Search

Homophily and heterophobia, the tendency for people with similar characteristics to preferentially interact with (or avoid) each other are pervasive in human social networks. Here, we develop an extension of the mathematical theory of urban scaling which describes the effects of homophily and heterophobia on social interactions and resulting economic outputs of cities. Empirical tests of our model show that increased residential racial heterophobia and segregation in U.S. cities are associated with reduced economic outputs and that the strength of this relationship increased throughout the 2010s. Our findings provide the means for the formal incorporation of general homophilic and heterophobic effects into theories of modern urban science and suggest that racial segregation is increasingly and adversely impacting the economic performance and connectivity of urban societies in the U.S.

We describe the current content moderation strategy employed by Meta to remove policy-violating content from its platforms. Meta relies on both handcrafted and learned risk models to flag potentially violating content for human review. Our approach aggregates these risk models into a single ranking score, calibrating them to prioritize more reliable risk models. A key challenge is that violation trends change over time, affecting which risk models are most reliable. Our system additionally handles production challenges such as changing risk models and novel risk models. We use a contextual bandit to update the calibration in response to such trends. Our approach increases Meta's top-line metric for measuring the effectiveness of its content moderation strategy by 13%.

Optoelectronic properties of van der Waals homostructures can be selectively engineered by the relative twist angle between layers. Here, we study the twist-dependent moire coupling in MoSe2 homobilayers. For small angles, we find a pronounced redshift of the K-K and {\Gamma}-K excitons accompanied by a transition from K-K to {\Gamma}-K emission. Both effects can be traced back to the underlying moire pattern in the MoSe2 homobilayers, as confirmed by our low-energy continuum model for different moire excitons. We identify two distinct intralayer moire excitons for R-stacking, while H-stacking yields two degenerate intralayer excitons due to inversion symmetry. In both cases, bright interlayer excitons are found at higher energies. The performed calculations are in excellent agreement with experiment and allow us to characterize the observed exciton resonances, providing insight about the layer composition and relevant stacking configuration of different moire exciton species.

Integrating quantum emitters into nanocavities which simultaneously couple to the photonic and mechanical modes is critical for interfacing electron spins, photons and phonons in the cavity QED system. Here, we investigate the interaction between the charged boron vacancy $V_B^-$, ultra-high-Q ($\sim10^5$) cavity photonic modes and local phonon modes. A pronounced asymmetry is observed in the emission spectrum for cavities with Q-factor above a threshold of 10$^4$. Similar asymmetries are not observed for cavities without $V_B^-$ centers. To explain our findings, we model the system with phonon-induced light-matter coupling based on $V_B^-$ centers, and compare to the Jaynes-Cummings model for usual emitters. Our results reveal that the multipartite interplay arises during the light-matter coupling of $V_B^-$ centers, illustrating that it is phonon-induced, rather than being caused by thermal population of phonon modes. Such emitter-optomechanical interaction between different photon ($V_B^-$ emission, cavity photonic) and phonon ($V_B^-$ phonon, cavity mechanical) modes provides a novel system to interface spin defects, photons and phonons in condensed matters.

Magnetic resonance imaging of spin systems combines scientific applications in medicine, chemistry and physics. Here, we investigate the pixel-wise coherent quantum dynamics of spins consisting of a 40 by 40 micron sized region of interest implanted with nitrogen vacancy centers (NV) coupled to a nano-magnetic flake of $\mathrm{CrTe_2}$. $\mathrm{CrTe_2}$ is an in-plane van der Waals ferromagnet, which we can probe quantitatively by the NV electron's spin signal even at room temperature. First, we combine the nano-scale sample shapes measured by atomic force microscope with the magnetic resonance imaging data. We then map out the coherent dynamics of the colour centers coupled to the van der Waals ferromagnet using pixel-wise coherent Rabi and Ramsey imaging of the NV sensor layer. Next, we fit the pixel-wise solution of the Hamiltonian to the quantum sensor data. Combining data and model, we can explore the detuning range of the spin oscillation with a quantum dynamic range of over $\left|\Delta_{max}\right|= 60 { }\mathrm{MHz} $ in the Ramsey interferometry mode. Finally, we show the effect of the $\mathrm{CrTe_2}$ van der Waals magnet on the coherence of the NV sensor layer and measure a 70 times increase in the maximum frequency of the quantum oscillation going from the Rabi to the Ramsey imaging mode.

Large scale two-dimensional (2D) moir\'e superlattices are driving a revolution in designer quantum materials. The electronic interactions in these superlattices, strongly dependent on the periodicity and symmetry of the moir\'e pattern, critically determine the emergent properties and phase diagrams. To date, the relative twist angle between two layers has been the primary tuning parameter for a given choice of constituent crystals. Here, we establish strain as a powerful mechanism to in-situ modify the moir\'e periodicity and symmetry. We develop an analytically exact mathematical description for the moir\'e lattice under arbitrary in-plane heterostrain acting on any bilayer structure. We demonstrate the ability to fine-tune the moir\'e lattice near critical points, such as the magic angle in bilayer graphene, or fully reconfigure the moir\'e lattice symmetry beyond that imposed by the unstrained constituent crystals. Due to this unprecedented simultaneous control over the strength of electronic interactions and lattice symmetry, 2D heterostrain provides a powerful platform to engineer, tune, and probe strongly correlated moir\'e materials., Comment: 9 pages,6 figures

Aerosol particles play an important role in the climate system by absorbing and scattering radiation and influencing cloud properties. They are also one of the biggest sources of uncertainty for climate modeling. Many climate models do not include aerosols in sufficient detail due to computational constraints. In order to represent key processes, aerosol microphysical properties and processes have to be accounted for. This is done in the ECHAM-HAM global climate aerosol model using the M7 microphysics, but high computational costs make it very expensive to run with finer resolution or for a longer time. We aim to use machine learning to emulate the microphysics model at sufficient accuracy and reduce the computational cost by being fast at inference time. The original M7 model is used to generate data of input-output pairs to train a neural network on it. We are able to learn the variables' tendencies achieving an average $R^2$ score of $77.1\% $. We further explore methods to inform and constrain the neural network with physical knowledge to reduce mass violation and enforce mass positivity. On a GPU we achieve a speed-up of up to over 64x compared to the original model., Comment: arXiv admin note: text overlap with arXiv:2109.10593

It is well known that Pennsylvania's K-12 schools are inadequately and inequitably funded. But the extent of the problem is not fully understood. This paper uses new data and methods to demonstrate just how unfair--and in fact, morally unsustainable--the funding of elementary and secondary education is in the Commonwealth. It shows that the proposal put forward in Governor Wolf's 2021-2022 Executive Budget, or something much like it, is a necessary first step toward reforming the shameful way K-12 schools in Pennsylvania are funded. The origins of the deep inadequacy and inequity in school funding go back decades. For the purposes of this paper, the source of the problem is divided into two parts. First, the state share of funding of K-12 schools has been declining since the early 1970s under Democratic and Republican governors--but it is mostly a result of Republican control of the General Assembly. Second, measuring the adequacy of funding by looking at the gap in per-student spending in school districts between what they spent in 2019 and what they should spend according to a 2020 update to the "Costing-Out Study" carried out by Augenblick, Palaich and Associates (APA) at the request of the State Board of Education pursuant to the bipartisan Act 114 of July 2005. Governor Wolf's proposal for K-12 education funding for the fiscal year beginning July 1, 2021, is a major step forward in providing adequate and equitable funding for Pennsylvania's schools. However, it does not close the funding gap for most schools. Nor does it eliminate all of the economic, racial, and ethnicity-based inequities in school funding.

Single spin defects in 2D transition-metal dichalcogenides are natural spin-photon interfaces for quantum applications. Here we report high-field magneto-photoluminescence spectroscopy from three emission lines (Q1, Q2 and Q*) of He-ion induced sulfur vacancies in monolayer $\text{MoS}_2$. Analysis of the asymmetric PL lineshapes in combination with the diamagnetic shift of Q1 and Q2 yields a consistent picture of localized emitters with a wavefunction extent of $\sim$ 3.5 nm. The distinct valley-Zeeman splitting in out-of-plane $B$-fields and the brightening of dark states through in-plane $B$-fields necessitates spin-valley selectivity of the defect states and lifted spin-degeneracy at zero field. Comparing our results to ab-initio calculations identifies the nature of Q1 and Q2 and suggests that Q* is the emission from a chemically functionalized defect. Analysis of the optical degree of circular polarization reveals that the Fermi level is a parameter that enables the tunability of the emitter. These results show that defects in 2D semiconductors may be utilized for quantum technologies.

An algorithm of Ross and Selinger for the factorization of diagonal elements of PU(2) to within distance $\varepsilon$ was adapted by Parzanchevski and Sarnak into an efficient probabilistic algorithm for any element of PU(2) using at most effective $3\log_p\frac{1}{\varepsilon^{3}}$ factors from certain well-chosen sets associated to a number field and a prime $p$. The icosahedral super golden gates are one such set associated to $\mathbb{Q}(\sqrt{5})$. We leverage recent work of Carvalho Pinto, Petit, and Stier to reduce this bound to $\frac{7}{3}\log_{59}\frac{1}{\varepsilon^3}$, and we implement the algorithm in Python. This represents an improvement by a multiplicative factor of $\log_259\approx5.9$ over the analogous result for the Clifford+$T$ gates. This is of interest because the icosahedral gates have shortest factorization lengths among all super golden gates.

We introduce an interactive image segmentation and visualization framework for identifying, inspecting, and editing tiny objects (just a few pixels wide) in large multi-megapixel high-dynamic-range (HDR) images. Detecting cosmic rays (CRs) in astronomical observations is a cumbersome workflow that requires multiple tools, so we developed an interactive toolkit that unifies model inference, HDR image visualization, segmentation mask inspection and editing into a single graphical user interface. The feature set, initially designed for astronomical data, makes this work a useful research-supporting tool for human-in-the-loop tiny-object segmentation in scientific areas like biomedicine, materials science, remote sensing, etc., as well as computer vision. Our interface features mouse-controlled, synchronized, dual-window visualization of the image and the segmentation mask, a critical feature for locating tiny objects in multi-megapixel images. The browser-based tool can be readily hosted on the web to provide multi-user access and GPU acceleration for any device. The toolkit can also be used as a high-precision annotation tool, or adapted as the frontend for an interactive machine learning framework. Our open-source dataset, CR detection model, and visualization toolkit are available at https://github.com/cy-xu/cosmic-conn., Comment: 6 pages, 4 figures. Accepted by CVPR 2022 Demo Program

Estimating the effects of continuous-valued interventions from observational data is a critically important task for climate science, healthcare, and economics. Recent work focuses on designing neural network architectures and regularization functions to allow for scalable estimation of average and individual-level dose-response curves from high-dimensional, large-sample data. Such methodologies assume ignorability (observation of all confounding variables) and positivity (observation of all treatment levels for every covariate value describing a set of units), assumptions problematic in the continuous treatment regime. Scalable sensitivity and uncertainty analyses to understand the ignorance induced in causal estimates when these assumptions are relaxed are less studied. Here, we develop a continuous treatment-effect marginal sensitivity model (CMSM) and derive bounds that agree with the observed data and a researcher-defined level of hidden confounding. We introduce a scalable algorithm and uncertainty-aware deep models to derive and estimate these bounds for high-dimensional, large-sample observational data. We work in concert with climate scientists interested in the climatological impacts of human emissions on cloud properties using satellite observations from the past 15 years. This problem is known to be complicated by many unobserved confounders., Comment: 33 pages

We report resonant Raman spectroscopy of neutral excitons X$^0$ and intravalley trions X$^-$ in hBN-encapsulated MoS$_2$ monolayer embedded in a nanobeam cavity. By temperature tuning the detuning between Raman modes of MoS$_2$ lattice phonons and X$^0$/X$^-$ emission peaks, we probe the mutual coupling of excitons, lattice phonons and cavity vibrational phonons. We observe an enhancement of X$^0$-induced Raman scattering and a suppression for X$^-$-induced, and explain our findings as arising from the tripartite exciton-phonon-phonon coupling. The cavity vibrational phonons provide intermediate replica states of X$^0$ for resonance conditions in the scattering of lattice phonons, thus enhancing the Raman intensity. In contrast, the tripartite coupling involving X$^-$ is found to be much weaker, an observation explained by the geometry-dependent polarity of the electron and hole deformation potentials. Our results indicate that phononic hybridization between lattice and nanomechanical modes plays a key role in the excitonic photophysics and light-matter interaction in 2D-material nanophotonic systems.

Negatively charged boron vacancies ($V_B^-$) in hexagonal boron nitride (hBN) exhibit a broad emission spectrum due to strong electron-phonon coupling and Jahn-Teller mixing of electronic states. As such, the direct measurement of zero-phonon line (ZPL) of $V_B^-$ has remained elusive. Here, we measure the room-temperature ZPL wavelength to be $773\pm2$ nm by coupling the hBN layer to the high-Q nanobeam cavity. As the wavelength of cavity mode is tuned, we observe a pronounced intensity resonance, indicating the coupling to $V_B^-$. Our observations are consistent with the spatial redistribution of $V_B^-$ emission. Spatially resolved measurements show a clear Purcell effect maximum at the midpoint of the nanobeam, in accord with the optical field distribution of the cavity mode. Our results are in good agreement with theoretical calculations, opening the way to using $V_B^-$ as cavity spin-photon interfaces.

In today's digital world, interaction with online platforms is ubiquitous, and thus content moderation is important for protecting users from content that do not comply with pre-established community guidelines. Having a robust content moderation system throughout every stage of planning is particularly important. We study the short-term planning problem of allocating human content reviewers to different harmful content categories. We use tools from fair division and study the application of competitive equilibrium and leximin allocation rules. Furthermore, we incorporate, to the traditional Fisher market setup, novel aspects that are of practical importance. The first aspect is the forecasted workload of different content categories. We show how a formulation that is inspired by the celebrated Eisenberg-Gale program allows us to find an allocation that not only satisfies the forecasted workload, but also fairly allocates the remaining reviewing hours among all content categories. The resulting allocation is also robust as the additional allocation provides a guardrail in cases where the actual workload deviates from the predicted workload. The second practical consideration is time dependent allocation that is motivated by the fact that partners need scheduling guidance for the reviewers across days to achieve efficiency. To address the time component, we introduce new extensions of the various fair allocation approaches for the single-time period setting, and we show that many properties extend in essence, albeit with some modifications. Related to the time component, we additionally investigate how to satisfy markets' desire for smooth allocation (e.g., partners for content reviewers prefer an allocation that does not vary much from time to time, to minimize staffing switch). We demonstrate the performance of our proposed approaches through real-world data obtained from Meta.

Recent volumetric 3D reconstruction methods can produce very accurate results, with plausible geometry even for unobserved surfaces. However, they face an undesirable trade-off when it comes to multi-view fusion. They can fuse all available view information by global averaging, thus losing fine detail, or they can heuristically cluster views for local fusion, thus restricting their ability to consider all views jointly. Our key insight is that greater detail can be retained without restricting view diversity by learning a view-fusion function conditioned on camera pose and image content. We propose to learn this multi-view fusion using a transformer. To this end, we introduce VoRTX, an end-to-end volumetric 3D reconstruction network using transformers for wide-baseline, multi-view feature fusion. Our model is occlusion-aware, leveraging the transformer architecture to predict an initial, projective scene geometry estimate. This estimate is used to avoid backprojecting image features through surfaces into occluded regions. We train our model on ScanNet and show that it produces better reconstructions than state-of-the-art methods. We also demonstrate generalization without any fine-tuning, outperforming the same state-of-the-art methods on two other datasets, TUM-RGBD and ICL-NUIM., Comment: 3DV 2021

We present 3DVNet, a novel multi-view stereo (MVS) depth-prediction method that combines the advantages of previous depth-based and volumetric MVS approaches. Our key idea is the use of a 3D scene-modeling network that iteratively updates a set of coarse depth predictions, resulting in highly accurate predictions which agree on the underlying scene geometry. Unlike existing depth-prediction techniques, our method uses a volumetric 3D convolutional neural network (CNN) that operates in world space on all depth maps jointly. The network can therefore learn meaningful scene-level priors. Furthermore, unlike existing volumetric MVS techniques, our 3D CNN operates on a feature-augmented point cloud, allowing for effective aggregation of multi-view information and flexible iterative refinement of depth maps. Experimental results show our method exceeds state-of-the-art accuracy in both depth prediction and 3D reconstruction metrics on the ScanNet dataset, as well as a selection of scenes from the TUM-RGBD and ICL-NUIM datasets. This shows that our method is both effective and generalizes to new settings., Comment: 10 pages, 6 figures, 3 tables. Accepted to 3DV 2021

Multimodal classification is a core task in human-centric machine learning. We observe that information is highly complementary across modalities, thus unimodal information can be drastically sparsified prior to multimodal fusion without loss of accuracy. To this end, we present Sparse Fusion Transformers (SFT), a novel multimodal fusion method for transformers that performs comparably to existing state-of-the-art methods while having greatly reduced memory footprint and computation cost. Key to our idea is a sparse-pooling block that reduces unimodal token sets prior to cross-modality modeling. Evaluations are conducted on multiple multimodal benchmark datasets for a wide range of classification tasks. State-of-the-art performance is obtained on multiple benchmarks under similar experiment conditions, while reporting up to six-fold reduction in computational cost and memory requirements. Extensive ablation studies showcase our benefits of combining sparsification and multimodal learning over naive approaches. This paves the way for enabling multimodal learning on low-resource devices., Comment: 11 pages, 4 figures, 5 tables, Yi Ding and Alex Rich contributed equally

deepstruct connects deep learning models and graph theory such that different graph structures can be imposed on neural networks or graph structures can be extracted from trained neural network models. For this, deepstruct provides deep neural network models with different restrictions which can be created based on an initial graph. Further, tools to extract graph structures from trained models are available. This step of extracting graphs can be computationally expensive even for models of just a few dozen thousand parameters and poses a challenging problem. deepstruct supports research in pruning, neural architecture search, automated network design and structure analysis of neural networks.

Magnetic skyrmions and their anti-particles, the antiskyrmions, are stable magnetic solitons existing down to the nanometer scale. Their stability and size as well as the possibility to propel them by, e.g., electric currents make them promising candidates for the use in memory devices, such as racetrack memories. Skyrmions and antiskyrmions share those same advantages individually, but may annihilate each other when they coexist in the same device. Yet, combining them to represent logical bits of 0 (skyrmion) and 1 (antiskyrmion) in one device opens new possibilities to create new types of densely packed racetrack memory devices. For this, a controlled creation and annihilation procedure, i.e., a writing or deletion operation, is necessary. Here, we propose a method to create arbitrary sequences of coexisting skyrmions and antiskyrmions by rotations of the magnetic moments at the edge of a rectangular slab. The skyrmions and antiskyrmions remain stable and do not annihilate each other.

Aerosol-cloud interactions include a myriad of effects that all begin when aerosol enters a cloud and acts as cloud condensation nuclei (CCN). An increase in CCN results in a decrease in the mean cloud droplet size (r$_{e}$). The smaller droplet size leads to brighter, more expansive, and longer lasting clouds that reflect more incoming sunlight, thus cooling the earth. Globally, aerosol-cloud interactions cool the Earth, however the strength of the effect is heterogeneous over different meteorological regimes. Understanding how aerosol-cloud interactions evolve as a function of the local environment can help us better understand sources of error in our Earth system models, which currently fail to reproduce the observed relationships. In this work we use recent non-linear, causal machine learning methods to study the heterogeneous effects of aerosols on cloud droplet radius.

We present a novel transparent conductive support structure for two-dimensional (2D) materials that provides an electronically passive 2D/3D interface while also enabling facile interfacial charge transport. This structure, which comprises an evaporated nanocrystalline carbon (nc-C) film beneath an atomic layer deposited alumina (ALD AlOx) layer, is thermally stable and allows direct chemical vapor deposition (CVD) of 2D materials onto the surface. When the nc-C/AlOx is deposited onto a 270 nm SiO2 layer on Si, strong optical contrast for monolayer flakes is retained. Raman spectroscopy reveals good crystal quality for MoS2 and we observe a ten-fold photoluminescence intensity enhancement compared to flakes on conventional Si/SiO2. Tunneling across the ultrathin AlOx enables interfacial charge injection, which we demonstrate by artifact-free scanning electron microscopy and photoemission electron microscopy. Thus, this combination of scalable fabrication and electronic conductivity across a weakly interacting 2D/3D interface opens up new application and characterization opportunities for 2D materials., Comment: 28 pages, 9 figures

We study the creation of quantum skyrmions in quadratic nanoscopic lattices of quantum spins coupled by Dzyaloshinkii-Moriya and exchange interactions. We numerically show that different kinds of quantum skyrmions, characterized by the magnitude of their spin expectation values and strong differences in their stability, can appear as ground state and as metastable excitations. In dependence on the coupling strengths and the lattice size, an adiabatic rotation scheme of the magnetization at the lattice boundaries allows for a controlled creation of quantum skyrmions.

Aerosol particles play an important role in the climate system by absorbing and scattering radiation and influencing cloud properties. They are also one of the biggest sources of uncertainty for climate modeling. Many climate models do not include aerosols in sufficient detail. In order to achieve higher accuracy, aerosol microphysical properties and processes have to be accounted for. This is done in the ECHAM-HAM global climate aerosol model using the M7 microphysics model, but increased computational costs make it very expensive to run at higher resolutions or for a longer time. We aim to use machine learning to approximate the microphysics model at sufficient accuracy and reduce the computational cost by being fast at inference time. The original M7 model is used to generate data of input-output pairs to train a neural network on it. By using a special logarithmic transform we are able to learn the variables tendencies achieving an average $R^2$ score of $89\%$. On a GPU we achieve a speed-up of 120 compared to the original model.

We report magneto-photoluminescence spectroscopy of gated MoS$_2$ monolayers in high magnetic fields to 28 T. At B = 0T and electron density $n_s\sim 10^{12}cm^-2$, we observe three trion resonances that cannot be explained within a single-particle picture. Employing ab initio calculations that take into account three-particle correlation effects as well as local and non-local electron-hole exchange interaction, we identify those features as quantum superpositions of inter- and intravalley spin states. We experimentally investigate the mixed character of the trion wave function via the filling factor dependent valley Zeeman shift in positive and negative magnetic fields. Our results highlight the importance of exchange interactions for exciton physics in monolayer MoS$_2$ and provide new insights into the microscopic understanding of trion physics in 2D multi-valley semiconductors for low excess carrier densities., Comment: Main manuscript: 7 pages, 3 figures ; Supplemental material: 16 pages, 4 figures

Large computer models are ubiquitous in the earth sciences. These models often have tens or hundreds of tuneable parameters and can take thousands of core-hours to run to completion while generating terabytes of output. It is becoming common practice to develop emulators as fast approximations, or surrogates, of these models in order to explore the relationships between these inputs and outputs, understand uncertainties and generate large ensembles datasets. While the purpose of these surrogates may differ, their development is often very similar. Here we introduce ESEm: an open-source tool providing a general workflow for emulating and validating a wide variety of models and outputs. It includes efficient routines for sampling these emulators for the purpose of uncertainty quantification and model calibration. It is built on well-established, high-performance libraries to ensure robustness, extensibility and scalability. We demonstrate the flexibility of ESEm through three case-studies using ESEm to reduce parametric uncertainty in a general circulation model, explore precipitation sensitivity in a cloud resolving model and scenario uncertainty in the CMIP6 multi-model ensemble.

Sparsity in the structure of Neural Networks can lead to less energy consumption, less memory usage, faster computation times on convenient hardware, and automated machine learning. If sparsity gives rise to certain kinds of structure, it can explain automatically obtained features during learning. We provide insights into experiments in which we show how sparsity can be achieved through prior initialization, pruning, and during learning, and answer questions on the relationship between the structure of Neural Networks and their performance. This includes the first work of inducing priors from network theory into Recurrent Neural Networks and an architectural performance prediction during a Neural Architecture Search. Within our experiments, we show how magnitude class blinded pruning achieves 97.5% on MNIST with 80% compression and re-training, which is 0.5 points more than without compression, that magnitude class uniform pruning is significantly inferior to it and how a genetic search enhanced with performance prediction achieves 82.4% on CIFAR10. Further, performance prediction for Recurrent Networks learning the Reber grammar shows an $R^2$ of up to 0.81 given only structural information.

Nanoplasmonic systems combined with optically-active two-dimensional materials provide intriguing opportunities to explore and control light-matter interactions at extreme sub-wavelength lengthscales approaching the exciton Bohr radius. Here, we present room- and cryogenic-temperature investigations of light-matter interactions between an MoSe$_2$ monolayer and individual lithographically defined gold dipole nanoantennas having sub-10 nm feed gaps. By progressively tuning the nanoantenna size, their dipolar resonance is tuned relative to the A-exciton transition in a proximal MoSe$_2$ monolayer achieving a total tuning of $\sim 130\;\mathrm{meV}$. Differential reflectance measurements performed on $> 100$ structures reveal an apparent avoided crossing between exciton and dipolar mode and an exciton-plasmon coupling constant of $g= 55\;\mathrm{meV}$, representing $g/(\hbar\omega_X)\geq3\%$ of the transition energy. This places our hybrid system in the intermediate-coupling regime where spectra exhibit a characteristic Fano-like shape, indicative of the interplay between pronounced light-matter coupling and significant damping. We also demonstrate active control of the optical response by varying the polarization of the excitation light to programmably suppress coupling to the dipole mode. We further study the emerging optical signatures of the monolayer localized at dipole nanoantennas at $10\;\mathrm{K}$. Our findings represent a key step towards realizing non-linear photonic devices based on 2D materials with potential for low-energy and ultrafast performance., Comment: 17 pages, 12 figures, https://pubs.acs.org/doi/10.1021/acs.nanolett.1c02676