搜索结果：harnessing

Agent skills externalise reusable agent-facing behavioural knowledge and guidance as persistent artefacts that can be discovered, activated, and interpreted by LLM agents. Although a skill artefact is static at rest, its architectural responsibilities arise in use, when the artefact is selected for a run, bound to context and authority constraints, interpreted by a stochastic agent, and recorded as run evidence. We call this run-specific relation skill-in-use. This paper studies agent skill harnessing: the architectural responsibilities that govern the transition from skill artefacts to skill-in-use, bound the executable consequences associated with skill-in-use, and capture evidence for attribution, verification, repair, and evolution. This paper provides a catalogue of ten empirically grounded architectural patterns (five core, five supporting) for skill harnessing and synthesises them into a reference architecture with four responsibility layers: Supply Chain, Mediation, Execution Control, and Evidence & Feedback. We evaluate the architecture through cross-instantiation across 8 selected systems. The resulting patterns and reference architecture provide a vocabulary and diag

Language models have changed from unreliable text generators to highly-capable large models with trillions of parameters. Capability increases come hand-in-hand with increases in scale, making understanding the internal representations of models more challenging. Since millions of users increasing rely on language models to interact with external tools or make decisions in medium or high-stakes scenarios, we need to establish control over model behavior and know when to trust model outputs. In this paper, we discuss our contributions on harnessing the latent spaces by proposing steering vectors for control and developing latent space-based model calibrators for trust. Together, our contributions help demystify the latent spaces of language models and offer new insights into how to harness model internals to build more trustworthy language technology.

Understanding what a user believes and intends is central to building effective agent assistants. This ability is often evaluated through Theory-of-Mind (ToM) tasks, where success requires reasoning from the user's perspective. However, many existing approaches address ToM with complex pipelines that model behavior indirectly, without explicitly reconstructing the user's mental state. This misses the core structure of the problem: users act based on their beliefs, which are updated through observations of the environment; beliefs and intentions jointly determine actions, which in turn change the environment; and social reasoning often requires nested beliefs about what others believe or intend. We propose UserHarness, a simple framework that reframes ToM reasoning as explicit user-mind reconstruction. UserHarness decomposes the user's mental state, its relation to the external environment, and the actions that follow from it, enabling agents to track what the user observes, believes, intends, and does. Across five benchmarks, UserHarness reaches up to 95.94% macro accuracy, improving over existing inference methods by more than 15% relative and over the strongest prompt-only harnes

Wire-harnessing tasks pose great challenges to be automated by the robot due to the complex dynamics and unpredictable behavior of the deformable wire. Traditional methods, often reliant on dual-robot arms or tactile sensing, face limitations in adaptability, cost, and scalability. This paper introduces a novel single-robot wire-harnessing pipeline that leverages a robot's twisting motion to generate necessary wire tension for precise insertion into clamps, using only one robot arm with an integrated force/torque (F/T) sensor. Benefiting from this design, the single robot arm can efficiently apply tension for wire routing and insertion into clamps in a narrow space. Our approach is structured around four principal components: a Model Predictive Control (MPC) based on the Koopman operator for tension tracking and wire following, a motion planner for sequencing harnessing waypoints, a suite of insertion primitives for clamp engagement, and a fix-point switching mechanism for wire constraint updating. Evaluated on an industrial-level wire harnessing task, our method demonstrated superior performance and reliability over conventional approaches, efficiently handling both single and mul

Most of today's educators are in no shortage of digital and online learning technologies available at their fingertips, ranging from Learning Management Systems such as Canvas, Blackboard, or Moodle, online meeting tools, online homework, and tutoring systems, exam proctoring platforms, computer simulations, and even virtual reality/augmented reality technologies. Furthermore, with the rapid development and wide availability of generative artificial intelligence (GenAI) services such as ChatGPT, we are just at the beginning of harnessing their potential to transform higher education. Yet, facing the large number of available options provided by cutting-edge technology, an imminent question on the mind of most educators is the following: how should I choose the technologies and integrate them into my teaching process so that they would best support student learning? We contemplate over these types of important and timely questions and share our reflections on evidence-based approaches to harnessing digital learning tools using a Self-regulated Engaged Learning Framework we have employed in our research in physics education that can be valuable for educators in other disciplines.

Many high-stakes decisions depend on forecasts made before outcomes are known. In this future prediction setting, the central challenge is that public evidence evolves over time, while the main supervision signal arrives only after resolution: the realized outcome mainly assesses final correctness, offering only coarse guidance on what to track, what to verify, and which judgments to leave uncertain along the way. Our key observation is that revisiting the same unresolved question over time creates informative temporal contrasts across evolving evidence and repeated forecasts, exposing what earlier attempts missed before resolution and yielding a diagnostic signal we call the pre-resolution signal. We instantiate this idea in Milkyway, a future prediction agent with a persistent future prediction harness, an editable external state that stores reusable procedural guidance across revisits to the same unresolved question. As the same unresolved question is revisited, Milkyway extracts pre-resolution signals from evolving evidence and repeated forecasts, uses them to update the harness, and improves later forecasts on that question before resolution. After resolution, the realized out

Time series are often embedded in rich contexts that are essential for holistic modeling. Moreover, real-world practitioners often require end-to-end workflows for analyzing temporal dynamics, where widely studied tasks such as forecasting are only one step in a broader solution loop. While generalist AI agents offer a promising interface for such workflows under complex contexts, they still operate primarily in textual spaces that are not fully aligned with structured temporal signals. In this work, we introduce TimeClaw, an agentic harness framework for time series that equips generalist LLM agents with the time series-native runtime support needed for contextualized temporal reasoning. TimeClaw integrates executable temporal tools for grounded and auditable analysis, experience-driven capability evolution for creating reusable analytical routines, and episodic multimodal memory for retrieving relevant reasoning traces. Together, these components unlock harnessed open-ended temporal reasoning with contextual information. Extensive evaluation on multiple benchmarks covering diverse tasks across energy, finance, weather, traffic, and other real-world domains demonstrates improved p

The intrinsic integration of Rydberg atomic receivers into wireless communication systems is proposed, by harnessing the principles of quantum physics in wireless communications. More particularly, we conceive a pair of Rydberg atomic receivers, one incorporates a local oscillator (LO), referred to as an LO-dressed receiver, while the other operates without an LO and is termed an LO-free receiver. The appropriate wireless model is developed for each configuration, elaborating on the receiver's responses to the radio frequency (RF) signal, on the potential noise sources, and on the signal-to-noise ratio (SNR) performance. The developed wireless model conforms to the classical RF framework, facilitating compatibility with established signal processing methodologies. Next, we investigate the associated distortion effects that might occur, specifically identifying the conditions under which distortion arises and demonstrating the boundaries of linear dynamic ranges. This provides critical insights into its practical implementations in wireless systems. Finally, extensive simulation results are provided for characterizing the performance of wireless systems, harnessing this pair of Rydb

Temperature modulations provide an alternative method for dynamically controlling the ferroelectric state. In this paper, we use scale-independent Landau potentials and predictive atomistic simulations to explore how temperature modulation can harness the depolarizing field to obtain non-electrical poling. We further predict that temperature gradients can be combined with strain to induce persistent polar textures such as multidomain states.

We present Logic Workout, a novel training method based on radical dynamic instability combining rolling, deformation and spring instability on small fitballs. By exploiting the "reactive falling effect", it re-engages innate balance mechanisms and forces real-time motor corrections, leading to better joint control, coordination, and movement precision. Preliminary results from a cohort of 18 participants show complete resolution of chronic pain, enhanced functional mobility, and surprising improvements in strength and performance -- challenging the belief that instability training impairs power output. We hypothesise that, by harnessing maximal instability, Logic Workout activates deep neuromuscular pathways and improve rehabilitation outcomes, training efficiency, and athletic performance.

This paper introduces LIMITER, a gamified digital musical instrument for harnessing and performing microtonal and justly intonated sounds. While microtonality in Western music remains a niche and esoteric system that can be difficult both to conceptualize and to perform with, LIMITER presents a novel, easy to pickup interface that utilizes color, geometric transformations, and game-like controls to create a simpler inlet into utilizing these sounds as a means of expression. We report on the background of the development of LIMITER, as well as explain the underlying musical and engineering systems that enable its function. Additionally, we offer a discussion and preliminary evaluation of the creativity-enhancing effects of the interface.

Although speckle is a powerful tool for high-precision metrology, large datasets and cumbersome training are always required to learn from the encoded speckle patterns, which is unfavorable for rapid deployment and multi-dimensional metrology. To enable high accuracy and fast training, physics-informed machine learning enforces physical laws to address high-dimensional problems. Here, we harness the modal fields in a few-mode fiber, which follow the law of beam propagation, to enable high-accuracy and fast-training parameter estimation. Anti-noise fast mode decomposition is implemented to retrieve the modal fields from the speckles. The accuracy is enhanced since the modal fields enable parameter estimation at random points in the continuous space-time domain. Artificial tactile perception and multi-dimensional metrology are achieved with high accuracy because the modal fields respond diversely to different parameters. Meanwhile, the number of specklegrams for training is reduced by around 5 times. The training time of machine learning is significantly reduced by 800 times, from 9 hours and 45 minutes to 40 seconds. Therefore, harnessing the modal fields paves a new way for the spe

Polar topology, an analogue of the magnetic topology, serves as a large playground for exotic physical phenomena with a wide range of multifunctional applications. Polar vortices and skyrmions are representative polar topologies that have been predicted to significantly enhance the functionality and information density of nanoelectronic devices due to their ultrasmall dimensions. Despite these advantages, the practical realization of polar topologies in devices is impeded by the intrinsic challenges associated with their controlled motion and manipulation. Therefore, harnessing vortex manipulation-such as motion, on demand creation, annihilation, and shape transformation-is essential for practical device integration. However, vortex motion is often challenged by intrinsic physical limitations in collective lattice distortions and strong pinning effects from the surrounding environment, which remains elusive. In this study, we present real time observation of vortex motion in PbTiO3/SrTiO3 heterostructures, achieved through the application of localized pulsed electric fields and trailing bias fields from a conductive tip. Notably, the vortices exhibit reversible motion in response t

We study the optimal performance of an information engine consisting of an overdamped Brownian bead confined in a controllable, $d$-dimensional harmonic trap and additionally subjected to gravity. The trap's center is updated dynamically via a feedback protocol designed such that no external work is done by the trap on the bead, while maximizing the extraction of gravitational potential energy and achieving directed motion. We show that performance strikingly improves when thermal fluctuations in directions perpendicular to gravity are harnessed. This improvement arises from feedback cooling of these transverse degrees of freedom, along which all heat is extracted; comparable performance can be achieved even without vertical measurements. This engine design modularizes the functions of harnessing fluctuations and storing free energy, drawing a close analogy to the Szilard engine.

Transcranial Electrical Stimulation (tES) is a neuromodulation technique that utilizes electrodes on the scalp to stimulate target brain regions. tES has shown promise in treating many neurological conditions, such as stroke rehabilitation and chronic pain. Several electrode placement algorithms have been proposed to optimize tES-based therapies by designing multi-electrode montages that create focal neural responses. We first extend a well-known unification result by Fernandez-Corazza et al. to unify all major traditional electrode placement algorithms. We utilize this unification result to identify a common restriction among traditional electrode placement algorithms: they do not harness the thresholding behavior of neural response. Consequently, these algorithms only partially harness the properties of neural response to optimize tES, particularly increasing the focality of neural response. We propose a new electrode placement algorithm, HingePlace, that utilizes a symmetrized hinge loss to harness the thresholding behavior of neural response. We extensively compare the HingePlace algorithm with traditional electrode placement algorithms in two simulation platforms. Across both

High-dimensional quantum entanglement is an essential resource in quantum technology since it provides benefits in increasing the information capacity and processing speed. Thus, the controlled harnessing of high-dimensional entanglement has long been hailed as a necessary prerequisite towards practical quantum applications. By using a deterministic quantum state filter that implemented through quantum interference, we present a generalised formulation for the complete high-dimensional symmetric and anti-symmetric Bell basis, and experimentally prepare four-dimensional orbital angular momentum Bell states that provide the well-behaved symmetric or anti-symmetric properties. Additionally, we use a concise yet efficient scan of temporal delay to directly observe high-dimensional two-photon interference effects in spatial modes. These results provide an alternative way for harnessing high-dimensional entanglement, and may facilitate the use of quantum interference for more complex quantum information processing tasks that beyond qubits.

Short-range surface plasmon polaritons (SR-SPPs) can arise due to the hybridization of surface plasmon polaritons propagating along the two interfaces of a thin metal slab. In optics, they have gained particular interest for imaging and sensing applications, because of their short wavelengths at optical frequencies along with strong field enhancement. However, mediating the interaction of SR-SPPs with photons in planar films is difficult because of the large momentum mismatch. For efficient coupling, nanostructuring such thin films (~20nm thickness), or placing metallic nanostructures in close proximity to the planar film is technologically challenging and can strongly influence the SR-SPP properties. In this paper, harnessing SR-SPPs in planar silver films is demonstrated using disorder-engineered metasurfaces. The disorder-engineering is realized by the light-controlled growth of silver nanoparticles. The dispersion of the hybrid modes with the silver thickness is measured and compared with simulations. We anticipate these results to introduce a novel and facile method for harnessing SR-SPPs in planar optical systems and make use of their promising properties for imaging, sensing

Despite the widespread integration of ambient light sensors (ALS) in smart devices commonly used for screen brightness adaptation, their application in human activity recognition (HAR), primarily through body-worn ALS, is largely unexplored. In this work, we developed ALS-HAR, a robust wearable light-based motion activity classifier. Although ALS-HAR achieves comparable accuracy to other modalities, its natural sensitivity to external disturbances, such as changes in ambient light, weather conditions, or indoor lighting, makes it challenging for daily use. To address such drawbacks, we introduce strategies to enhance environment-invariant IMU-based activity classifications through augmented multi-modal and contrastive classifications by transferring the knowledge extracted from the ALS. Our experiments on a real-world activity dataset for three different scenarios demonstrate that while ALS-HAR's accuracy strongly relies on external lighting conditions, cross-modal information can still improve other HAR systems, such as IMU-based classifiers.Even in scenarios where ALS performs insufficiently, the additional knowledge enables improved accuracy and macro F1 score by up to 4.2 % and

With the rapid development of LLMs, it is natural to ask how to harness their capabilities efficiently. In this paper, we explore whether it is feasible to direct each input query to a single most suitable LLM. To this end, we propose LLM routing for challenging reasoning tasks. Our extensive experiments suggest that such routing shows promise but is not feasible in all scenarios, so more robust approaches should be investigated to fill this gap.