arXiv subjects

Shan Zhao

Publications and source records attributed to Shan Zhao.

At least 19 recordsLinked to original sources

arXiv2026

ECHO: A Cognitively Inspired, Auditable Memory Plane for Long-Horizon Agents

Long-horizon agents need memory that identifies relevant experience, resolves revisions, and exposes checkable provenance. We present ECHO (Embodied Context and History Orchestration), an auditable memory architecture and service prototype inspired by episodic encoding, consolidation, contextual reinstatement, reconsolidation, and executive control. This is functional inspiration, not neural equivalence; the empirical analysis focuses on retrieval and context construction. Development runs reach 96.29% Hit@10 and 73.64% turn Recall@5 on 1,536 LoCoMo category 1-4 questions, and 97.60% Hit@10, 88.84% turn Recall@5, and 88.71% session Recall@5 on all 500 LongMemEval-S questions. A five-history BEAM gate fails, and in a separate matched 91-question QA sample Mem0 OSS scores 64.84% versus ECHO's 41.76% (exact McNemar p = 0.00107), with a history-cluster interval crossing zero. A post-hoc audit found source-specific phrases in the query-expansion rules. Although no gold answer field entered the runtime, expansion-enabled retrieval scores are therefore descriptive development measurements, not independent confirmation.

cs.AI↗

arXiv2026

On calculating polar solvation energy of nonrigid proteins in the Poisson-Boltzmann theory

The Poisson-Boltzmann (PB) theory is a cornerstone of implicit solvent models for electrostatic analysis, and has found a great success in various biomolecular applications. However, in calculating polar solvation energy, one should consider that the structure of the protein changes upon transition from vacuum to water phases. To address this, here we report for the first time a generalized PB framework capable of accommodating nonrigid conformational changes without suffering from self-energy artifacts. For regularized PB models, in which the charge singularities are captured by the Green's functions, self-energies in the water and vacuum states will be analytically canceled. For non-regularized PB solvers, such as APBS and DelPhi, a simple thermodynamic cycle is proposed for nonrigid proteins by adding a Coulombic correction in vacuum. The generalized PB theory is validated using a perturbed two-atom system and a diverse set of proteins with different structures in vacuum and water, demonstrating its accuracy and robustness, regardless of the choice of sharp-interface and diffuse-interface PB models and different numerical solvers.

math.NA↗

arXiv2026

What to Remove, What to Preserve: Dual-Ambiguity Rectification for All-in-One Image Restoration

All-in-one image restoration aims to handle diverse degradations within a unified framework. Existing methods commonly encode heterogeneous degradation conditions in a shared latent space, where degradation-related cues and scene content can remain entangled. We characterize the resulting challenge as dual ambiguity: semantic ambiguity in channel-wise modulation and spatial ambiguity in restoration responses, which can lead to content corruption and residual artifacts. To mitigate this issue, we propose DAR-Net, a Dual-Ambiguity Rectification Network for all-in-one image restoration. DAR-Net first introduces a Degradation Archetype Representation (DAR) module to construct a structured degradation state through simplex-constrained archetype mixture modeling. Based on this state, a Semantic Ambiguity Rectification (SeAR) module generates degradation-aware prompts to improve channel-wise conditioning in the decoder. A Spatial Ambiguity Rectification (SpAR) module further regularizes degradation-aware and complementary features toward orthogonal response subspaces, reducing spatial interference between removal and preservation cues. Extensive experiments on standard all-in-one restoration benchmarks show that DAR-Net achieves the best overall performance under both three-degradation and five-degradation settings, improving the average PSNR over the strongest competitor by 0.14 dB and 0.34 dB, respectively; it additionally shows superior performance on CDD-11 and WeatherBench.

cs.CV↗

arXiv2026

LA-RL: Label-Aware Self-Reflection for Reinforcement Learning in Information Extraction

Large language models show strong promise for information extraction (IE), but existing reflection-based correction methods are often misaligned with structured extraction outputs. Free-form self-reflection can flag an error, yet it rarely identifies whether the failure is a missing span, wrong label, boundary mismatch, invalid relation type, or reversed argument order. We introduce LA-RL (Label-Aware Reflective Reinforcement Learning), an outcome-supervised framework that guides IE self-correction with task-grounded diagnostic labels. A single backbone first predicts an extraction, diagnoses task-specific error labels, and then revises its output conditioned on the diagnosis. Training starts from diagnostic data labeled by an annotation model for cold-start supervised fine-tuning and proceeds through two GRPO stages that reward final extraction quality, format validity, and first-pass correctness, without a process reward model. Experiments on named entity recognition, relation extraction, and event extraction show consistent same-backbone gains over SFT, including 6.83 average F1 on SciER relation extraction, about 20 F1 on out-of-distribution relation extraction, and 14.80 trigger F1 plus 17.50 argument F1 on DuEE1.0. Ablations show that reflection structure is task-sensitive: stronger constraints benefit relation extraction, whereas named entity recognition needs less restrictive correction under domain shift.

cs.CL↗

arXiv2026

LC-ICL: Label-Guided Contrastive In-Context Learning for Robust Information Extraction

There has been increasing interest in exploring the capabilities of advanced large language models (LLMs) in the field of information extraction (IE), specifically focusing on tasks related to named entity recognition (NER) and relation extraction (RE).Although researchers are exploring the use of few-shot information extraction through in-context learning with LLMs, they tend to focus only on using correct or positive examples for demonstration, neglecting the potential value of incorporating incorrect or negative examples into the learning process.In this paper, we present LC-ICL a novel few-shot technique that leverages both correct and incorrect sample constructions to create in-context learning demonstrations. This approach enhances the ability of LLMs to extract entities and relations by combining positive samples with negative samples annotated by error-cause labels. These labels expose more detailed error features in erroneous examples, enabling the model to understand why similar predictions fail and avoid repeating such errors during inference.Specifically, our proposed method taps into the inherent contextual information and valuable information in hard negative samples and the nearest positive neighbors to the test and then applies the in-context learning demonstrations based on LLMs. Our experiments on various datasets indicate that LC-ICL outperforms previous few-shot in-context learning methods, delivering substantial enhancements in performance across a broad spectrum of related tasks. These improvements are noteworthy, showcasing the versatility of our approach in diverse scenarios.

cs.CL↗

arXiv2026

Xiaomi Auto World Model: A Joint World Model Integrating Reconstruction and Generation for Autonomous Driving

This report presents a unified technical system addressing the two core capabilities of world models for autonomous driving: world representation and world generation. For world representation, we propose WorldRec, a feed-forward reconstruction architecture driven by sparse scene queries. WorldRec initializes structured queries in 3D space, leveraging them to aggregate cross-view, cross-temporal features, thereby naturally enforcing spatial consistency across frames and yielding compact yet high-fidelity 3D Gaussian scene representations. For world generation, we propose WorldGen, a two-stage training framework of bidirectional pretraining followed by causal fine-tuning through three progressive stages (Teacher Forcing, ODE distillation, and DMD), enabling high-quality online causal video generation in as few as 4 denoising steps. Building on both modules, we further introduce the JWM, which deeply integrates WorldRec and WorldGen to achieve synergistic gains in generation stability, cross-frame consistency, and visual fidelity, providing a solid foundation for closed-loop simulation, data synthesis, and end-to-end training in autonomous driving.

cs.CV↗

arXiv2026

Where Does Vision Meet Language? Understanding and Refining Visual Fusion in MLLMs via Contrastive Attention

Multimodal Large Language Models (MLLMs) have achieved remarkable progress in vision-language understanding, yet how they internally integrate visual and textual information remains poorly understood. To bridge this gap, we perform a systematic layer-wise masking analysis across multiple architectures, revealing how visual-text fusion evolves within MLLMs. The results show that fusion emerges at several specific layers rather than being uniformly distributed across the network, and certain models exhibit a late-stage "review" phenomenon where visual signals are reactivated before output generation. Besides, we further analyze layer-wise attention evolution and observe persistent high-attention noise on irrelevant regions, along with gradually increasing attention on text-aligned areas. Guided by these insights, we introduce a training-free contrastive attention framework that models the transformation between early fusion and final layers to highlight meaningful attention shifts. Extensive experiments across various MLLMs and benchmarks validate our analysis and demonstrate that the proposed approach improves multimodal reasoning performance.

cs.CV↗

arXiv2026

ParkGaussian: Surround-view 3D Gaussian Splatting for Autonomous Parking

Parking is a critical task for autonomous driving systems (ADS), with unique challenges in crowded parking slots and GPS-denied environments. However, existing works focus on 2D parking slot perception, mapping, and localization, 3D reconstruction remains underexplored, which is crucial for capturing complex spatial geometry in parking scenarios. Naively improving the visual quality of reconstructed parking scenes does not directly benefit autonomous parking, as the key entry point for parking is the slots perception module. To address these limitations, we curate the first benchmark named ParkRecon3D, specifically designed for parking scene reconstruction. It includes sensor data from four surround-view fisheye cameras with calibrated extrinsics and dense parking slot annotations. We then propose ParkGaussian, the first framework that integrates 3D Gaussian Splatting (3DGS) for parking scene reconstruction. To further improve the alignment between reconstruction and downstream parking slot detection, we introduce a slot-aware reconstruction strategy that leverages existing parking perception methods to enhance the synthesis quality of slot regions. Experiments on ParkRecon3D demonstrate that ParkGaussian achieves state-of-the-art reconstruction quality and better preserves perception consistency for downstream tasks. The code and dataset will be released at: https://github.com/wm-research/ParkGaussian

cs.CV↗

arXiv2025

A regularized Matched Interface and Boundary Method (MIB) for Solving Polarizable Multipole Poisson-Boltzmann model

To accurately model the electron density and polarization, a polarizable multipole (PM) model using the AMOEBA force field has been introduced \cite{Ren:2003, Shi:2013} recently. In the AMOEBA force field, the traditional point atomic representation is updated with permanent multipoles including additional dipoles and quadrupoles at atom centers in terms of derivatives of delta functions. Meanwhile, the polarization of the solute is considered by the introduction of induced dipoles. The AMOEBA forcefield thus shows significantly better agreement with experimental and high-level {\it ab initio} results. Moreover, the AMOEBA force field keeps the simple atomic structure, so that it can conviniently replace the traditional partial charge model. In this paper, we address the numerical challenges associated with the Polarizable Multipole Poisson--Boltzamnnn (PM-PB) model, which couples the AMOEBA force field with a linear Poisson-Boltzmann equation for implicit solvent and polarization modeling. To solve the PM-PB model, we designed a regularized Matched Interface and Boundary (MIB) method to analytically regularizes the singular source term in the PMPB model while maintains 2nd order accuracy by rigorously treating the interface conditions. The accuracy of the method is validated on Kirkwood sphere with available analytical solutions and on proteins whose charge distribution are assigned using AMOEBA force field.

math-ph↗

arXiv2025

The Cost of Simplicity: How Reducing EEG Electrodes Affects Source Localization and BCI Accuracy

Electrode density optimization in electroencephalography (EEG)-based Brain-Computer Interfaces (BCIs) requires balancing practical usability against signal fidelity, particularly for source localization. Reducing electrodes enhances portability but its effects on neural source reconstruction quality and source connectivity - treated as proxies to BCI performance - remain understudied. We address this gap through systematic evaluation of 62-, 32-, and 16-channel configurations using a fixed, fully automated processing pipeline applied to the well-characterized P300 potential. This approach's rationale is to minimize variability and bias inherent to EEG analysis by leveraging the P300's stimulus-locked reproducibility and pipeline standardization. Analyzing 63 sessions (31 subjects) from the Eye-BCI dataset with rigorous artifact correction and channel validation, we demonstrate: (1) Progressive degradation in source reconstruction quality with sparser configurations, including obscured deep neural generators and spatiotemporal distortions; (2) A novel sqrt(Re) scaling law linking electrode reduction ratio (Re) to localization accuracy - a previously unquantified relationship to the best of our knowledge; (3) While reduced configurations preserve basic P300 topography and may suffice for communicative BCIs, higher-density channels are essential for reliable deep source reconstruction. Overall, this study establishes a first step towards quantitative benchmarks for electrode selection, with critical implications for clinical BCIs requiring anatomical precision in applications like neurodegenerative disease monitoring, where compromised spatial resolution could mask pathological signatures. Most importantly, the sqrt(Re) scaling law may provide the first principled method to determine the minimal electrode density required based on acceptable error margins or expected effect sizes.

q-bio.NC↗

arXiv2025

From Noise to Insight: Visualizing Neural Dynamics with Segmented SNR Topographies for Improved EEG-BCI Performance

Electroencephalography (EEG)-based wearable brain-computer interfaces (BCIs) face challenges due to low signal-to-noise ratio (SNR) and non-stationary neural activity. We introduce in this manuscript a mathematically rigorous framework that combines data-driven noise interval evaluation with advanced SNR visualization to address these limitations. Analysis of the publicly available Eye-BCI multimodal dataset demonstrates the method's ability to recover canonical P300 characteristics across frequency bands (delta: 0.5-4 Hz, theta: 4-7.5 Hz, broadband: 1-15 Hz), with precise spatiotemporal localization of both P3a (frontocentral) and P3b (parietal) subcomponents. To the best of our knowledge, this is the first study to systematically assess the impact of noise interval selection on EEG signal quality. Cross-session correlations for four different choices of noise intervals spanning from early to late pre-stimulus phases also indicate that alertness and task engagement states modulate noise interval sensitivity, suggesting broader applications for adaptive BCI systems. While validated in healthy participants, our results represent a first step towards providing clinicians with an interpretable tool for detecting neurophysiological abnormalities and provides quantifiable metrics for system optimization.

q-bio.NC↗

arXiv2025

ExtraGS: Geometric-Aware Trajectory Extrapolation with Uncertainty-Guided Generative Priors

Synthesizing extrapolated views from recorded driving logs is critical for simulating driving scenes for autonomous driving vehicles, yet it remains a challenging task. Recent methods leverage generative priors as pseudo ground truth, but often lead to poor geometric consistency and over-smoothed renderings. To address these limitations, we propose ExtraGS, a holistic framework for trajectory extrapolation that integrates both geometric and generative priors. At the core of ExtraGS is a novel Road Surface Gaussian(RSG) representation based on a hybrid Gaussian-Signed Distance Function (SDF) design, and Far Field Gaussians (FFG) that use learnable scaling factors to efficiently handle distant objects. Furthermore, we develop a self-supervised uncertainty estimation framework based on spherical harmonics that enables selective integration of generative priors only where extrapolation artifacts occur. Extensive experiments on multiple datasets, diverse multi-camera setups, and various generative priors demonstrate that ExtraGS significantly enhances the realism and geometric consistency of extrapolated views, while preserving high fidelity along the original trajectory.

cs.CV↗

arXiv2025

Automatic Blink-based Bad EEG channels Detection for BCI Applications

In Brain-Computer Interface (BCI) applications, noise presents a persistent challenge, often compromising the quality of EEG signals essential for accurate data interpretation. This paper focuses on optimizing the signal-to-noise ratio (SNR) to improve BCI performance, with channel selection being a key method for achieving this enhancement. The Eye-BCI multimodal dataset is used to address the issue of detecting and eliminating faulty EEG channels caused by non-biological artifacts, such as malfunctioning electrodes and power line interference. The core of this research is the automatic detection of problematic channels through the Adaptive Blink-Correction and De-Drifting (ABCD) algorithm. This method utilizes blink propagation patterns to identify channels affected by artifacts or malfunctions. Additionally, segmented SNR topographies and source localization plots are employed to illustrate the impact of channel removal by comparing Left and Right hand grasp Motor Imagery (MI). Classification accuracy further supports the value of the ABCD algorithm, reaching an average classification accuracy of 93.81% [74.81%; 98.76%] (confidence interval at 95% confidence level) across 31 subjects (63 sessions), significantly surpassing traditional methods such as Independent Component Analysis (ICA) (79.29% [57.41%; 92.89%]) and Artifact Subspace Reconstruction (ASR) (84.05% [62.88%; 95.31%]). These results underscore the critical role of channel selection and the potential of using blink patterns for detecting bad EEG channels, offering valuable insights for improving real-time or offline BCI systems by reducing noise and enhancing signal quality.

q-bio.NC↗

arXiv2025

ExEBench: Benchmarking Foundation Models on Extreme Earth Events

Our planet is facing increasingly frequent extreme events, which pose major risks to human lives and ecosystems. Recent advances in machine learning (ML), especially with foundation models (FMs) trained on extensive datasets, excel in extracting features and show promise in disaster management. Nevertheless, these models often inherit biases from training data, challenging their performance over extreme values. To explore the reliability of FM in the context of extreme events, we introduce \textbf{ExE}Bench (\textbf{Ex}treme \textbf{E}arth Benchmark), a collection of seven extreme event categories across floods, wildfires, storms, tropical cyclones, extreme precipitation, heatwaves, and cold waves. The dataset features global coverage, varying data volumes, and diverse data sources with different spatial, temporal, and spectral characteristics. To broaden the real-world impact of FMs, we include multiple challenging ML tasks that are closely aligned with operational needs in extreme events detection, monitoring, and forecasting. ExEBench aims to (1) assess FM generalizability across diverse, high-impact tasks and domains, (2) promote the development of novel ML methods that benefit disaster management, and (3) offer a platform for analyzing the interactions and cascading effects of extreme events to advance our understanding of Earth system, especially under the climate change expected in the decades to come. The dataset and code are public https://github.com/zhaoshan2/EarthExtreme-Bench.

cs.LG↗

arXiv2024

MOSABench: Multi-Object Sentiment Analysis Benchmark for Evaluating Multimodal Large Language Models Understanding of Complex Image

Multimodal large language models (MLLMs) have shown remarkable progress in high-level semantic tasks such as visual question answering, image captioning, and emotion recognition. However, despite advancements, there remains a lack of standardized benchmarks for evaluating MLLMs performance in multi-object sentiment analysis, a key task in semantic understanding. To address this gap, we introduce MOSABench, a novel evaluation dataset designed specifically for multi-object sentiment analysis. MOSABench includes approximately 1,000 images with multiple objects, requiring MLLMs to independently assess the sentiment of each object, thereby reflecting real-world complexities. Key innovations in MOSABench include distance-based target annotation, post-processing for evaluation to standardize outputs, and an improved scoring mechanism. Our experiments reveal notable limitations in current MLLMs: while some models, like mPLUG-owl and Qwen-VL2, demonstrate effective attention to sentiment-relevant features, others exhibit scattered focus and performance declines, especially as the spatial distance between objects increases. This research underscores the need for MLLMs to enhance accuracy in complex, multi-object sentiment analysis tasks and establishes MOSABench as a foundational tool for advancing sentiment analysis capabilities in MLLMs.

cs.CV↗

arXiv2024

Beyond Grid Data: Exploring Graph Neural Networks for Earth Observation

Earth Observation (EO) data analysis has been significantly revolutionized by deep learning (DL), with applications typically limited to grid-like data structures. Graph Neural Networks (GNNs) emerge as an important innovation, propelling DL into the non-Euclidean domain. Naturally, GNNs can effectively tackle the challenges posed by diverse modalities, multiple sensors, and the heterogeneous nature of EO data. To introduce GNNs in the related domains, our review begins by offering fundamental knowledge on GNNs. Then, we summarize the generic problems in EO, to which GNNs can offer potential solutions. Following this, we explore a broad spectrum of GNNs' applications to scientific problems in Earth systems, covering areas such as weather and climate analysis, disaster management, air quality monitoring, agriculture, land cover classification, hydrological process modeling, and urban modeling. The rationale behind adopting GNNs in these fields is explained, alongside methodologies for organizing graphs and designing favorable architectures for various tasks. Furthermore, we highlight methodological challenges of implementing GNNs in these domains and possible solutions that could guide future research. While acknowledging that GNNs are not a universal solution, we conclude the paper by comparing them with other popular architectures like transformers and analyzing their potential synergies.

cs.LG↗

arXiv2024

Temporal and Spacial Studies of Infectious Diseases: Mathematical Models and Numerical Solvers

The SIR model is a classical model characterizing the spreading of infectious diseases. This model describes the time-dependent quantity changes among Susceptible, Infectious, and Recovered groups. By introducing space-depend effects such as diffusion and creation in addition to the SIR model, the Fisher's model is in fact a more advanced and comprehensive model. However, the Fisher's model is much less popular than the SIR model in simulating infectious disease numerically due to the difficulties from the parameter selection, the involvement of 2-d/3-d spacial effects, the configuration of the boundary conditions, etc. This paper aim to address these issues by providing numerical algorithms involving space and time finite difference schemes and iterative methods, and its open-source Python code for solving the Fisher's model. This 2-D Fisher's solver is second order in space and up to the second order in time, which is rigorously verified using test cases with analytical solutions. Numerical algorithms such as SOR, implicit Euler, Staggered Crank-Nicolson, and ADI are combined to improve the efficiency and accuracy of the solver. It can handle various boundary conditions subject to different physical descriptions. In addition, real-world data of Covid-19 are used by the model to demonstrate its practical usage in providing prediction and inferences.

q-bio.QM↗

arXiv2024

DIM: Dynamic Integration of Multimodal Entity Linking with Large Language Model

Our study delves into Multimodal Entity Linking, aligning the mention in multimodal information with entities in knowledge base. Existing methods are still facing challenges like ambiguous entity representations and limited image information utilization. Thus, we propose dynamic entity extraction using ChatGPT, which dynamically extracts entities and enhances datasets. We also propose a method: Dynamically Integrate Multimodal information with knowledge base (DIM), employing the capability of the Large Language Model (LLM) for visual understanding. The LLM, such as BLIP-2, extracts information relevant to entities in the image, which can facilitate improved extraction of entity features and linking them with the dynamic entity representations provided by ChatGPT. The experiments demonstrate that our proposed DIM method outperforms the majority of existing methods on the three original datasets, and achieves state-of-the-art (SOTA) on the dynamically enhanced datasets (Wiki+, Rich+, Diverse+). For reproducibility, our code and collected datasets are released on \url{https://github.com/season1blue/DIM}.

cs.CL↗