arXiv subjects

Alex Zhang

Publications and source records attributed to Alex Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Numerical Model Simulation of the Carruthers GCI Images

astro-ph.IM↗

arXiv2026

Carruthers Data Processing Pipeline: Photon Background Removal

The Carruthers Geocorona Observatory, launched in September 2025, is NASA's first mission devoted to investigating the fundamental nature of Earth's exosphere from its distant vantage in halo orbit around the Earth-Sun Lagrange 1 (L1) point. Its primary payload, the GeoCoronal Imager, consists of two coaligned photometric imagers that measure ultraviolet Lyman-alpha emission radiance from exospheric hydrogen simultaneously at wide- and narrow- fields of view. These observations will map the exosphere's global spatial structure and observe its temporal variability in response to geomagnetic storms. However, a critical step in that analysis is isolating the in-band exospheric hydrogen Lyman-alpha signal from any other source of photons, including in-band InterPlanetary Hydrogen photon background and out-of-band photon backgrounds, which are emitted from Earth's limb. This paper details the algorithms used to retrieve and remove photon backgrounds from the GCI science images acquired on-orbit. Finally, the science data processing pipeline that transforms instrument-effect corrected images (L1B science data product) into absolutely-calibrated exospheric H measurements in physical units (L1C science data product) is detailed. Evaluation of algorithm performance based on a realistic pre-flight case study using synthetic data demonstrates that these photon background removal algorithms achieve high accuracy, leaving a residual systematic bias in isolated exospheric Lyman-alpha of only 3% under beginning-of-life conditions.

astro-ph.IM↗

arXiv2026

Carruthers Data Processing Pipeline: Radiometric Responsivity Calibration

The Carruthers Geocorona Observatory is NASA's first mission dedicated to investigating the fundamental nature of Earth's exosphere and its dynamic response to space weather. Its primary payload, the GeoCoronal Imager, consists of two co-aligned broadband photometric imagers that support simultaneous, common-volume sensing of ultraviolet emission at 121.6 nm (Lyman-alpha) by exospheric hydrogen atoms. Accurate exospheric parameter retrieval from these images requires accurate knowledge of the instrument's optical responsivity, which enables conversion of measured exospheric signal rates into the scientifically relevant quantity of emission radiance. The Carruthers mission achieves absolute sensitivity calibration through the acquisition of photometric measurements of stars and subsequent inversion of the observed fluxes to retrieve the wavelength-dependent responsivity across the passband for each imaging configuration. The retrieval algorithm performance is enhanced by its incorporation of an objective, algorithm-driven ranking criterion to systematically select target calibration stars from a stellar spectral library. The end-to-end workflow, from the stellar selection criterion to the passband inversion, is validated using synthetically generated stellar measurements. These validation tests establish that our optical responsivity retrieval approach has high recovery fidelity, achieving error rates of <5% at the Lyman-alpha wavelength for all primary science imaging modes.

astro-ph.IM↗

arXiv2026

Deep RL for Fast Long-Horizon Operations Scheduling on NASA's Carruthers Geocorona Observatory Mission

Spacecraft operations scheduling is a highly constrained, long-horizon combinatorial optimization problem that traditionally relies on heuristics, constraint programming, or manual planning. We present a scalable deep reinforcement learning framework developed and deployed for NASA's Carruthers Geocorona Observatory mission. Our framework introduces a macro-action abstraction known as activity blocks coupled with dynamic action-masking to navigate the intractably large search space and strictly enforce complex power, thermal, and instrument constraints. The resulting architecture generates globally feasible schedules with overwhelming probability, establishes operational trust, and executes a full training cycle in under six hours, circumventing the need for policy robustness by enabling rapid, on-demand retraining. Further, resulting schedules outperform baseline heuristics in scheduled science quality. The deep reinforcement learning framework was deployed as the default operational scheduler for the Carruthers Geocorona Observatory mission from the outset of the mission, demonstrating that deep reinforcement learning can be trusted for real spacecraft operations under complex, evolving constraints.

astro-ph.IM↗

arXiv2026

On-orbit Calibration of the Carruthers GCI: Instrument Effect Correction

The Carruthers Geocorona Observatory -- launched in September 2025 -- is NASA's first mission devoted to investigating the fundamental nature of Earth's exosphere from its distant vantage in halo orbit around the Earth-Sun L1 Lagrange point. Its primary payload, the GeoCoronal Imager, consists of two coaligned photometric imagers that measure ultraviolet Lyman-alpha emission radiance from exospheric hydrogen simultaneously at wide- and narrow- fields of view. The imagers use Micro Channel Plate intensified Complimentary Metal Oxide Semiconductor detectors, which are known to add various artifacts to the final image telemetered from the spacecraft, hereby known as instrument effects. This paper details the algorithms used to retrieve and remove instrument effects from raw telemetry on-orbit, including detector voltage bias, thermal dark current, particle radiation, flat-field, and distortion. Finally, the science data processing pipeline from raw telemetry to instrument-effect corrected images is detailed. Algorithm performance is measured via a synthetic numerical image generator or validated on pre-launch experiments.

astro-ph.IM↗

arXiv2026

LLMZero: Discovering Adaptive Training Strategies for RL Post-Training via LLM Agents

RL post-training strategies are dataset-dependent and reveal a recurring empirical pattern: capacity parameters accumulate monotonically across stages, while regularization parameters predominantly oscillate in response to shifting training dynamics. This distinction highlights a potential flaw in fixed training schedules: by forcing all parameters along rigid paths, they fail to capture the dynamic exploration-exploitation tradeoffs that regularization must track. We uncover this through LLMZero, an agentic system that optimizes training trajectories via tree search by diagnosing pathologies at each checkpoint and proposing coordinated multi-parameter transitions. Across four diverse GRPO tasks, LLMZero discovers strategies that improve over the base model by 9% to 140% and over grid search by 6% to 15% (relative), consistently outperforming random search and a skill-based agent under a matched compute budget. The capacity--regularization asymmetry is consistent across all four tasks, offering a candidate design heuristic for multi-stage training.

cs.LG↗

arXiv2026

Private Vertical Federated Inference for Time-Series

Institutions may benefit from collaborative inference on time-series data. In settings where privacy is necessary, multi-party computation (MPC) is a straightforward approach to providing strong guarantees, yet it remains prohibitively expensive and scales poorly with modern transformer architectures. Vertical Federated Learning (VFL) offers efficiency but suffers from privacy leakage at the embedding level, and securing the entire VFL model head via MPC remains prohibitively slow and communication-heavy for larger models. To enable practical, secure inference at scale, we propose "Public/Private Hybrid Head-VFL" (PPHH-VFL). This hybrid architecture splits the model head into an efficient plaintext public head and a secure, lightweight MPC private head. By applying adversarial training to the public embeddings, we mitigate privacy leakage; concurrently, the small private head securely preserves the flow of sensitive information needed for high downstream utility. Empirical evaluations on models ranging up to 86 million parameters demonstrate that PPHH-VFL accelerates inference by up to six orders of magnitude compared to end-to-end MPC. Compared to a standard VFL+MPC baseline, our approach scales significantly better, achieving a speedup of up to 44.4x in WAN and a 91.2x reduction in communication costs (dropping from 1.7 GB to 19 MB per batch), while simultaneously improving downstream classification accuracy by 2.50% and regression RMSE by 40.7%.

cs.LG↗

arXiv2026

Transition Waves in Mechanical Metamaterials with Neighbor-Programmable Energy Landscapes

Transition waves in mechanical metamaterials manifest themselves as propagating interfaces between different stable states in lattices composed of arrays of coupled, intrinsically bistable elements. Here, we show experimentally and numerically that arrays of elastic unit cells that are individually monostable, yet whose energy landscapes can be programmed through interactions with neighboring units, provide a rich and largely unexplored platform for transition wave propagation. We implement this concept by designing a unit cell comprising a von Mises truss supported by two vertical elastic beams. In one-dimensional arrays of such units, we demonstrate that each cell's energy landscape can change from monostable to bistable depending on the state of its neighbors. This neighbor-programmable energy landscape enables the controlled initiation and propagation of transition waves, giving rise to highly discrete, directionally unbiased, domino-like wave propagation. Experiments and numerical simulations show that the existence and speed of the waves are governed by geometric design and mass distribution. Our results establish neighboring effects as a distinct mechanism for transition wave propagation, expanding the design space of mechanical metamaterials beyond architectures that rely on intrinsically multistable building blocks.

physics.app-ph↗

arXiv2026

Now You See That: Learning End-to-End Humanoid Locomotion from Raw Pixels

Achieving robust vision-based humanoid locomotion remains challenging due to two fundamental issues: the sim-to-real gap introduces significant perception noise that degrades performance on fine-grained tasks, and training a unified policy across diverse terrains is hindered by conflicting learning objectives. To address these challenges, we present an end-to-end framework for vision-driven humanoid locomotion. For robust sim-to-real transfer, we develop a high-fidelity depth sensor simulation that captures stereo matching artifacts and calibration uncertainties inherent in real-world sensing. We further propose a vision-aware behavior distillation approach that combines latent space alignment with noise-invariant auxiliary tasks, enabling effective knowledge transfer from privileged height maps to noisy depth observations. For versatile terrain adaptation, we introduce terrain-specific reward shaping integrated with multi-critic and multi-discriminator learning, where dedicated networks capture the distinct dynamics and motion priors of each terrain type. We validate our approach on two humanoid platforms equipped with different stereo depth cameras. The resulting policy demonstrates robust performance across diverse environments, seamlessly handling extreme challenges such as high platforms and wide gaps, as well as fine-grained tasks including bidirectional long-term staircase traversal.

cs.RO↗

arXiv2025

Multi-Scale High-Resolution Logarithmic Grapher Module for Efficient Vision GNNs

Vision graph neural networks (ViG) have demonstrated promise in vision tasks as a competitive alternative to conventional convolutional neural nets (CNN) and transformers (ViTs); however, common graph construction methods, such as k-nearest neighbor (KNN), can be expensive on larger images. While methods such as Sparse Vision Graph Attention (SVGA) have shown promise, SVGA's fixed step scale can lead to over-squashing and missing multiple connections to gain the same information that could be gained from a long-range link. Through this observation, we propose a new graph construction method, Logarithmic Scalable Graph Construction (LSGC) to enhance performance by limiting the number of long-range links. To this end, we propose LogViG, a novel hybrid CNN-GNN model that utilizes LSGC. Furthermore, inspired by the successes of multi-scale and high-resolution architectures, we introduce and apply a high-resolution branch and fuse features between our high-resolution and low-resolution branches for a multi-scale high-resolution Vision GNN network. Extensive experiments show that LogViG beats existing ViG, CNN, and ViT architectures in terms of accuracy, GMACs, and parameters on image classification and semantic segmentation tasks. Our smallest model, Ti-LogViG, achieves an average top-1 accuracy on ImageNet-1K of 79.9% with a standard deviation of 0.2%, 1.7% higher average accuracy than Vision GNN with a 24.3% reduction in parameters and 35.3% reduction in GMACs. Our work shows that leveraging long-range links in graph construction for ViGs through our proposed LSGC can exceed the performance of current state-of-the-art ViGs. Code is available at https://github.com/mmunir127/LogViG-Official.

cs.CV↗

arXiv2025

RAISE: A self-driving laboratory for interfacial property formulation discovery

Surface wettability is a critical design parameter for biomedical devices, coatings, and textiles. Contact angle measurements quantify liquid-surface interactions, which depend strongly on liquid formulation. Herein, we present the Robotic Autonomous Imaging Surface Evaluator (RAISE), a closed-loop, self-driving laboratory that is capable of linking liquid formulation optimization with surface wettability assessment. RAISE comprises a full experimental orchestrator with the ability of mixing liquid ingredients to create varying formulation cocktails, transferring droplets of prepared formulations to a high-throughput stage, and using a pick-and-place camera tool for automated droplet image capture. The system also includes an automated image processing pipeline to measure contact angles. This closed loop experiment orchestrator is integrated with a Bayesian Optimization (BO) client, which enables iterative exploration of new formulations based on previous contact angle measurements to meet user-defined objectives. The system operates in a high-throughput manner and can achieve a measurement rate of approximately 1 contact angle measurement per minute. Here we demonstrate RAISE can be used to explore surfactant wettability and how surfactant combinations create tunable formulations that compensate for purity-related variations. Furthermore, multi-objective BO demonstrates how precise and optimal formulations can be reached based on application-specific goals. The optimization is guided by a desirability score, which prioritizes formulations that are within target contact angle ranges, minimize surfactant usage and reduce cost. This work demonstrates the capabilities of RAISE to autonomously link liquid formulations to contact angle measurements in a closed-loop system, using multi-objective BO to efficiently identify optimal formulations aligned with researcher-defined criteria.

cs.RO↗

arXiv2025

VCMamba: Bridging Convolutions with Multi-Directional Mamba for Efficient Visual Representation

Recent advances in Vision Transformers (ViTs) and State Space Models (SSMs) have challenged the dominance of Convolutional Neural Networks (CNNs) in computer vision. ViTs excel at capturing global context, and SSMs like Mamba offer linear complexity for long sequences, yet they do not capture fine-grained local features as effectively as CNNs. Conversely, CNNs possess strong inductive biases for local features but lack the global reasoning capabilities of transformers and Mamba. To bridge this gap, we introduce \textit{VCMamba}, a novel vision backbone that integrates the strengths of CNNs and multi-directional Mamba SSMs. VCMamba employs a convolutional stem and a hierarchical structure with convolutional blocks in its early stages to extract rich local features. These convolutional blocks are then processed by later stages incorporating multi-directional Mamba blocks designed to efficiently model long-range dependencies and global context. This hybrid design allows for superior feature representation while maintaining linear complexity with respect to image resolution. We demonstrate VCMamba's effectiveness through extensive experiments on ImageNet-1K classification and ADE20K semantic segmentation. Our VCMamba-B achieves 82.6% top-1 accuracy on ImageNet-1K, surpassing PlainMamba-L3 by 0.3% with 37% fewer parameters, and outperforming Vision GNN-B by 0.3% with 64% fewer parameters. Furthermore, VCMamba-B obtains 47.1 mIoU on ADE20K, exceeding EfficientFormer-L7 by 2.0 mIoU while utilizing 62% fewer parameters. Code is available at https://github.com/Wertyuui345/VCMamba.

cs.CV↗

arXiv2025

Learning-Based Modeling of a Magnetically Steerable Soft Suction Device for Endoscopic Endonasal Interventions

This paper introduces a learning-based modeling framework for a magnetically steerable soft suction device designed for endoscopic endonasal brain tumor resection. The device is miniaturized (4 mm outer diameter, 2 mm inner diameter, 40 mm length), 3D printed using biocompatible SIL 30 material, and integrates embedded Fiber Bragg Grating (FBG) sensors for real-time shape feedback. Shape reconstruction is represented using four Bezier control points, providing a compact representation of deformation. A data-driven model was trained on 5,097 experimental samples to learn the mapping from magnetic field parameters (magnitude: 0-14 mT, frequency: 0.2-1.0 Hz, vertical tip distances: 90-100 mm) to Bezier control points defining the robot's 3D shape. Both Neural Network (NN) and Random Forest (RF) architectures were compared. The RF model outperformed the NN, achieving a mean RMSE of 0.087 mm in control point prediction and 0.064 mm in shape reconstruction error. Feature importance analysis revealed that magnetic field components predominantly influence distal control points, while frequency and distance affect the base configuration. Unlike prior studies applying general machine learning to soft robotic data, this framework introduces a new paradigm linking magnetic actuation inputs directly to geometric Bezier control points, creating an interpretable, low-dimensional deformation representation. This integration of magnetic field characterization, embedded FBG sensing, and Bezier-based learning provides a unified strategy extensible to other magnetically actuated continuum robots. By enabling sub-millimeter shape prediction and real-time inference, this work advances intelligent control of magnetically actuated soft robotic tools in minimally invasive neurosurgery.

cs.RO↗

arXiv2025

DTVM: Revolutionizing Smart Contract Execution with Determinism and Compatibility

We introduce the DeTerministic Virtual Machine (DTVM) Stack, a next-generation smart contract execution framework designed to address critical performance, determinism, and ecosystem compatibility challenges in blockchain networks. Building upon WebAssembly (Wasm) while maintaining full Ethereum Virtual Machine (EVM) ABI compatibility, DTVM introduces a Deterministic Middle Intermediate Representation (dMIR) and a hybrid lazy-JIT compilation engine to balance compilation speed and execution efficiency. DTVM further accommodates diverse instruction set architectures (e.g., EVM, RISC-V) through modular adaptation layers. This enables seamless integration with DTVM's hybrid lazy-JIT compilation engine, which dynamically optimizes performance while preserving deterministic execution guarantees across heterogeneous environments. The key contributions including: 1). The framework achieves up to 2$\times$ acceleration over evmone in dominant Ethereum contract (e.g. ERC20/721/1155) execution and reduces fibonacci computation latency by 11.8$\sim$40.5% compared to Wasm based VMs. 2). A novel trampoline hot-switch mechanism enables sub-millisecond (0.95ms) post-deployment invocation times, outperforming up to about 23$\times$ in compilation and invocation efficiency. 3). It supports multi-language development (Solidity, C++, Rust, Java, Go, and AssemblyScript) through unified bytecode conversion while maintaining EVM ABI compatibility for seamless invocation. It reduces machine code object sizes by 30.0$\sim$72.6%, coupled with a minimized Trusted Computing Base. 4). It offers SmartCogent, an AI-driven full-stack development experience, leveraging fine-tuned LLMs and retrieval-augmented generation to automate tasks across the smart contract lifecycle: development, debugging, security auditing, and deployment. DTVM Stack has been open-sourced (https://github.com/DTVMStack).

cs.DC↗

arXiv2025

The Amazon Nova Family of Models: Technical Report and Model Card

We present Amazon Nova, a new generation of state-of-the-art foundation models that deliver frontier intelligence and industry-leading price performance. Amazon Nova Pro is a highly-capable multimodal model with the best combination of accuracy, speed, and cost for a wide range of tasks. Amazon Nova Lite is a low-cost multimodal model that is lightning fast for processing images, video, documents and text. Amazon Nova Micro is a text-only model that delivers our lowest-latency responses at very low cost. Amazon Nova Canvas is an image generation model that creates professional grade images with rich customization controls. Amazon Nova Reel is a video generation model offering high-quality outputs, customization, and motion control. Our models were built responsibly and with a commitment to customer trust, security, and reliability. We report benchmarking results for core capabilities, agentic performance, long context, functional adaptation, runtime performance, and human evaluation.

cs.AI↗

arXiv2024

Language-Guided World Models: A Model-Based Approach to AI Control

This paper introduces the concept of Language-Guided World Models (LWMs) -- probabilistic models that can simulate environments by reading texts. Agents equipped with these models provide humans with more extensive and efficient control, allowing them to simultaneously alter agent behaviors in multiple tasks via natural verbal communication. In this work, we take initial steps in developing robust LWMs that can generalize to compositionally novel language descriptions. We design a challenging world modeling benchmark based on the game of MESSENGER (Hanjie et al., 2021), featuring evaluation settings that require varying degrees of compositional generalization. Our experiments reveal the lack of generalizability of the state-of-the-art Transformer model, as it offers marginal improvements in simulation quality over a no-text baseline. We devise a more robust model by fusing the Transformer with the EMMA attention mechanism (Hanjie et al., 2021). Our model substantially outperforms the Transformer and approaches the performance of a model with an oracle semantic parsing and grounding capability. To demonstrate the practicality of this model in improving AI safety and transparency, we simulate a scenario in which the model enables an agent to present plans to a human before execution, and to revise plans based on their language feedback.

cs.CL↗

arXiv2023

Appearance Codes using Joint Embedding Learning of Multiple Modalities

The use of appearance codes in recent work on generative modeling has enabled novel view renders with variable appearance and illumination, such as day-time and night-time renders of a scene. A major limitation of this technique is the need to re-train new appearance codes for every scene on inference, so in this work we address this problem proposing a framework that learns a joint embedding space for the appearance and structure of the scene by enforcing a contrastive loss constraint between different modalities. We apply our framework to a simple Variational Auto-Encoder model on the RADIATE dataset \cite{sheeny2021radiate} and qualitatively demonstrate that we can generate new renders of night-time photos using day-time appearance codes without additional optimization iterations. Additionally, we compare our model to a baseline VAE that uses the standard per-image appearance code technique and show that our approach achieves generations of similar quality without learning appearance codes for any unseen images on inference.

cs.CV↗

arXiv2023

Transaction Fee Mining and Mechanism Design

Transaction fees represent a major incentive in many blockchain systems as a way to incentivize processing transactions. Unfortunately, they also introduce an enormous amount of incentive asymmetry compared to alternatives like fixed block rewards. We analyze some of the incentive compatibility issues that arise from transaction fees, which relate to the bids that users submit, the allocation rules that miners use to choose which transactions to include, and where they choose to mine in the context of longest-chain consensus. We start by surveying a variety of mining attacks including undercutting, fee sniping, and fee-optimized selfish mining. Then, we move to analyzing mechanistic notions of user incentive compatibility, myopic miner incentive compatibility, and off-chain-agreement-proofness, as well as why they are provably incompatible in their full form. Then, we discuss weaker notions of nearly and $\gamma$-weak incentive compatibility, and how all of these forms of incentive compatibility hold or fail in the trustless auctioneer setup of blockchains, examining classical mechanisms as well as more recent ones such as Ethereum's EIP-1559 mechanism and \cite{chung}'s burning second-price auction. Throughout, we generalize and interrelate existing notions, provide new unifying perspectives and intuitions on analysis, and discuss both specific and overarching open problems for future work.

cs.GT↗