搜索结果：environments

Decoherence is often modeled using Markovian master equations that predict exponential suppression of coherence and are frequently used as effective bounds on quantum behavior in complex environments. Such descriptions, however, correspond to the singular physical limit of vanishing environmental memory. Here we formulate decoherence using a general time-nonlocal decoherence functional determined solely by the environmental force correlation function, with Markovian dynamics recovered explicitly as a limiting case. For arbitrary stationary environments with finite temporal correlations, we show that the decoherence functional exhibits quadratic short-time growth that is model-independent within the finite-memory class considered. Consequently, the decoherence time defined operationally-without assuming exponential decay-scales as the square root of the environmental correlation time, independent of the detailed form of the bath correlation kernel. These results are illustrated analytically for Gaussian-correlated, soft power-law, and Ornstein-Uhlenbeck environments. In the Ornstein-Uhlenbeck case, the non-Markovian dynamics admit an exact analytical closure, yielding a closed evolu

Aligning autonomous agents with human intent remains a central challenge in modern AI. A key manifestation of this challenge is reward hacking, whereby agents appear successful under the evaluation signal while violating the intended objective. Reward hacking has been observed across a wide range of settings, yet methods for reliably measuring it at scale remain lacking. In this work, we introduce a new evaluation paradigm for measuring reward hacking. Whereas prior studies have primarily analyzed it post hoc by inspecting agent trajectories, we instead embed detectable reward hacking opportunities directly into environments. This makes their exploitation verifiable by design, enabling deterministic and automated measurement of whether and how agents exploit such vulnerabilities. We instantiate this approach in $\textit{TextArena}$ and release $\textit{Hack-Verifiable TextArena}$, a testbed in which reward hacking can be measured reliably. Using this benchmark, we analyze reward hacking behavior across language models in diverse environments and settings. We open source the code at https://github.com/MajoRoth/hack-verifiable-environments/.

Making calibrated online predictions is a central challenge in modern AI systems. Much of the existing literature focuses on fully adversarial environments where outcomes may be arbitrary, leading to conservative algorithms that can perform suboptimally in more benign settings, such as when outcomes are nearly stationary. This gap raises a natural question: can we design online prediction algorithms whose calibration error automatically adapts to the degree of non-stationarity in the environment, smoothly interpolating between i.i.d. and adversarial regimes? We answer this question in the affirmative and develop a suite of algorithms that achieve adaptive calibration guarantees under multiple calibration measures. Specifically, with $T$ being the number of rounds, $K$ being the unknown number of i.i.d. segments of the environment, and $C\in[0,T]$ being another unknown non-stationary measure defined as the minimal $\ell_1$ deviation of the mean outcomes, our algorithms attain $\widetilde{O}(\min\{\sqrt{T}+(TC)^{\frac{1}{3}}, \sqrt{KT}\})$ for $\ell_1$ calibration error and $\widetilde{O}(\min\{(1+C)^{\frac{1}{3}}, K\})$ for both $\ell_2$ and pseudo KL calibration error. These bounds

Large language model (LLM) agents have shown impressive capabilities in human language comprehension and reasoning, yet their potential in cybersecurity remains underexplored. We introduce DefenderBench, a practical, open-source toolkit for evaluating language agents across offense, defense, and cybersecurity knowledge-based tasks. DefenderBench includes environments for network intrusion, malicious content detection, code vulnerability analysis, and cybersecurity knowledge assessment. It is intentionally designed to be affordable and easily accessible for researchers while providing fair and rigorous assessment. We benchmark several state-of-the-art (SoTA) and popular LLMs, including both open- and closed-weight models, using a standardized agentic framework. Our results show that Claude-3.7-sonnet performs best with a DefenderBench score of 81.65, followed by Claude-3.7-sonnet-think with 78.40, while the best open-weight model, Llama 3.3 70B, is not far behind with a DefenderBench score of 71.81. DefenderBench's modular design allows seamless integration of custom LLMs and tasks, promoting reproducibility and fair comparisons. An anonymized version of DefenderBench is available a

We study how dark matter environments influence nonlinear gravitational memory from intermediate-mass-ratio binaries. Incorporating environmental effects from the dark matter gravitational potential, dynamical friction, and accretion, we compute the leading-order nonlinear memory for both bound and unbound orbits under dark matter minispikes and Navarro-Frenk-White haloes. For quasi-circular inspirals in a minispike, we additionally include an empirical prescription for the time-dependent evolution of the dark matter profile, which gradually evolves along the inspiral and captures the cumulative environmental response. We find that dark matter can modify the time evolution and mode content of the memory relative to the vacuum case, with the cumulative effect depending sensitively on the density profile and on how the environment accelerates the inspiral. We use these waveforms to assess signal-to-noise ratios and mismatches in representative space-based detector configurations, highlighting where memory-driven differences may be large enough to warrant targeted parameter-estimation studies. Our results emphasize that astrophysical environments can leave a hereditary imprint on grav

Despite extensive research in robust visual-inertial navigation systems~(VINS) in dynamic environments, many approaches remain vulnerable to objects that suddenly start moving, which are referred to as \textit{abruptly dynamic objects}. In addition, most approaches have considered the effect of dynamic objects only at the feature association level. In this study, we observed that the state estimation diverges when errors from false correspondences owing to moving objects incorrectly propagate into the IMU bias terms. To overcome these problems, we propose a robust VINS framework called \mbox{\textit{DynaVINS++}}, which employs a) adaptive truncated least square method that adaptively adjusts the truncation range using both feature association and IMU preintegration to effectively minimize the effect of the dynamic objects while reducing the computational cost, and b)~stable state recovery with bias consistency check to correct misestimated IMU bias and to prevent the divergence caused by abruptly dynamic objects. As verified in both public and real-world datasets, our approach shows promising performance in dynamic environments, including scenes with abruptly dynamic objects.

We analyze a set of volume limited samples from the SDSS to study the dependence of galaxy colour on different environments of the cosmic web. We measure the local dimension of galaxies to determine the geometry of their embedding environments and find that filaments host a higher fraction of red galaxies than sheets at each luminosity. We repeat the analysis at a fixed density and recover the same trend which shows that galaxy colours depend on geometry of environments besides local density. At a fixed luminosity, the fraction of red galaxies in filaments and sheets increases with the extent of these environments. This suggests that the bigger structures have a larger baryon reservoir favouring higher accretion and larger stellar mass. We find that the mean colour of the red and blue populations are systematically higher in the environments with smaller local dimension and increases monotonically in all the environments with luminosity. We observe that the bimodal nature of the galaxy colour distribution persists in all environments and all luminosities, which suggests that the transformation from blue to red galaxy can occur in all environments.

Open-Vocabulary Mobile Manipulation (OVMM) is a crucial capability for autonomous robots, especially when faced with the challenges posed by unknown and dynamic environments. This task requires robots to explore and build a semantic understanding of their surroundings, generate feasible plans to achieve manipulation goals, adapt to environmental changes, and comprehend natural language instructions from humans. To address these challenges, we propose a novel framework that leverages the zero-shot detection and grounded recognition capabilities of pretraining visual-language models (VLMs) combined with dense 3D entity reconstruction to build 3D semantic maps. Additionally, we utilize large language models (LLMs) for spatial region abstraction and online planning, incorporating human instructions and spatial semantic context. We have built a 10-DoF mobile manipulation robotic platform JSR-1 and demonstrated in real-world robot experiments that our proposed framework can effectively capture spatial semantics and process natural language user instructions for zero-shot OVMM tasks under dynamic environment settings, with an overall navigation and task success rate of 80.95% and 73.33% o

In this paper we tackle the problem of persistently covering a complex non-convex environment with a team of robots. We consider scenarios where the coverage quality of the environment deteriorates with time, requiring to constantly revisit every point. As a first step, our solution finds a partition of the environment where the amount of work for each robot, weighted by the importance of each point, is equal. This is achieved using a power diagram and finding an equitable partition through a provably correct distributed control law on the power weights. Compared to other existing partitioning methods, our solution considers a continuous environment formulation with non-convex obstacles. In the second step, each robot computes a graph that gathers sweep-like paths and covers its entire partition. At each planning time, the coverage error at the graph vertices is assigned as weights of the corresponding edges. Then, our solution is capable of efficiently finding the optimal open coverage path through the graph with respect to the coverage error per distance traversed. Simulation and experimental results are presented to support our proposal.

Self-supervised monocular depth estimation has seen significant progress in recent years, especially in outdoor environments. However, depth prediction results are not satisfying in indoor scenes where most of the existing data are captured with hand-held devices. As compared to outdoor environments, estimating depth of monocular videos for indoor environments, using self-supervised methods, results in two additional challenges: (i) the depth range of indoor video sequences varies a lot across different frames, making it difficult for the depth network to induce consistent depth cues for training; (ii) the indoor sequences recorded with handheld devices often contain much more rotational motions, which cause difficulties for the pose network to predict accurate relative camera poses. In this work, we propose a novel framework-MonoIndoor++ by giving special considerations to those challenges and consolidating a set of good practices for improving the performance of self-supervised monocular depth estimation for indoor environments. First, a depth factorization module with transformer-based scale regression network is proposed to estimate a global depth scale factor explicitly, and t

This work uses multiscale environments and structures of galaxies in the Sloan Digital Sky Survey as consistency checks of the evolution from starburst to quiescence at redshift $z < 0.2$. The environmental indicators include fixed aperture mass overdensities ($δ_{x\mathrm{Mpc}}$, $x \in \{0.5, 1, 2, 4, 8\}\,h^{-1}$Mpc), $k$-nearest neighbor distances, the tidal parameter, halo mass ($M_h$), and satellite/central classification. The residuals of specific star formation rates ($Δ\,\mathrm{SSFR}$) is used to select starbursts ($Δ\,\mathrm{SSFR} > 0.6\,$dex, $N \approx 8,\,600$). Quenched post-starbursts (QPSBs) are selected using H$α< 3\,$angstrom in emission and H$δ_A > 4\,$ angstrom in absorption ($N \approx 750$). The environments of starbursts and QPSBs are compared with those of active galactic nuclei (AGNs) and inactive galaxies of varying $Δ\,\mathrm{SSFR}$. The environments of starbursts, AGNs, and QPSBs are unlike the environments of most quiescent galaxies (QGs). About $70\%-90\%$ of starbursts, AGNs with H$δ_A > 4$, and QPSBs are centrals, $\sim 80\%-90\%$ have $M_h < 10^{13}\,M_\odot$, and only $\sim 2\%-4\%$ have $M_h > 10^{14}\,M_\odot$ or live in c

Simulation engines are widely adopted in robotics. However, they lack either full simulation control, ROS integration, realistic physics, or photorealism. Recently, synthetic data generation and realistic rendering has advanced tasks like target tracking and human pose estimation. However, when focusing on vision applications, there is usually a lack of information like sensor measurements or time continuity. On the other hand, simulations for most robotics tasks are performed in (semi)static environments, with specific sensors and low visual fidelity. To solve this, we introduced in our previous work a fully customizable framework for generating realistic animated dynamic environments (GRADE) [1]. We use GRADE to generate an indoor dynamic environment dataset and then compare multiple SLAM algorithms on different sequences. By doing that, we show how current research over-relies on known benchmarks, failing to generalize. Our tests with refined YOLO and Mask R-CNN models provide further evidence that additional research in dynamic SLAM is necessary. The code, results, and generated data are provided as open-source at https://eliabntt.github.io/grade-rrSimulation of Dynamic Environ

Advances in computer networks and rendering systems facilitate the creation of distributed collaborative environments in which the distribution of information at remote locations allows efficient communication. One of the challenges in networked virtual environments is maintaining a consistent view of the shared state in the presence of inevitable network latency and jitter. A consistent view in a shared scene may significantly increase the sense of presence among participants and facilitate their interactivity. The dynamic shared state is directly affected by the frequency of actions applied on the objects in the scene. Mixed Reality (MR) and Virtual Reality (VR) environments contain several types of action producers including human users, a wide range of electronic motion sensors, and haptic devices. In this paper, the authors propose a novel criterion for categorization of distributed MR/VR systems and present an adaptive synchronization algorithm for distributed MR/VR collaborative environments. In spite of significant network latency, results show that for low levels of update frequencies the dynamic shared state can be maintained consistent at multiple remotely located sites.

We propose a new framework to improve automatic speech recognition (ASR) systems in resource-scarce environments using a generative adversarial network (GAN) operating on acoustic input features. The GAN is used to enhance the features of mismatched data prior to decoding, or can optionally be used to fine-tune the acoustic model. We achieve improvements that are comparable to multi-style training (MTR), but at a lower computational cost. With less than one hour of data, an ASR system trained on good quality data, and evaluated on mismatched audio is improved by between 11.5% and 19.7% relative word error rate (WER). Experiments demonstrate that the framework can be very useful in under-resourced environments where training data and computational resources are limited. The GAN does not require parallel training data, because it utilises a baseline acoustic model to provide an additional loss term that guides the generator to create acoustic features that are better classified by the baseline.

We have made a survey of quasar environments at 0.5 < z < 0.8, using a sample of both radio-loud and radio-quiet quasars matched in B-band luminosity. Our observations include images of background control fields to provide a good determination of the field galaxy counts. About 10 per cent of the quasars appear to live in rich clusters, whereas approximately 45 per cent live in environments similar to that of field galaxies. The richness of galaxies within a 0.5 Mpc radius around the radio-quiet quasars is found to be indistinguishable from the richness around the radio-loud quasars, corresponding on average to groups or poorer clusters of galaxies. Comparing the galaxy richness in the radio-loud quasar fields with quasar fields in the literature, we find no evidence of an evolution in the environment with epoch. Instead, a weak, but significant correlation between quasar radio luminosity and environmental richness is present. It is thus possible that the environments of quasars, at least the powerful ones, do not evolve much between the present epoch and z \approx 0.8.

The relationships between the listener, physical world and virtual environment (VE) should not only inspire the design of natural multimodal interfaces but should be discovered to make sense of the mediating action of VR technologies. This chapter aims to transform an archipelago of studies related to sonic interactions in virtual environments (SIVE) into a research field equipped with a first theoretical framework with an inclusive vision of the challenges to come: the egocentric perspective of the auditory digital twin. In a VE with immersive audio technologies implemented, the role of VR simulations must be enacted by a participatory exploration of sense-making in a network of human and non-human agents, called actors. The guardian of such locus of agency is the auditory digital twin that fosters intra-actions between humans and technology, dynamically and fluidly redefining all those configurations that are crucial for an immersive and coherent experience. The idea of entanglement theory is here mainly declined in an egocentric-spatial perspective related to emerging knowledge of the listener's perceptual capabilities. This is an actively transformative relation with the digita

This paper proposes a conceptual model for a secure and performance-efficient workload management model in cloud environments. In this model, a resource management unit is employed for energy and performance proficient allocation of virtual machines while ensuring the secure processing of users' applications by defending against data breaches due to unauthorized access to virtual machines in real-time. The resource management unit is guided by a secure virtual machine management unit which is designed to generate information regarding unauthorized access or inter-communication links among active virtual machines. Also, a workload analyzer unit operates concurrently to estimate resource utilization information to assist the resource management unit in the performance-efficient allocation of virtual machines. Contrary to prior works which engage access control mechanisms, encryption, and decryption of data before the transfer and the use of tunneling for prevention of unauthorized access to virtual machines which raises excess computational cost overhead, the proposed model operates diversely for efficiently serving the same purpose.