arXiv subjects

Chen Xie

Publications and source records attributed to Chen Xie.

At least 19 recordsLinked to original sources

arXiv2026

LD4WAM: Learning Latent Dynamics from Human Videos for World Action Models

Human video is playing an increasingly central role in training World Action Models (WAMs), owing to its diversity and low collection cost relative to teleoperated robot data. However, most WAMs learn from such video only by predicting pixel-level future frames, giving dynamics that are not directly actionable, whereas motion retargeting recovers directly actionable actions but leaves a large visual gap across embodiments. We therefore propose motion-aligned latent dynamics as an embodiment-agnostic representation to bridge video priors and low-level actions. We further present LD4WAM, which pairs a Latent Dynamics Model trained with semantic reconstruction and real motion alignment with a World Dynamics Action Model built as a mixture-of-transformers (MoT), which preserves full future-video generation and uses learnable queries to distill these latent dynamics from generated futures for action conditioning. Pretrained on our curated unified dataset of over 5{,}000 hours of human and robot data, LD4WAM performs strongly in RoboTwin simulation and on real robots equipped with both grippers and dexterous hands, while generalizing well to unseen objects and backgrounds.

cs.RO↗

arXiv2026

The Roman Coronagraph Community Participation Program: corgisim - a simulation suite for the Nancy Grace Roman Space Telescope Coronagraph Instrument

NASA's Roman Space Telescope will feature a pathfinder Coronagraph Instrument to demonstrate advanced high-contrast imaging from space, paving the way for future missions like the Habitable Worlds Observatory. The Coronagraph Instrument could obtain imaging, polarimetry and spectroscopy of Jupiter analogs in reflected visible light for the first time. We present the development of an open-source simulation package ``corgisim'' as part of the Roman Coronagraph Community Participate Program. Built on established optical propagation libraries including PROPER and CGISim, corgisim provides a user-friendly, publicly available Python framework for end-to-end simulations of the Coronagraph Instrument observations. The package produces high-fidelity, format-compliant data for pre-launch calibration, pipeline testing, and community applications such as target selection and observation planning. We will give an overview of corgisim's infrastructure, functionalities, and current implementation across planned imaging, polarimetry, and spectroscopy modes, including the ability to simulate host stars, injected companions, and extended disks. We will also highlight suitable applications of corgisim and provide guidance on how users can access and employ the software.

astro-ph.IM↗

arXiv2026

The Roman Coronagraph Community Participation Program: early calibration plan and pilot observation of a companion

Roman is set to launch in weeks! The Coronagraph Instrument - technology pathfinder for future direct imaging missions - is ready to fly too. According to predictions, laboratory tests and high fidelity simulations, it will open a new contrast regime enabling the imaging of mature, giant planets in visible reflected light. The Community Participation Program is responsible for preparing a comprehensive observing program with associated data processing software and calibrations. We give a brief update about the on-going "baseline" calibration plan for the first months. Additionally, we describe a pilot program aiming for the stellar companion HD 29992 B at moderate ~1e-5 to ~1e-6 Band 1 (575 nm) contrast, to be carried out as soon as the instrument is operational. The idea is to generate a canonical data set with a self luminous companion that is easily recoverable. This functional checkout will be precious to best prepare our community, exercise our calibration plan and suite of tools.

astro-ph.IM↗

arXiv2026

The Roman Coronagraph Community Participation Program: trials and triumphs of designing an observing program for a technology demonstration instrument

The Coronagraph Instrument onboard the Nancy Grace Roman Space Telescope serves as a crucial technology pathfinder for the Habitable Worlds Observatory, with on-sky verification of high-contrast imaging techniques and the potential to image a Jupiter analog in reflected light for the first time. Together with the Roman Project Team, the Community Participation Program (CPP) is responsible for target selection, preparatory observations, developing an exposure time calculator, target database, data reduction pipeline, simulation tools, and engagement with the broader community. Here we present an overview of the CPP activities over the past two years with an emphasis on observation planning activities for the initial in-orbit checkout and the first six months of the observation phase. Finally, we present future opportunities for the astronomical community to interact with the data as it becomes public early in the mission.

astro-ph.IM↗

arXiv2026

The Roman Coronagraph Community Participation Program: Observation planning and data reduction for polarimetric mode

Reflected-light polarimetry of exoplanets constrains and resolves degeneracies in atmospheric properties, while polarized light observations of debris disks enable the characterization of dust-grain properties. The best-effort polarimetric mode of the Roman Coronagraph Instrument will be able to perform multi-wavelength observations of planetary systems using both the Hybrid Lyot Coronagraph (HLC) and the Shaped Pupil Coronagraph (SPC). This paper presents an overview of observation planning, simulations, and data reduction procedures for the polarimetric mode of the Roman Coronagraph. As an initial test of simulation and data reduction, a dataset of polarimetric observing sequences for the debris disk HD 172555 in HLC mode was generated using corgisim with estimated observation parameters, and data reduction was performed using corgidrp, incorporating all relevant noise factors and calibration products. Currently, mock calibration products are used in corgidrp; these will be replaced with simulated calibration products in future updates

astro-ph.IM↗

arXiv2026

SSPO: Structure-Aware Similarity-Weighted Preference Optimization for Neural Combinatorial Optimization

Neural combinatorial optimization (NCO) relies on parallel solution sampling for training, yet existing methods fail to fully exploit the rich information latent in a co-sampled solution group. Preference-optimization methods anchor on the single best solution and discard fine-grained quality and structural signal from all other peers-a failure we term gradient signal polarization. Mean-based baselines instead weight peers uniformly, so structurally near-identical peers flood the baseline with redundant information and keep gradient variance high-a failure we term baseline redundancy. We propose SSPO (Structure-Aware Similarity-Weighted Preference Optimization), which scores all $B$ sampled solutions jointly through a dissimilarity-weighted leave-one-out baseline: structurally distinct peers receive higher weight, resolving both failures in a single mechanism. The baseline uses zero-parameter, problem-adaptive solution embeddings built from the encoder's existing node representations. Experiments on TSP, EFL, and JSP benchmarks show consistent gains over prior best-anchor and uniform-weight baselines. A direct comparison against uniform RLOO on TSP and EFL confirms that structure-aware weighting is the primary driver of improvement. The SSPO-trained EFL policy has been deployed in a production facility-location system at JD$\mathord{.}$com, confirming practical viability at scale.

stat.ML↗

arXiv2026

Standalone LLM and a Pre-specified Agentic Pipeline for Explaining ICU Mortality Predictions: a Feasibility Study on the eICU Demo Dataset

Machine-learning models can predict ICU mortality accurately, but feature-attribution methods alone rarely provide the clinical narrative needed for bedside use. Large language models (LLMs) may bridge this gap, and multi-step agentic pipelines are a plausible extension because they separate data interpretation, guideline checking, and final explanation. This revised feasibility study preserves the original standalone-versus-agentic comparison while making the main clinical findings more explicit. Using the retained local eICU Demo artifact set (2,353 ICU stays; 8.1\% mortality), XGBoost achieved an AUROC of 0.855 (95\% CI 0.796--0.906) and an AUPRC of 0.332 (95\% CI 0.217--0.494). On a stratified 38-case explanation subset, the standalone LLM produced 1 explanation with explicit outcome leakage, whereas the four-step agentic pipeline produced none. Among the 14 cases that overlapped with the SHAP review subset, the standalone LLM showed higher SHAP alignment (mean Jaccard 0.171 versus 0.077) and higher direction consistency (92.9\% versus 78.6\%), while the agentic pipeline showed higher guideline grounding (0.762 versus 0.143), higher value specificity (0.236 versus 0.143), and slightly higher plausibility (0.700 versus 0.671). Clinically, the results suggest that agentic decomposition may improve safety-relevant grounding and patient-specific detail, but it should be paired with attribution-based checks before use in high-stakes risk explanation.

cs.AI↗

arXiv2026

RoboWM-Bench: A Benchmark for Evaluating World Models in Robotic Manipulation

Recent advances in large-scale video world models have enabled increasingly realistic future prediction, raising the prospect of using generated videos as scalable supervision for robot learning. However, for embodied manipulation, perceptual realism alone is not sufficient: generated interactions must also be physically consistent and executable by robotic agents. Existing benchmarks provide valuable assessments of visual quality and physical plausibility, but they do not systematically evaluate whether predicted behaviors can be translated into executable actions that complete manipulation tasks. We introduce RoboWM-Bench, a manipulation-centric benchmark for embodiment-grounded evaluation of video world models. RoboWM-Bench converts generated human-hand and robotic manipulation videos into embodied action sequences and validates them through execution in physically grounded simulation environments. Built on real-to-sim scene reconstruction and diverse manipulation tasks, RoboWM-Bench enables standardized, reproducible, and scalable evaluation of physical executability. Using RoboWM-Bench, we evaluate state-of-the-art video world models and observe that visual plausibility and embodied executability are not always aligned. Our analysis highlights several recurring factors that affect execution performance, including spatial reasoning, contact prediction, and non-physical geometric distortions, particularly in complex and long-horizon interactions. These findings provide a more fine-grained view of current model capabilities and underscore the value of embodiment-aware evaluation for guiding physically grounded world modeling in robotic manipulation.

cs.RO↗

arXiv2026

From Seeing to Simulating: Generative High-Fidelity Simulation with Digital Cousins for Generalizable Robot Learning and Evaluation

Learning robust robot policies in real-world environments requires diverse data augmentation, yet scaling real-world data collection is costly due to the need for acquiring physical assets and reconfiguring environments. Therefore, augmenting real-world scenes into simulation has become a practical augmentation for efficient learning and evaluation. We present a generative framework that establishes a generative real-to-sim mapping from real-world panoramas to high-fidelity simulation scenes, and further synthesize diverse cousin scenes via semantic and geometric editing. Combined with high-quality physics engines and realistic assets, the generated scenes support interactive manipulation tasks. Additionally, we incorporate multi-room stitching to construct consistent large-scale environments for long-horizon navigation across complex layouts. Experiments demonstrate a strong sim-to-real correlation validating our platform's fidelity, and show that extensively scaling up data generation leads to significantly better generalization to unseen scene and object variations, demonstrating the effectiveness of Digital Cousins for generalizable robot learning and evaluation.

cs.RO↗

arXiv2026

Rethinking Code Complexity Through the Lens of Large Language Models

Code complexity metrics such as cyclomatic complexity have long been used to assess software quality and maintainability. With the rapid advancement of large language models (LLMs) on coding tasks, an important yet underexplored question arises: do traditional complexity metrics meaningfully characterize the coding difficulty that LLMs perceive? In this work, we empirically demonstrate that classical complexity metrics exhibit no consistent correlation with LLM performance, revealing a fundamental mismatch with model-perceived difficulty. To address this gap, we propose LM-CC, a novel code complexity metric tailored for LLMs, grounded in the hypothesis that model-perceived code difficulty is fundamentally driven by semantic nonlinearity. LM-CC quantifies complexity through an entropy-guided semantic compositional hierarchy, capturing the cumulative uncertainty encountered by LLMs during code understanding. Our experimental results demonstrate that LM-CC exhibits strong and consistent partial correlations with LLM performance, while semantics-preserving reductions in LM-CC consistently lead to improved downstream task performance. The source code is available at: https://github.com/xchen121/lm-cc.

cs.SE↗

arXiv2026

T2M Mamba: Motion Periodicity-Saliency Coupling Approach for Stable Text-Driven Motion Generation

Text-to-motion generation, which converts motion language descriptions into coherent 3D human motion sequences, has attracted increasing attention in fields, such as avatar animation and humanoid robotic interaction. Though existing models have achieved significant fidelity, they still suffer from two core limitations: (i) They treat motion periodicity and keyframe saliency as independent factors, overlooking their coupling and causing generation drift in long sequences. (ii) They are fragile to semantically equivalent paraphrases, where minor synonym substitutions distort textual embeddings, propagating through the decoder and producing unstable or erroneous motions. In this work, we propose T2M Mamba to address these limitations by (i) proposing Periodicity-Saliency Aware Mamba, which utilizes novel algorithms for keyframe weight estimation via enhanced Density Peaks Clustering and motion periodicity estimation via FFT-accelerated autocorrelation to capture coupled dynamics with minimal computational overhead, and (ii) constructing a Periodic Differential Cross-modal Alignment Module (PDCAM) to enhance robust alignment of textual and motion embeddings. Extensive experiments on HumanML3D and KIT-ML datasets have been conducted, confirming the effectiveness of our approach, achieving an FID of 0.068 and consistent gains on all other metrics.

cs.CV↗

arXiv2025

Evolutionary Physics-Informed Temporal Fusion for Lane-Change Intention Prediction

Early lane-change intention prediction is essential for autonomous driving and ADAS, but it remains challenging because lane-changing behavior depends on evolving traffic risk, surrounding-vehicle interactions, and target-lane feasibility rather than only instantaneous vehicle states. This study proposes an evolutionary physics-informed temporal fusion framework for three-class lane-change intention prediction, including left lane change, right lane change, and no lane change. Instead of using static physics-informed variables alone, the proposed method derives temporal descriptors from conventional traffic signals, including risk evolution, gap persistence, counterfactual lane utility, interaction pressure gradient, maneuver feasibility, and intent consistency. These descriptors are fused with temporal embeddings learned from raw trajectory sequences through a sequence encoder, and the fused representation is used for final classification. Experiments are conducted on the highD and exiD datasets under 1\,s, 2\,s, and 3\,s prediction horizons. The proposed model achieves Macro F1-scores of 0.9514, 0.9256, and 0.8872 on highD, and 0.9386, 0.9070, and 0.8531 on exiD, respectively. The improvement is especially pronounced in exiD ramp-adjacent scenarios, indicating that temporal physical evolution is particularly useful in interaction-rich environments. These results demonstrate that combining evolutionary physics-informed descriptors with learned temporal representations provides a more dynamic and interpretable solution for early lane-change intention prediction.

cs.LG↗

arXiv2025

JWST/NIRSpec Detects Warm CO Emission in the Terrestrial-Planet Zone of HD 131488

We have obtained a high-resolution, JWST NIRSpec $2.87$ -- $5.14$ $\mu$m spectrum of the debris disk around HD 131488. We discover CO fundamental emission indicating the presence of warm fluorescent gas within $\sim10$ AU of the star. The large discrepancy in CO's vibrational and rotational temperature indicates that CO is out of thermal equilibrium and is excited with UV fluorescence. Our UV fluorescence model gives a best fit of $1150\,$K with an effective temperature of $450$, $332$, and $125\,$K for the warm CO gas kinetic temperature within $0.5$, $1$, and $10\,$AU to the star and a gas vibrational temperature of $8800\,$K. The newly discovered warm CO gas population likely resides between sub-AU scales and $\sim\,10\,$AU, interior to the cold CO reservoir detected beyond $35\,$AU with HST STIS and ALMA. The discovery of warm, fluorescent gas in a debris disk is the first such detection ever made. The detection of warm CO raises the possibility of unseen molecules (H$_2$O, H$_2$, etc) as collisional partners to excite the warm gas. We estimated a lower mass limit for CO of $1.25\times 10^{-7}\text{M}_{\oplus}$, which is $10^{-5}$ of the cold CO mass detected with ALMA and HST. We demonstrate that UV fluorescence emerges as a promising avenue for detecting tenuous gas at $10^{-7}$ Earth-mass level in debris disks with JWST.

astro-ph.EP↗

arXiv2025

RaX-Crash: A Resource Efficient and Explainable Small Model Pipeline with an Application to City Scale Injury Severity Prediction

New York City reports over one hundred thousand motor vehicle collisions each year, creating substantial injury and public health burden. We present RaX-Crash, a resource efficient and explainable small model pipeline for structured injury severity prediction on the official NYC Motor Vehicle Collisions dataset. RaX-Crash integrates three linked tables with tens of millions of records, builds a unified feature schema in partitioned storage, and trains compact tree based ensembles (Random Forest and XGBoost) on engineered tabular features, which are compared against locally deployed small language models (SLMs) prompted with textual summaries. On a temporally held out test set, XGBoost and Random Forest achieve accuracies of 0.7828 and 0.7794, clearly outperforming SLMs (0.594 and 0.496); class imbalance analysis shows that simple class weighting improves fatal recall with modest accuracy trade offs, and SHAP attribution highlights human vulnerability factors, timing, and location as dominant drivers of predicted severity. Overall, RaX-Crash indicates that interpretable small model ensembles remain strong baselines for city scale injury analytics, while hybrid pipelines that pair tabular predictors with SLM generated narratives improve communication without sacrificing scalability.

cs.LG↗

arXiv2025

BEAM: Brainwave Empathy Assessment Model for Early Childhood

Empathy in young children is crucial for their social and emotional development, yet predicting it remains challenging. Traditional methods often only rely on self-reports or observer-based labeling, which are susceptible to bias and fail to objectively capture the process of empathy formation. EEG offers an objective alternative; however, current approaches primarily extract static patterns, neglecting temporal dynamics. To overcome these limitations, we propose a novel deep learning framework, the Brainwave Empathy Assessment Model (BEAM), to predict empathy levels in children aged 4-6 years. BEAM leverages multi-view EEG signals to capture both cognitive and emotional dimensions of empathy. The framework comprises three key components: 1) a LaBraM-based encoder for effective spatio-temporal feature extraction, 2) a feature fusion module to integrate complementary information from multi-view signals, and 3) a contrastive learning module to enhance class separation. Validated on the CBCP dataset, BEAM outperforms state-of-the-art methods across multiple metrics, demonstrating its potential for objective empathy assessment and providing a preliminary insight into early interventions in children's prosocial development.

cs.LG↗

arXiv2025

EvoC2Rust: A Skeleton-guided Framework for Project-Level C-to-Rust Translation

Translating legacy C codebases to Rust is increasingly demanded for building safety-critical systems. While various approaches have emerged for this task, they face inherent trade-offs: rule-based methods often struggle to satisfy code safety and idiomaticity requirements, while LLM-based methods frequently fail to generate semantically equivalent Rust code, due to the heavy dependencies of modules across the entire codebase. Recent studies have revealed that both solutions are limited to small-scale programs. In this paper, we propose EvoC2Rust, an automated framework for converting complete C projects to equivalent Rust ones. EvoC2Rust employs a skeleton-guided translation strategy for project-level translation. The pipeline consists of three stages: 1) it first decomposes the C project into functional modules, employs a feature-mapping-enhanced LLM to transform definitions and macros, and generates type-checked function stubs, which form a compilable Rust skeleton; 2) it then incrementally translates functions, replacing the corresponding stub placeholders; 3) finally, it repairs compilation errors by integrating LLM and static analysis. Through evolutionary augmentation, EvoC2Rust combines the advantages of both rule-based and LLM-based solutions. Our evaluation on open-source benchmarks and six industrial projects demonstrates the superior performance of EvoC2Rust in project-level C-to-Rust translation. The results show that our approach outperforms the strongest LLM-based baseline by 17.24% in syntax accuracy and 14.32% in semantic accuracy, while also achieving a 43.59% higher code safety rate than the best rule-based tool.

cs.SE↗

arXiv2025

Water ice in the debris disk around HD 181327

Debris disks are exoplanetary systems that contain planets, minor bodies (i.e., asteroids, Kuiper belt objects, comets, etc.), and micron-sized debris dust. Since water ice is the most common frozen volatile, it plays an essential role in the formation of planets and minor bodies. Although water ice has been commonly found in Kuiper belt objects and comets in the Solar System, no definitive evidence for water ice in debris disks has been obtained to date. Here, we report the discovery of water ice in the HD 181327 disk using the James Webb Space Telescope Near-Infrared Spectrograph. We detect the solid-state broad absorption feature of water ice at 3 $\mu$m and a distinct Fresnel peak feature at 3.1 $\mu$m, a characteristic of large water-ice particles. This implies the presence of a water-ice reservoir in the HD 181327 exoKuiper belt. Gradients of water-ice features at different stellocentric distances reveal a dynamic process of destroying and replenishing water ice in the disk, with estimated water-ice mass fractions ranging from 0.1% at ~85 au to 14% at ~113 au. It is highly plausible that the icy bodies that release water ice in HD 181327 could be the extra-solar counterparts of some of the Kuiper belt objects in our Solar System, supported by their spectral similarity.

astro-ph.EP↗

arXiv2025

EdgePoint2: Compact Descriptors for Superior Efficiency and Accuracy

The field of keypoint extraction, which is essential for vision applications like Structure from Motion (SfM) and Simultaneous Localization and Mapping (SLAM), has evolved from relying on handcrafted methods to leveraging deep learning techniques. While deep learning approaches have significantly improved performance, they often incur substantial computational costs, limiting their deployment in real-time edge applications. Efforts to create lightweight neural networks have seen some success, yet they often result in trade-offs between efficiency and accuracy. Additionally, the high-dimensional descriptors generated by these networks poses challenges for distributed applications requiring efficient communication and coordination, highlighting the need for compact yet competitively accurate descriptors. In this paper, we present EdgePoint2, a series of lightweight keypoint detection and description neural networks specifically tailored for edge computing applications on embedded system. The network architecture is optimized for efficiency without sacrificing accuracy. To train compact descriptors, we introduce a combination of Orthogonal Procrustes loss and similarity loss, which can serve as a general approach for hypersphere embedding distillation tasks. Additionally, we offer 14 sub-models to satisfy diverse application requirements. Our experiments demonstrate that EdgePoint2 consistently achieves state-of-the-art (SOTA) accuracy and efficiency across various challenging scenarios while employing lower-dimensional descriptors (32/48/64). Beyond its accuracy, EdgePoint2 offers significant advantages in flexibility, robustness, and versatility. Consequently, EdgePoint2 emerges as a highly competitive option for visual tasks, especially in contexts demanding adaptability to diverse computational and communication constraints.

cs.CV↗