a significant portion of the world’s population underrepresented who communicate in low-resource languages. To address this, we introduce CREST (CRoss-lingual Efficient Safety Transfer), a parameter-efficient multilingual safety classification model that supports 100 languages with only 0.5B parameters. By training on a strategically chosen subset of only 13 high-resource languages, our model utilizes cluster-based cross-lingual transfer from a few to 100 languages, enabling effective generalization to both unseen high-resource and low-resource languages. This approach addresses the challenge of limited training data in low-resource settings. We conduct comprehensive evaluations across six safety benchmarks to demonstrate that CREST outperforms existing state-of-the-art guardrails of comparable scale and achieves competitive results against models with significantly larger parameter counts (2.5B parameters and above). Our findings highlight the limitations of language-specific guardrails and underscore the importance of developing universal, language-agnostic safety systems that can scale effectively to serve global populations.

arxiv html pdf kimi

2025/12/03 05:00 GTM

AI self-evolution has long been envisioned as a path toward superintelligence, where models autonomously acquire, refine, and internalize knowledge from their own learning experiences. Yet in practice, unguided self-evolving systems often plateau quickly or even degrade as training progresses. These failures arise from issues such as concept drift, diversity collapse, and mis-evolution, as models reinforce their own biases and converge toward low-entropy behaviors. To enable models to self-evolve in a stable and controllable manner while minimizing reliance on human supervision, we introduce R-Few, a guided Self-Play Challenger-Solver framework that incorporates lightweight human oversight through in-context grounding and mixed training. At each iteration, the Challenger samples a small set of human-labeled examples to guide synthetic question generation, while the Solver jointly trains on human and synthetic examples under an online, difficulty-based curriculum. Across math and general reasoning benchmarks, R-Few achieves consistent and iterative improvements. For example, Qwen3-8B-Base improves by +3.0 points over R-Zero on math tasks and achieves performance on par with General-Reasoner, despite the latter being trained on 20 times more human data. Ablation studies confirm the complementary contributions of grounded challenger training and curriculum-based solver training, and further analysis shows that R-Few mitigates drift, yielding more stable and controllable co-evolutionary dynamics.

3.157See, Think, Learn: A Self-Taught Multimodal Reasoner¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

3.158When Refusals Fail: Unstable Safety Mechanisms in Long-Context LLM Agents¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

Solving complex or long-horizon problems often requires large language models (LLMs) to use external tools and operate over a significantly longer context window. New LLMs enable longer context windows and support tool calling capabilities. Prior works have focused mainly on evaluation of LLMs on long-context prompts, leaving agentic setup relatively unexplored, both from capability and safety perspectives. Our work addresses this gap. We find that LLM agents could be sensitive to length, type, and placement of the context, exhibiting unexpected and inconsistent shifts in task performance and in refusals to execute harmful requests. Models with 1M-2M token context windows show severe degradation already at 100K tokens, with performance drops exceeding 50% for both benign and harmful tasks. Refusal rates shift unpredictably: GPT-4.1-nano increases from $\sim$ 5% to $\sim$ 40% while Grok 4 Fast decreases from $\sim$ 80% to $\sim$ 10% at 200K tokens. Our work shows potential safety issues with agents operating on longer context and opens additional questions on the current metrics and paradigm for evaluating LLM agent safety on long multi-step tasks. In particular, our results on LLM agents reveal a notable divergence in both capability and safety performance compared to prior evaluations of LLMs on similar criteria.

3.188Experimental Characterization of Fingertip Trajectory following for a 3-DoF Series-Parallel Hybrid Robotic Finger¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

Task-space control of robotic fingers is a critical enabler of dexterous manipulation, as manipulation objectives are most naturally specified in terms of fingertip motions and applied forces rather than individual joint angles. While task-space planning and control have been extensively studied for larger, arm-scale manipulators, demonstrations of precise task-space trajectory tracking in compact, multi-DoF robotic fingers remain scarce. In this paper, we present the physical prototyping and experimental characterization of a three-degree-of-freedom, linkage-driven, series-parallel robotic finger with analytic forward kinematics and a closed-form Jacobian. A resolved motion rate control (RMRC) scheme is implemented to achieve closed-loop task-space trajectory tracking. We experimentally evaluate the fingertip tracking performance across a variety of trajectories, including straight lines, circles, and more complex curves, and report millimeter-level accuracy. To the best of our knowledge, this work provides one of the first systematic experimental demonstrations of precise task-space trajectory tracking in a linkage-driven robotic finger, thereby establishing a benchmark for future designs aimed at dexterous in-hand manipulation.

3.189VLA Models Are More Generalizable Than You Think: Revisiting Physical and Spatial Modeling¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

Vision-language-action (VLA) models achieve strong in-distribution performance but degrade sharply under novel camera viewpoints and visual perturbations. We show that this brittleness primarily arises from misalignment in Spatial Modeling, rather than Physical Modeling. To address this, we propose a one-shot adaptation framework that recalibrates visual representations through lightweight, learnable updates. Our first method, Feature Token Modulation (FTM), applies a global affine transformation to visual tokens and improves Libero viewpoint accuracy from 48.5% to 87.1% with only 4K parameters. Building on this, Feature Linear Adaptation (FLA) introduces low-rank updates to the ViT encoder, achieving 90.8% success with 4.7M parameters -- matching LoRA-scale finetuning at far lower cost. Together, these results reveal substantial untapped robustness in pretrained VLA models and demonstrate that targeted, minimal visual adaptation is sufficient to restore viewpoint generalization.

3.190Polar Perspectives: Evaluating 2-D LiDAR Projections for Robust Place Recognition with Visual Foundation Models¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

arxiv html pdf kimi

2025/12/03 05:00 GTM

Visual traversability estimation is critical for autonomous navigation, but existing VLM-based methods rely on hand-crafted prompts, generalize poorly across embodiments, and output only traversability maps, leaving trajectory generation to slow external planners. We propose SwarmDiffusion, a lightweight end-to-end diffusion model that jointly predicts traversability and generates a feasible trajectory from a single RGB image. To remove the need for annotated or planner-produced paths, we introduce a planner-free trajectory construction pipeline based on randomized waypoint sampling, Bezier smoothing, and regularization enforcing connectivity, safety, directionality, and path thinness. This enables learning stable motion priors without demonstrations. SwarmDiffusion leverages VLM-derived supervision without prompt engineering and conditions the diffusion process on a compact embodiment state, producing physically consistent, traversable paths that transfer across different robot platforms. Across indoor environments and two embodiments (quadruped and aerial), the method achieves 80-100% navigation success and 0.09 s inference, and adapts to a new robot using only-500 additional visual samples. It generalizes reliably to unseen environments in simulation and real-world trials, offering a scalable, prompt-free approach to unified traversability reasoning and trajectory generation.

3.192VLM as Strategist: Adaptive Generation of Safety-critical Testing Scenarios via Guided Diffusion¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

The safe deployment of autonomous driving systems (ADSs) relies on comprehensive testing and evaluation. However, safety-critical scenarios that can effectively expose system vulnerabilities are extremely sparse in the real world. Existing scenario generation methods face challenges in efficiently constructing long-tail scenarios that ensure fidelity, criticality, and interactivity, while particularly lacking real-time dynamic response capabilities to the vehicle under test (VUT). To address these challenges, this paper proposes a safety-critical testing scenario generation framework that integrates the high-level semantic understanding capabilities of Vision Language Models (VLMs) with the fine-grained generation capabilities of adaptive guided diffusion models. The framework establishes a three-layer hierarchical architecture comprising a strategic layer for VLM-directed scenario generation objective determination, a tactical layer for guidance function formulation, and an operational layer for guided diffusion execution. We first establish a high-quality fundamental diffusion model that learns the data distribution of real driving scenarios. Next, we design an adaptive guided diffusion method that enables real-time, precise control of background vehicles (BVs) in closed-loop simulation. The VLM is then incorporated to autonomously generate scenario generation objectives and guidance functions through deep scenario understanding and risk reasoning, ultimately guiding the diffusion model to achieve VLM-directed scenario generation. Experimental results demonstrate that the proposed method can efficiently generate realistic, diverse, and highly interactive safety-critical testing scenarios. Furthermore, case studies validate the adaptability and VLM-directed generation performance of the proposed method.

3.193Steering Vision-Language-Action Models as Anti-Exploration: A Test-Time Scaling Approach¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

Vision-Language-Action (VLA) models, trained via flow-matching or diffusion objectives, excel at learning complex behaviors from large-scale, multi-modal datasets (e.g., human teleoperation, scripted policies). However, since VLAs incorporate diverse data modes in the pre-training stage, and the finetuning dataset often contains demonstration data collected in a kinematically suboptimal or undesirable way, it exists redundant action modes that are irrelevant to the success action modes of the downstream task. Specifically, we observe a critical inference-time fragility among various sampled noises after supervised finetuning of pre-trained VLAs. In this paper, we attribute this instability to the distribution shift between the VLA policy and the policy induced by stable success modes of the downstream task dataset. Thus, we propose \textbf{TACO}, a test-time-scaling (TTS) framework that applies a lightweight pseudo-count estimator as a high-fidelity verifier of action chunks. The VLA models integrated with TACO can execute the actions with maximum pseudo-count from all sampled action chunks, thereby preventing distribution shifts while preserving the generalization ability of VLAs since the constraint is applied only during inference. Our method resembles the classical anti-exploration principle in offline reinforcement learning (RL), and being gradient-free, it incurs significant computational benefits compared to RL update, especially for flow or diffusion-based VLAs which are difficult to perform RL update due to denoising process. Extensive experiments across four simulation benchmarks (RoboTwin2.0, Robotwin, LIBERO, SimplerEnv) and a dual-arm platform demonstrate that our method significantly improves the inference stability and success rates in downstream-task adaptations.

3.194Phase-Adaptive LLM Framework with Multi-Stage Validation for Construction Robot Task Allocation: A Systematic Benchmark Against Traditional Optimizati...¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

Multi-robot task allocation in construction automation has traditionally relied on optimization methods such as Dynamic Programming and Reinforcement Learning. This research introduces the LangGraph-based Task Allocation Agent (LTAA), an LLM-driven framework that integrates phase-adaptive allocation strategies, multi-stage validation with hierarchical retries, and dynamic prompting for efficient robot coordination. Although recent LLM approaches show potential for construction robotics, they largely lack rigorous validation and benchmarking against established algorithms. This paper presents the first systematic comparison of LLM-based task allocation with traditional methods in construction scenarios.The study validates LLM feasibility through SMART-LLM replication and addresses implementation challenges using a Self-Corrective Agent Architecture. LTAA leverages natural-language reasoning combined with structured validation mechanisms, achieving major computational gains reducing token usage by 94.6% and allocation time by 86% through dynamic prompting. The framework adjusts its strategy across phases: emphasizing execution feasibility early and workload balance in later allocations.The authors evaluate LTAA against Dynamic Programming, Q-learning, and Deep Q-Network (DQN) baselines using construction operations from the TEACh human-robot collaboration dataset. In the Heavy Excels setting, where robots have strong task specializations, LTAA achieves 77% task completion with superior workload balance, outperforming all traditional methods. These findings show that LLM-based reasoning with structured validation can match established optimization algorithms while offering additional advantages such as interpretability, adaptability, and the ability to update task logic without retraining.

3.195Diagnose, Correct, and Learn from Manipulation Failures via Visual Symbols¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

3.196CogDrive: Cognition-Driven Multimodal Prediction-Planning Fusion for Safe Autonomy¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

Safe autonomous driving in mixed traffic requires a unified understanding of multimodal interactions and dynamic planning under uncertainty. Existing learning based approaches struggle to capture rare but safety critical behaviors, while rule based systems often lack adaptability in complex interactions. To address these limitations, CogDrive introduces a cognition driven multimodal prediction and planning framework that integrates explicit modal reasoning with safety aware trajectory optimization. The prediction module adopts cognitive representations of interaction modes based on topological motion semantics and nearest neighbor relational encoding. With a differentiable modal loss and multimodal Gaussian decoding, CogDrive learns sparse and unbalanced interaction behaviors and improves long horizon trajectory prediction. The planning module incorporates an emergency response concept and optimizes safety stabilized trajectories, where short term consistent branches ensure safety during replanning cycles and long term branches support smooth and collision free motion under low probability switching modes. Experiments on Argoverse2 and INTERACTION datasets show that CogDrive achieves strong performance in trajectory accuracy and miss rate, while closed loop simulations confirm adaptive behavior in merge and intersection scenarios. By combining cognitive multimodal prediction with safety oriented planning, CogDrive offers an interpretable and reliable paradigm for safe autonomy in complex traffic.

3.197RoboWheel: A Data Engine from Real-World Human Demonstrations for Cross-Embodiment Robotic Learning¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

We introduce Robowheel, a data engine that converts human hand object interaction (HOI) videos into training-ready supervision for cross morphology robotic learning. From monocular RGB or RGB-D inputs, we perform high precision HOI reconstruction and enforce physical plausibility via a reinforcement learning (RL) optimizer that refines hand object relative poses under contact and penetration constraints. The reconstructed, contact rich trajectories are then retargeted to cross-embodiments, robot arms with simple end effectors, dexterous hands, and humanoids, yielding executable actions and rollouts. To scale coverage, we build a simulation-augmented framework on Isaac Sim with diverse domain randomization (embodiments, trajectories, object retrieval, background textures, hand motion mirroring), which enriches the distributions of trajectories and observations while preserving spatial relationships and physical plausibility. The entire data pipeline forms an end to end pipeline from video,reconstruction,retargeting,augmentation data acquisition. We validate the data on mainstream vision language action (VLA) and imitation learning architectures, demonstrating that trajectories produced by our pipeline are as stable as those from teleoperation and yield comparable continual performance gains. To our knowledge, this provides the first quantitative evidence that HOI modalities can serve as effective supervision for robotic learning. Compared with teleoperation, Robowheel is lightweight, a single monocular RGB(D) camera is sufficient to extract a universal, embodiment agnostic motion representation that could be flexibly retargeted across embodiments. We further assemble a large scale multimodal dataset combining multi-camera captures, monocular videos, and public HOI corpora for training and evaluating embodied models.

arxiv html pdf kimi

2025/12/03 05:00 GTM

3.199Reframing Human-Robot Interaction Through Extended Reality: Unlocking Safer, Smarter, and More Empathic Interactions with Virtual Robots and Foundatio...¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

This perspective reframes human-robot interaction (HRI) through extended reality (XR), arguing that virtual robots powered by large foundation models (FMs) can serve as cognitively grounded, empathic agents. Unlike physical robots, XR-native agents are unbound by hardware constraints and can be instantiated, adapted, and scaled on demand, while still affording embodiment and co-presence. We synthesize work across XR, HRI, and cognitive AI to show how such agents can support safety-critical scenarios, socially and cognitively empathic interaction across domains, and outreaching physical capabilities with XR and AI integration. We then discuss how multimodal large FMs (e.g., large language model, large vision model, and vision-language model) enable context-aware reasoning, affect-sensitive situations, and long-term adaptation, positioning virtual robots as cognitive and empathic mediators rather than mere simulation assets. At the same time, we highlight challenges and potential risks, including overtrust, cultural and representational bias, privacy concerns around biometric sensing, and data governance and transparency. The paper concludes by outlining a research agenda for human-centered, ethically grounded XR agents - emphasizing multi-layered evaluation frameworks, multi-user ecosystems, mixed virtual-physical embodiment, and societal and ethical design practices to envision XR-based virtual agents powered by FMs as reshaping future HRI into a more efficient and adaptive paradigm.

3.200Robotic capabilities framework: A boundary object and intermediate-level knowledge artifact for co-designing robotic processes¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

As robots become more adaptable, responsive, and capable of interacting with humans, the design of effective human-robot collaboration becomes critical. Yet, this design process is typically led by monodisciplinary approaches, often overlooking interdisciplinary knowledge and the experiential knowledge of workers who will ultimately share tasks with these systems. To address this gap, we introduce the robotic capabilities framework, a vocabulary that enables transdisciplinary collaborations to meaningfully shape the future of work when robotic systems are integrated into the workplace. Rather than focusing on the internal workings of robots, the framework centers discussion on high-level capabilities, supporting dialogue around which elements of a task should remain human-led and which can be delegated to robots. We developed the framework through reflexive and iterative processes, and applied it in two distinct settings: by engaging roboticists in describing existing commercial robots using its vocabulary, and through a design activity with students working on robotics-related projects. The framework emerges as an intermediate-level knowledge artifact and a boundary object that bridges technical and experiential domains, guiding designers, empowering workers, and contributing to more just and collaborative futures of work.

3.201AID: Agent Intent from Diffusion for Multi-Agent Informative Path Planning¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

Information gathering in large-scale or time-critical scenarios (e.g., environmental monitoring, search and rescue) requires broad coverage within limited time budgets, motivating the use of multi-agent systems. These scenarios are commonly formulated as multi-agent informative path planning (MAIPP), where multiple agents must coordinate to maximize information gain while operating under budget constraints. A central challenge in MAIPP is ensuring effective coordination while the belief over the environment evolves with incoming measurements. Recent learning-based approaches address this by using distributions over future positions as “intent” to support coordination. However, these autoregressive intent predictors are computationally expensive and prone to compounding errors. Inspired by the effectiveness of diffusion models as expressive, long-horizon policies, we propose AID, a fully decentralized MAIPP framework that leverages diffusion models to generate long-term trajectories in a non-autoregressive manner. AID first performs behavior cloning on trajectories produced by existing MAIPP planners and then fine-tunes the policy using reinforcement learning via Diffusion Policy Policy Optimization (DPPO). This two-stage pipeline enables the policy to inherit expert behavior while learning improved coordination through online reward feedback. Experiments demonstrate that AID consistently improves upon the MAIPP planners it is trained from, achieving up to 4x faster execution and 17% increased information gain, while scaling effectively to larger numbers of agents. Our implementation is publicly available at https://github.com/marmotlab/AID.

arxiv html pdf kimi

2025/12/03 05:00 GTM

3.206Property-Guided Cyber-Physical Reduction and Surrogation for Safety Analysis in Robotic Vehicles¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

We propose a methodology for falsifying safety properties in robotic vehicle systems through property-guided reduction and surrogate execution. By isolating only the control logic and physical dynamics relevant to a given specification, we construct lightweight surrogate models that preserve property-relevant behaviors while eliminating unrelated system complexity. This enables scalable falsification via trace analysis and temporal logic oracles. We demonstrate the approach on a drone control system containing a known safety flaw. The surrogate replicates failure conditions at a fraction of the simulation cost, and a property-guided fuzzer efficiently discovers semantic violations. Our results suggest that controller reduction, when coupled with logic-aware test generation, provides a practical and scalable path toward semantic verification of cyber-physical systems.

3.207Robust Geospatial Coordination of Multi-Agent Communications Networks Under Attrition¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

Fast, efficient, robust communication during wildfire and other emergency responses is critical. One way to achieve this is by coordinating swarms of autonomous aerial vehicles carrying communications equipment to form an ad-hoc network connecting emergency response personnel to both each other and central command. However, operating in such extreme environments may lead to individual networking agents being damaged or rendered inoperable, which could bring down the network and interrupt communications. To overcome this challenge and enable multi-agent UAV networking in difficult environments, this paper introduces and formalizes the problem of Robust Task Networking Under Attrition (RTNUA), which extends connectivity maintenance in multi-robot systems to explicitly address proactive redundancy and attrition recovery. We introduce Physics-Informed Robust Employment of Multi-Agent Networks ( $Φ$ IREMAN), a topological algorithm leveraging physics-inspired potential fields to solve this problem. Through simulation across 25 problem configurations, $Φ$ IREMAN consistently outperforms the DCCRS baseline, and on large-scale problems with up to 100 tasks and 500 drones, maintains $>99.9\%$ task uptime despite substantial attrition, demonstrating both effectiveness and scalability.

3.208Reinforcement Learning for Robotic Safe Control with Force Sensing¶

arxiv html pdf kimi

2025/12/03 05:00 GTM

For the task with complicated manipulation in unstructured environments, traditional hand-coded methods are ineffective, while reinforcement learning can provide more general and useful policy. Although the reinforcement learning is able to obtain impressive results, its stability and reliability is hard to guarantee, which would cause the potential safety threats. Besides, the transfer from simulation to real world also will lead in unpredictable situations. To enhance the safety and reliability of robots, we introduce the force and haptic perception into reinforcement learning. Force and tactual sensation play key roles in robotic dynamic control and human-robot interaction. We demonstrate that the force-based reinforcement learning method can be more adaptive to environment, especially in sim-to-real transfer. Experimental results show in object pushing task, our strategy is safer and more efficient in both simulation and real world, thus it holds prospects for a wide variety of robotic applications.

1 Daily News

2feed¶

2.1悟空杀进NeurIPS！把七十二变写成数学公式，这才是我们想看的“东方魔法”¶

2.2差分隐私失效！揭秘GNN拓扑推断攻击：PGR用「零重叠」合成图破解危局¶

3paper¶

3.1OneThinker: All-in-one Reasoning Model for Image and Video¶

3.2MultiShotMaster: A Controllable Multi-Shot Video Generation Framework¶

3.3U4D: Uncertainty-Aware 4D World Modeling from LiDAR Sequences¶

3.4InEx: Hallucination Mitigation via Introspection and Cross-Modal Multi-Agent Collaboration¶

3.5Contextual Image Attack: How Visual Context Exposes Multimodal Safety Vulnerabilities¶

3.6BEVDilation: LiDAR-Centric Multi-Modal Fusion for 3D Object Detection¶

3.7A Lightweight Real-Time Low-Light Enhancement Network for Embedded Automotive Vision Systems¶

3.8Layout Anything: One Transformer for Universal Room Layout Estimation¶

3.9Benchmarking Scientific Understanding and Reasoning for Video Generation using VideoScience-Bench¶

3.10LoVoRA: Text-guided and Mask-free Video Object Removal and Addition with Learnable Object-aware Localization¶

3.11EGGS: Exchangeable 2D/3D Gaussian Splatting for Geometry-Appearance Balanced Novel View Synthesis¶

3.12DiverseAR: Boosting Diversity in Bitwise Autoregressive Image Generation¶

3.13Learning Multimodal Embeddings for Traffic Accident Prediction and Causal Estimation¶

3.14MRD: Multi-resolution Retrieval-Detection Fusion for High-Resolution Image Understanding¶

3.15Glance: Accelerating Diffusion Models with 1 Sample¶

3.16Polar Perspectives: Evaluating 2-D LiDAR Projections for Robust Place Recognition with Visual Foundation Models¶

3.17MindGPT-4ov: An Enhanced MLLM via a Multi-Stage Post-Training Paradigm¶

3.18Taming Camera-Controlled Video Generation with Verifiable Geometry Reward¶

3.19MICCAI STSR 2025 Challenge: Semi-Supervised Teeth and Pulp Segmentation and CBCT-IOS Registration¶

3.20RFOP: Rethinking Fusion and Orthogonal Projection for Face-Voice Association¶

3.21Are Detectors Fair to Indian IP-AIGC? A Cross-Generator Study¶

3.22Action Anticipation at a Glimpse: To What Extent Can Multimodal Cues Replace Video?¶

3.23ReVSeg: Incentivizing the Reasoning Chain for Video Segmentation with Reinforcement Learning¶

3.24Defense That Attacks: How Robust Models Become Better Attackers¶

3.25HUD: Hierarchical Uncertainty-Aware Disambiguation Network for Composed Video Retrieval¶

3.26TrackNetV5: Residual-Driven Spatio-Temporal Refinement and Motion Direction Decoupling for Fast Object Tracking¶

3.27Diagnose, Correct, and Learn from Manipulation Failures via Visual Symbols¶

3.28LumiX: Structured and Coherent Text-to-Intrinsic Generation¶

3.29Rethinking Surgical Smoke: A Smoke-Type-Aware Laparoscopic Video Desmoking Method and Dataset¶

3.30AttMetNet: Attention-Enhanced Deep Neural Network for Methane Plume Detection in Sentinel-2 Satellite Imagery¶

3.31Reasoning-Aware Multimodal Fusion for Hateful Video Detection¶

3.32Beyond Paired Data: Self-Supervised UAV Geo-Localization from Reference Imagery Alone¶

3.33DF-Mamba: Deformable State Space Modeling for 3D Hand Pose Estimation in Interactions¶

3.34Emergent Bayesian Behaviour and Optimal Cue Combination in LLMs¶

3.35GeoViS: Geospatially Rewarded Visual Search for Remote Sensing Visual Grounding¶

3.36Tissue-mask supported inter-subject whole-body image registration in the UK Biobank - A method benchmarking study¶

3.37VLM-Pruner: Buffering for Spatial Sparsity in an Efficient VLM Centrifugal Token Pruning Paradigm¶

3.38GeoBridge: A Semantic-Anchored Multi-View Foundation Model Bridging Images and Text for Geo-Localization¶

3.39ALDI-ray: Adapting the ALDI Framework for Security X-ray Object Detection¶

3.40ClimaOoD: Improving Anomaly Segmentation via Physically Realistic Synthetic Data¶

3.41Unsupervised Structural Scene Decomposition via Foreground-Aware Slot Attention with Pseudo-Mask Guidance¶

3.42PGP-DiffSR: Phase-Guided Progressive Pruning for Efficient Diffusion-based Image Super-Resolution¶

3.43UAUTrack: Towards Unified Multimodal Anti-UAV Visual Tracking¶

3.44PolarGuide-GSDR: 3D Gaussian Splatting Driven by Polarization Priors and Deferred Reflection for Real-World Reflective Scenes¶

3.45Spatially-Grounded Document Retrieval via Patch-to-Region Relevance Propagation¶

3.46Real-Time Multimodal Data Collection Using Smartwatches and Its Visualization in Education¶

3.47Hear What Matters! Text-conditioned Selective Video-to-Audio Generation¶

3.48PoreTrack3D: A Benchmark for Dynamic 3D Gaussian Splatting in Pore-Scale Facial Trajectory Tracking¶

3.49Leveraging Large-Scale Pretrained Spatial-Spectral Priors for General Zero-Shot Pansharpening¶

3.50Joint Distillation for Fast Likelihood Evaluation and Sampling in Flow-based Models¶

3.51PPTBench: Towards Holistic Evaluation of Large Language Models for PowerPoint Layout and Design Understanding¶

3.52RULER-Bench: Probing Rule-based Reasoning Abilities of Next-level Video Generation Models for Vision Foundation Intelligence¶

3.53Content-Aware Texturing for Gaussian Splatting¶

3.54SAM2Grasp: Resolve Multi-modal Grasping via Prompt-conditioned Temporal Action Prediction¶

3.55Co-speech Gesture Video Generation via Motion-Based Graph Retrieval¶

3.56From Panel to Pixel: Zoom-In Vision-Language Pretraining from Biomedical Scientific Literature¶

3.57OmniPerson: Unified Identity-Preserving Pedestrian Generation¶

3.58AVGGT: Rethinking Global Attention for Accelerating VGGT¶

3.59WeMMU: Enhanced Bridging of Vision-Language Models and Diffusion Models via Noisy Query Tokens¶

3.60On the Problem of Consistent Anomalies in Zero-Shot Anomaly Detection¶

3.61SkyMoE: A Vision-Language Foundation Model for Enhancing Geospatial Interpretation with Mixture of Experts¶

3.62Two-Stage Vision Transformer for Image Restoration: Colorization Pretraining + Residual Upsampling¶

3.63GeoDiT: A Diffusion-based Vision-Language Model for Geospatial Understanding¶

3.64dots.ocr: Multilingual Document Layout Parsing in a Single Vision-Language Model¶

3.65A Large Scale Benchmark for Test Time Adaptation Methods in Medical Image Segmentation¶

3.66Attention-guided reference point shifting for Gaussian-mixture-based partial point set registration¶

3.67YingVideo-MV: Music-Driven Multi-Stage Video Generation¶

3.68Masking Matters: Unlocking the Spatial Reasoning Capabilities of LLMs for 3D Scene-Language Understanding¶

3.69UCAgents: Unidirectional Convergence for Visual Evidence Anchored Multi-Agent Medical Decision-Making¶

3.70G-SHARP: Gaussian Surgical Hardware Accelerated Real-time Pipeline¶

3.71WorldPack: Compressed Memory Improves Spatial Consistency in Video World Modeling¶

3.72TGDD: Trajectory Guided Dataset Distillation with Balanced Distribution¶

3.73Vision to Geometry: 3D Spatial Memory for Sequential Embodied MLLM Reasoning and Exploration¶

3.74Does Hearing Help Seeing? Investigating Audio-Video Joint Denoising for Video Generation¶

3.75See, Think, Learn: A Self-Taught Multimodal Reasoner¶

3.88Nav- $R^2$ Dual-Relation Reasoning for Generalizable Open-Vocabulary Object-Goal Navigation¶