Searcharxiv⌕ Search

arXiv subjects

Satish Kumar

Publications and source records attributed to Satish Kumar.

At least 19 recordsLinked to original sources
arXiv2026

VFNet: Multi-View Spatio-Temporal Model for Void Fraction Estimation in Gas-Liquid Two-Phase Flow

Void fraction, which quantifies the proportion of the fluid flow volume occupied by the gas phase, is a key parameter in the characterization of gas-liquid two-phase flow. Existing estimation methods either rely on flow assumptions that do not generalize across different fluids or on intrusive sensing that disturbs the flow behavior. We propose VFNet, a dual-branch spatio-temporal neural network for void-fraction prediction from synchronized multi-view videos of two-phase flow. A local branch extracts features from confined spatial regions and fuses the synchronized dual views, while a spatio-temporal branch captures the global evolution of the flow across space and time to refine a coarse geometric estimate. Trained on simulated computational fluid dynamics (CFD) data with known ground-truth void fractions and evaluated against both learning-based and traditional baselines, VFNet achieves the best performance across a broad range of metrics and also improves downstream flow-pattern classification on real two-phase flow data.

cs.CV↗
arXiv2026

Rationally Enriched Chebyshev Trunk Bases for DeepONet Surrogates of High P\'eclet Entrance Transport

This study demonstrates a rationally enriched Chebyshev (REC) trunk for deep operator network (DeepONet) surrogate models of singularly perturbed and high-P\'eclet transport problems whose solution profiles are characterized by thin localized boundary or wall layers. The REC trunk combines Chebyshev polynomial dictionary elements with rational dictionary elements constructed using the adaptive Antoulas-Anderson (AAA) algorithm. Over five independent training runs, the resulting REC-trunk DeepONet is evaluated against a vanilla DeepONet and a Chebyshev-trunk DeepONet whose prescribed dictionary consists only of Chebyshev polynomials across three problems whose singular perturbation parameters are diffusion-to-advection ratios: a singularly perturbed scalar boundary-value problem (BVP), the thermal entrance problem with a prescribed wall temperature, and the concentration entrance problem with an absorbing wall. Across the held-out test profiles, the REC-trunk DeepONet improves over the vanilla DeepONet and remains comparable to the Chebyshev-trunk DeepONet in predicting the scalar profile, with its clearest advantage over the Chebyshev-trunk DeepONet appearing when the perturbation parameter lies between $1.00\times10^{-4}$ and $1.78\times10^{-4}$, where it reduces the profile-error metrics by up to $19.5\,\%$ relative to the Chebyshev-trunk DeepONet. In predicting the wall-normal temperature and concentration profiles, the REC-trunk DeepONet reduces the profile-error metrics by up to $60.2\,\%$ and $32.2\,\%$ relative to the vanilla and Chebyshev-trunk DeepONets, respectively, while suppressing artificial near-wall oscillations as the P\'eclet or mass-transfer P\'eclet number ranges from $10^{2}$ to $10^{4}$.

cs.LG↗
arXiv2026

Quantum Remote Implementation of Hybrid Operations on Hyperstates Using Hyperentangled States

Quantum remote control, also known as quantum remote implementation of an operator (QRIO), enables the remote manipulation of an arbitrary quantum state by implementing a desired quantum operation at a distant location. Significant progress has recently been made in developing QRIO protocols and their variants. Most existing schemes employ hyperentangled states where entanglement is shared across multiple degrees of freedom (DOFs). However, these protocols typically exploit only one degree of freedom at a time. In this work, we propose a QRIO protocol that simultaneously utilizes the polarization and spatial DOFs of a two-qubit hyperentangled state to remotely implement an arbitrary hybrid operator on an unknown single-photon two-qubit hyperstate. The shared hyperentangled resource is realized using the polarization and spatial modes of photons, while the protocol is constructed using linear optical elements and cross-Kerr nonlinear interactions to facilitate effective photon-photon coupling. Furthermore, the effects of measurement errors arising from finite coherent state distinguishability and coherent state dissipation are analyzed and the corresponding success probability of the protocol is evaluated. The results demonstrate that an appropriate choice of the cross-Kerr phase shift and coherent state amplitude significantly enhances the protocol performance, making the proposed scheme a promising candidate for hybrid quantum communication and distributed quantum information processing.

quant-ph↗
arXiv2026

A Comprehensive Design Framework for Vertical Power Delivery in High-Performance Computing

Power delivery -- including high-to-low voltage conversion, complex power distribution across heterogeneously integrated chiplets, and efficient interconnect allocation -- remains a critical bottleneck in high-performance computing (HPC) systems. Existing vertical power delivery (VPD) solutions are estimated to achieve less than 70\% system-wide end-to-end power delivery efficiency, defined from platform input power to delivered on-chip load power, with substantial energy lost as heat before reaching on-chip point-of-loads (POLs). In the absence of systematic design methodologies, evaluating power quality, exploring architectural alternatives, and optimizing performance rely on computationally prohibitive simulations, resulting in suboptimal designs. This paper introduces an end-to-end scalable power delivery framework for HPC systems, including distributed VPD (DVPD) architecture, DVPD design optimization methodology, and analytical models. The framework leverages substrate-embedded GaN power switches together with arrays of unit inductors and capacitors tailored for HPC applications. Multi-stage power conversion schemes (48V-to-1V, 48V-to-24V-to-1V, and 48V-to-12V-to-1V) are explored, with system-wide voltage drops and power losses evaluated under steady-state conditions. Design specifications for passive and active devices are formulated to meet next-generation efficiency targets. For the 48V-to-1V case, the proposed DVPD approach achieves 84\% system-wide efficiency while occupying 54\% of the area beneath the load system, with efficiency increasing to 87.6\% at 75\% area utilization across a 1--50~kW load range. Furthermore, steady-state voltage drops peak at 2.7\% and transient drops at 9\% (without decoupling capacitors), demonstrating the viability of DVPD for future wafer-scale HPC platforms.

eess.SY↗
arXiv2026

Coexistence of static order and spin dynamics in an S = 5/2 frustrated triangular antiferromagnet

Frustrated triangular-lattice antiferromagnets in the classical high-spin limit provide a paradigmatic setting in which the interplay of competing exchange interactions, anisotropy, and collective degrees of freedom can lead to unconventional low-energy excitations, anomalous criticality, and persistent dynamical responses. Here, we present comprehensive thermodynamic, $\mu$SR, and neutron diffraction experiments, along with first-principles calculations, on a triangular-lattice antiferromagnet, MnSnB$_2$O$_6$, where Mn$^{2+}$ ($S=5/2$) moments form a nearly perfect 2D triangular network without any anti-site disorder. The Curie-Weiss fit to the magnetic susceptibility yields a moderate Curie-Weiss temperature of $-12$ K, indicating dominant antiferromagnetic interactions between Mn$^{2+}$ moments, which is supported by first-principles calculations. Specific-heat measurements reveal the onset of long-range magnetic order at $T_{\rm N}\approx 1$ K, which is ascribed to intraplane exchange interactions. The specific heat exhibits pronounced short-range correlations above $T_{\rm N}$ and an unconventional power-law behavior, $C\propto T^{1.37}$, deep in the ordered state, suggesting the presence of non-trivial low-energy excitations. Zero-field $\mu$SR experiments down to 50~mK confirm the presence of magnetic ordering below $T_{\rm N}$, in agreement with thermodynamic and neutron diffraction experiments. The $\mu$SR measurements detect persistent spin dynamics coexisting with static magnetic order. The temperature evolution of the order parameter down to 50~mK from neutron diffraction suggests that the ordered state is consistent with a 3D Ising-like antiferromagnet. This family of archetypal frustrated magnets offers a promising venue for the experimental realization of emergent phenomena governed by competing exchange interactions and exotic low-energy excitations.

cond-mat.str-el↗
arXiv2026

Strain induced magnetic phase transitions in Fe3GeTe2 monolayer

We investigate the magnetic properties of a monolayer of Fe3GeTe2 as a function of the lattice constant by combining first-principles calculations with atomistic spin dynamics simulations. The calculated magnetic exchange interactions reveal a competition between ferromagnetic and antiferromagnetic couplings, with the latter being significantly strengthened under compressive strain. Stochastic Landau-Lifshitz-Gilbert simulations reveal a substantial decrease in the Curie temperature with decreasing lattice constant, and predict a transition of the magnetic ground state from a ferromagnetic configuration to a conical spin-spiral state. We introduce a simple spin-model which explains the stabilization of the spiral phase due to competing exchange interactions. We found multiple magnetic phase transitions involving ferromagnetic, conical spin-spiral, and planar Neel states, depending on both the lattice constant and the temperature. The absence of Dzyaloshinskii-Moriya interactions is found to significantly reduce the Neel temperature, while leaving the Curie temperature largely unaffected. Our findings reveal the importance of lattice distortions in controlling complex magnetic phases and their evolution with temperature.

cond-mat.mtrl-sci↗
arXiv2026

Shear-driven dynamics of surfactant-laden droplets on rough substrates

The depinning of liquid droplets due to flow of a surrounding immiscible fluid plays a crucial role in applications such as enhanced oil recovery, surface cleaning, and crossflow emulsification. Although surfactants are often present in these systems, the role of Marangoni stresses on droplet depinning by an external flow remains unclear. To address this, we develop a lubrication-theory-based model for a thin Newtonian droplet laden with insoluble surfactant on a substrate with Gaussian-shaped defects which are used to account for the effects of surface roughness. The droplet is surrounded by a surfactant-free immiscible Newtonian fluid in a long, narrow rectangular channel, with flow driven by an applied pressure gradient. Using a precursor-film/disjoining-pressure approach for contact-line motion, we derive nonlinear evolution equations for the droplet thickness and interfacial surfactant concentration, which are solved numerically. The pressure gradient transports surfactant from the receding to the advancing contact line, generating a Marangoni flow opposing the pressure-driven flow. This reduces the net shear force on the droplet, leading to depinning at a higher critical pressure gradient. These findings reveal a previously unexamined regime in which interfacial Marangoni stresses, rather than uniform interfacial-tension reduction, govern the critical flow rate. The results provide a mechanistic basis for using surfactant-concentration gradients as a tunable handle to control droplet motion on rough substrates.

physics.flu-dyn↗
arXiv2026

Anharmonic Quantum Transport Analysis of Thermal Transport Anomalies in Ultrathin Silicon Nanowires

Thermal transport in low-dimensional semiconductors is crucial for advancing thermal management in nanoelectronics, quantum devices, and thermoelectric devices. Recent molecular dynamics (MD) studies have identified a nonmonotonic dependence of thermal conductivity (k) on diameter in ultrathin silicon nanowires (NWs). However, classical MD methods are limited at low temperatures and in strongly confined regimes. This work introduces a fully quantum-mechanical perspective on this anomaly by employing anharmonic non-equilibrium Green's function (NEGF) simulations combined with density-functional-theory-trained neuroevolution potentials. For [001]- and [110]-oriented NWs, k decreases with diameter d to a minimum at d_c = 6.24 nm and 5.50 nm, respectively, then rises with d, for a temperature range of 10-300 K. At room temperature, this behavior arises from dominant momentum-conserving normal scattering relative to Umklapp processes in confined regimes, thereby enabling Poiseuille-like hydrodynamic phonon flow that competes with boundary scattering. At cryogenic temperatures, strong radial confinement quantizes the phonon spectrum, and only low-frequency phonons (< 2 THz) significantly contribute to heat transport through quasi-ballistic propagation of long-wavelength modes, as demonstrated by the spectral thermal conductance. In contrast to classical MD, which is inaccurate at low temperatures due to overexcitation of high-frequency vibrations by Boltzmann statistics, neglect of quantum suppression, and overestimation of thermal conductivity in thinner NWs with stronger quantum confinement, the NEGF framework provides quantitative accuracy even at low temperatures, such as 10 K.

cond-mat.mes-hall↗
arXiv2026

Signature of spin liquid state in a frustrated 3D antiferromagnet

Frustrated pyrochlore lattices in transition-metal oxides provide an ideal platform for realizing exotic quantum states, including spin liquids with unconventional low-energy excitations arising from the macroscopic ground-state degeneracy of corner-sharing tetrahedral networks. Here, we report the synthesis and comprehensive characterization of ZnCrGaO$4$, a frustrated three-dimensional pyrochlore-like magnet in which intrinsic cation ordering gives rise to unavoidable atomic-site disorder. A Curie--Weiss analysis of the high-temperature magnetic susceptibility yields a large negative Curie--Weiss temperature, $\theta{\mathrm{CW}} \approx -205$ K, indicating dominant antiferromagnetic exchange interactions ($J/k_{\mathrm{B}} \sim 55$ K) between Cr$^{3+}$ ($S = 3/2$) moments. Despite the presence of strong antiferromagnetic interactions, no signature of long-range magnetic ordering is observed down to 125 mK, as evidenced by specific-heat and ac-susceptibility measurements. Furthermore, the absence of bifurcation between zero-field-cooled and field-cooled dc magnetic susceptibilities measured at 0.01 T indicates the absence of spin freezing, which is further supported by the frequency-independent ac susceptibility down to 250 mK. The presence of broad maxima in the magnetic specific heat and ac susceptibility at low temperatures suggests the development of short-range spin correlations within a dynamic magnetic state. In addition, the low-temperature specific heat follows a power-law behavior below 1 K, indicating the presence of unconventional low-energy excitations and algebraic spin correlations. These results provide compelling evidence for a dynamic correlated ground state in ZnCrGaO$_4$, establishing it as a promising platform for exploring highly frustrated $S > 1/2$ three-dimensional quantum magnets and potential spin-liquid behavior.

cond-mat.str-el↗
arXiv2026

RareSpot+: A Benchmark, Model, and Active Learning Framework for Small and Rare Wildlife in Aerial Imagery

Automated wildlife monitoring from aerial imagery is vital for conservation but remains limited by two persistent challenges: the difficulty of detecting small, rare species and the high cost of large-scale expert annotation. Prairie dogs exemplify this problem -- they are ecologically important yet appear tiny, sparsely distributed, and visually indistinct from their surroundings, posing a severe challenge for conventional detection models. To overcome these limitations, we present RareSpot+, a detection framework that integrates multi-scale consistency learning, context-aware augmentation, and geospatially guided active learning to address these issues. A novel multi-scale consistency loss aligns intermediate feature maps across detection heads, enhancing localization of small (approx. 30 pixels wide) objects without architectural changes, while context-aware augmentation improves robustness by synthesizing hard, ecologically plausible examples. A geospatial active learning module exploits domain-specific spatial priors linking prairie dogs and burrows, together with test-time augmentation and a meta-uncertainty model, to reduce redundant labeling. On a 2 km^2 aerial dataset, RareSpot+ improves detection over the baseline mAP@50 by +35.2% (absolute +0.13). Cross-dataset tests on HerdNet, AED, and several other wildlife benchmarks demonstrate robust detector-level transferability. The active learning module further boosts prairie dog AP by 14.5% using an annotation budget of just 1.7% of the unlabeled tiles. Beyond detection, RareSpot+ enables spatial ecological analyses such as clustering and co-occurrence, linking vision-based detection with quantitative ecology.

cs.CV↗
arXiv2026

Identifying Topological Differences in Two Populations of Random Geometric Objects

We propose a statistical framework to identify topological differences in two populations of random geometric objects. The proposed framework involves first associating a topological signature with random geometric objects and then performing a two-sample test using the observed topological signatures. We associate persistence barcodes, a topological signature from topological data analysis, with each observed random geometric object. This, in turn, yields a two-sample problem on the space of persistence barcodes. As the space of persistence barcodes is not suitable for standard statistical analysis, we translate the two-sample problem on a suitable subset of a Euclidean space. In the course of this study, we embed the topological signatures in an ordered convex cone in a Euclidean space using functions from tropical geometry. We show that the embedding is a sufficient statistic for the persistence barcodes. This fact leads to the proposal of a two-sample test based on this sufficient statistic, and its equivalence to the two-sample problem on the barcode space is established. Finally, the consistency of the proposed test is studied.

stat.ME↗
arXiv2026

Emergent dimensional reduction in a distorted kagome magnet $\mathrm{YCa_3(CrO)_3(BO_3)_4}$ driven by exchange hierarchy

Frustrated kagome magnets provide a fertile platform for unconventional collective quantum phenomena, yet the role of lattice distortion in reorganizing magnetic degrees of freedom and controlling low-energy physics remains poorly understood. Here we report a rare realization of dimensional reduction in the distorted kagome material $\mathrm{YCa_3(CrO)_3(BO_3)_4}$, combining thermodynamic experiments with first-principles calculations and large-scale Monte Carlo simulations. Magnetic susceptibility and specific heat show no signatures of spin freezing or long-range magnetic order down to $65~\mathrm{mK}$ despite strong antiferromagnetic interactions. Instead, the susceptibility exhibits a broad maximum characteristic of quasi-one-dimensional spin correlations, while the magnetic specific heat follows a robust power law $C_{\mathrm{mag}}\sim T^2$ over more than a decade in temperature that remains unchanged in applied magnetic fields. This field-independent scaling rules out impurity or conventional magnon contributions and points to a collective low-energy excitation spectrum governed by frustration and local constraints. We show that a strongly hierarchical exchange network reorganizes the system into local antiferromagnetic dimers and weakly coupled spin chains, with frustrated inter-unit couplings suppressing three-dimensional order to ultralow temperatures. Our results demonstrate how a hierarchy of competing exchange interactions can reorganize a frustrated three-dimensional magnet into effectively lower-dimensional correlated units, stabilizing extended regimes of quantum-disordered behavior in realistic materials.

cond-mat.str-el↗
arXiv2026

Inertia-Dilatancy Interplay Governs Shear-Thickening Drop Impact

Combining high-speed photography with direct force measurements, we investigate the impact dynamics of drops of cornstarch-water mixtures -- a premier example of shear-thickening fluids -- across a wide range of impact conditions. Our study identifies three distinct impact regimes. In addition to the liquid-like and solid-like behaviors generally expected for the impact-induced response of shear-thickening fluids, we uncover a counterintuitive regime in which high-concentration cornstarch-water mixtures display a liquid-like response at the onset of impact when shear rates are high and only transition to a solid-like behavior at later times as shear rates reduce. By integrating the classic drop-impact theory with the Reynolds-Darcy mechanism for dilatancy, we develop a unified model that quantitatively describes the impact dynamics of shear-thickening drops across all regimes. Our work reveals the unexpected response of shear-thickening fluids to ultra-fast deformation and advances fundamental understanding of drop impact for complex fluids.

physics.flu-dyn↗
arXiv2025

Revealing Phonon Bridge Effect for Amorphous vs Crystalline Metal-Silicide Layers at Si/Ti Interfaces by a Machine Learning Potential

Metal-semiconductor interfaces play a central role in micro and nano-electronic devices as heat dissipation or temperature drop across these interfaces can significantly affect device performance. Prediction of accurate thermal boundary resistance (TBR) across these interfaces, considering realistic structures and their correlation with underlying thermal transport, remains challenging. In this work we develop a unified Neuroevolution Potential (NEP) for the Si-Ti system that accurately reproduces energies, forces, and phonon properties of bulk Si, Ti, and TiSi2 and extends naturally to interfacial environments to analyze interfacial transport. An important development over current machine-learned interatomic potentials is the capability to model complex structures at metal-semiconductor interfaces, as the NEP enables large scale non-equilibrium molecular dynamics simulations of epitaxial Si/Ti interfaces to elucidate the effect of amorphous or crystalline silicide interfacial layers. Simulated TBRs show excellent agreement with our time-domain thermoreflectance (TDTR) measurements. Spectral analyses reveal that amorphous TiSi2 interfacial layer helps in efficient interfacial transport when the thickness is less than 1.5 nm compared to the crystalline TiSi2 layer, but this trend reverses when the interfacial layer thickness increases beyond 1.5 nm. Comparison of TBRs at Si/TiSi2 interface for different crystalline phases of TiSi2 establishes that C54 phase has reduced TBR compared to C49 phase, which is correlated with the difference in their phonon density of states (PDOS) overlap with Si. These results provide atomistic insight into the role of crystalline versus amorphous silicides in interfacial heat transport and demonstrate a transferable machine-learned potential for studying heat dissipation in advanced semiconductor devices.

cond-mat.mtrl-sci↗
arXiv2025

Seeing new depths: Three-dimensional flow of a free-swimming alga

A swimming microorganism stirs the surrounding fluid, creating a flow field that governs not only its locomotion and nutrient uptake, but also its interactions with other microorganisms and the environment. Despite its fundamental importance, capturing this flow field and unraveling its biological implications remains a challenge. Here, we report the first direct, time-resolved measurements of the three-dimensional (3D) flow field generated by a single, free-swimming microalga, Chlamydomonas reinhardtii, a model organism for microbial locomotion and flagellar dynamics. Supported by hydrodynamic modeling and simulations, our measurements resolve how established two-dimensional (2D) flow features such as in-plane vortices and the stagnation point emerge from and shape the full algal flow in 3D. Moreover, we reveal unexpected low-Reynolds-number flow phenomena including micron-sized vortex rings and periodically recurring translating vortices and uncover topological changes in the underlying flow structure associated with the puller-to-pusher transition of an alga. Biologically, access to the 3D flow field enables rigorous quantification of the alga's energy expenditure, as well as its swimming and feeding efficiency, improving the precision of these physiological metrics. Taken together, our study demonstrates rich vortex dynamics in inertialess flows and shows their influence on microbial motility. The work also introduces a new experimental method for mapping the fluid environment sculpted by beating flagella.

physics.flu-dyn↗
arXiv2025

A Novel Post-Quantum Secure Digital Signature Scheme Based on Neural Network

Digital signatures are fundamental cryptographic primitives that ensure the authenticity and integrity of digital documents. In the post-quantum era, classical public key-based signature schemes become vulnerable to brute-force and key-recovery attacks due to the computational power of quantum algorithms. Multivariate polynomial based signature schemes are among the one of the cryptographic constructions that offers strong security guarantees against such quantum threats. With the growing capabilities of neural networks, it is natural to explore their potential application in the design of cryptographic primitives. Neural networks inherently captures the non-linear relationships within the data, which are encoded in their synaptic weight matrices and bias vectors. In this paper, we propose a novel construction of a multivariate polynomial based digital signature scheme that leverages neural network architectures. A neural network with binary weights is employed to define the central structure of the signature scheme. The design introduces a recurrent random vector, functionally analogous to an attention mechanism, which contributes dynamic randomness based on the previous state, thereby enhancing the scheme's security. It is demonstrated that the proposed signature scheme provide security against Existential Unforgeability under adaptive Chosen-Message Attacks (EUF-CMA). Furthermore, it is proven that direct attacks aimed to recover the private keys are computationally infeasible within polynomial time, even in the presence of quantum computing abilities. The operational characteristics of the proposed scheme are also evaluated, with results indicating notable efficiency and practical viability in post-quantum cryptographic applications.

cs.CR↗
arXiv2025

Design and Experimental Realization of Various Protocols for Secure Quantum Computation and Communication

A set of new schemes for quantum computation and communication have been either designed or experimentally realized using optimal quantum resources. A multi-output quantum teleportation scheme, where a sender (Alice) teleports an m and m+1-qubit GHZ-like unknown state to a receiver (Bob), has been demonstrated using two copies of the Bell state instead of a five-qubit cluster state and implemented on IBM's quantum computer for the m=1 case. Another scheme, known as quantum broadcasting where a known state is sent to two spatially separated parties (Bob and Charlie) has also been realized using two Bell states. It is shown that existing quantum broadcasting schemes can be reduced to multiparty remote state preparation. After achieving teleportation of unknown and known states, sending a quantum operator becomes the next step. A scheme for remote implementation of operators (RIO), specifically a controlled joint-RIO (CJRIO), has been proposed using a four-qubit hyper-entangled state involving spatial and polarization degrees of freedom. In this direction, two more variants, remote implementation of hidden and partially unknown operators (RIHO and RIPUO) have also been proposed. Their success probabilities are analyzed considering dissipation of an auxiliary coherent state interacting with the environment. For secure multiparty tasks like quantum voting or auction, secure multiparty quantum computation (SMQC) becomes essential. A quantum anonymous voting (QAV) scheme has been experimentally implemented on IBM's quantum computer. Finally, two quantum key distribution (QKD) protocols, coherent one-way (COW) and differential phase shift (DPS), are experimentally demonstrated and the key rates are analyzed as functions of post-processing parameters and detector dead times across various distances.

quant-ph↗
arXiv2025

RareSpot: Spotting Small and Rare Wildlife in Aerial Imagery with Multi-Scale Consistency and Context-Aware Augmentation

Automated detection of small and rare wildlife in aerial imagery is crucial for effective conservation, yet remains a significant technical challenge. Prairie dogs exemplify this issue: their ecological importance as keystone species contrasts sharply with their elusive presence--marked by small size, sparse distribution, and subtle visual features--which undermines existing detection approaches. To address these challenges, we propose RareSpot, a robust detection framework integrating multi-scale consistency learning and context-aware augmentation. Our multi-scale consistency approach leverages structured alignment across feature pyramids, enhancing fine-grained object representation and mitigating scale-related feature loss. Complementarily, context-aware augmentation strategically synthesizes challenging training instances by embedding difficult-to-detect samples into realistic environmental contexts, significantly boosting model precision and recall. Evaluated on an expert-annotated prairie dog drone imagery benchmark, our method achieves state-of-the-art performance, improving detection accuracy by over 35% compared to baseline methods. Importantly, it generalizes effectively across additional wildlife datasets, demonstrating broad applicability. The RareSpot benchmark and approach not only support critical ecological monitoring but also establish a new foundation for detecting small, rare species in complex aerial scenes.

cs.CV↗