Zhenyu Yu

Zhenyu Yu, Shuigeng Zhou

Sequential recommenders weight historical interactions either through positional self-attention as in Transformers or through a single implicit decay schedule as in State-Space Models. Neither makes the multi-scale temporal structure of real user behaviour explicit. We propose MARS, an encoder-agnostic aggregation operator that consumes real timestamps and produces K summaries emphasising distinct recency scales, fused by a context-adaptive gate. MARS adds at most 6% parameters and runs in $\mathcal{O}(LdK)$ time. MARS adapts to data density by automatically selecting between two encoder instantiations: MARS-T (Transformer) for sparse data and MARS-M (Mamba) for dense data, based on the average sequence length of the training set. On five public benchmarks against ten Transformer- and Mamba-based baselines under a unified RecBole protocol, MARS attains the best HR@10 on every benchmark, with mean relative gain +19.7% over the strongest content-only Transformer baseline on sparse data (reaching +36.2% on Games) and +3.2% HR@10 / +0.9% NDCG over SIGMA on dense ML-1M at 42% fewer MFLOPs, occupying the accuracy-efficiency Pareto frontier across the data-density spectrum. A backbone-only ablation isolates the marginal contribution of MARS at +4% to +19% HR@10 on sparse data and motivates the dual-instantiation design. The code is included in the supplementary material.

Zhenyu Yu, Jihong Guan, Shuigeng Zhou

A publisher who releases check-in trajectories inadvertently publishes a strong predictor of every user's future locations. We address this risk by generating unlearnable trajectories, perturbed sequences that yield victim models with degraded next-Point-of-Interest (next-POI) accuracy on clean test inputs. Direct ports of image-domain unlearnable examples fail on two counts. The published data must remain geographically and semantically plausible, and the perturbation must resist purification adversaries that exploit the structure of randomized defences. We propose Ghost, a manifold-aligned framework whose perturbations look like plausible human check-in sequences yet leave no learnable signal behind. Ghost steers each substitution onto the real-trajectory manifold through a frozen trajectory language model, so a denoising-bridge adversary has nothing to invert and a context-free frequency-table adversary recovers a near-uniform distribution. Across two standard benchmarks, and four attacker postures, Ghost achieves protection-gap competitive with the strongest deterministic baseline (PGD) while attaining the lowest restored accuracy under the bigram adaptive purification adversary on both datasets, and lies within one per-cell standard deviation of PGD on the protection-versus-purification-resistance plane. Ablations confirm the manifold prior subsumes the entropy-floor knob of prior randomized defences, with the frequency-table adversary's survival gap remaining within 0.04 even when twenty percent of the pairs are leaked.

Zhenyu Yu, Shuigeng Zhou

Latent reasoning has improved sequential recommendation by iteratively refining representations before prediction, but does it help spatial prediction? We find that the answer depends on whether reasoning is grounded in the underlying metric space. Without such grounding, latent reasoning degrades spatial prediction below the unmodified baseline, while a learned metric-space bias derived from pairwise distances produces consistent gains. We formalize this finding through MeRa (Metric-space Reasoning), a lightweight backbone-agnostic module that can be inserted between any sequence encoder and its prediction heads. On the GETNext backbone, the gap between reasoning without and with metric-space bias reaches 4.5% NDCG@10. MeRa achieves the best NDCG@10 on all three spatial prediction benchmarks among the compared methods, surpassing recent approaches such as GeoMamba and HMST. We prove that metric-space-constrained reasoning converges to a unique fixed point and that N-step reasoning is strictly more expressive than (N-1)-step reasoning. A controlled experiment on CLEVR with Euclidean distance confirms that the finding generalizes beyond geographic coordinates. The code is included in the supplementary material.

Zhenyu Yu, Shuigeng Zhou

Large language models reach 50 to 70% accuracy on causal reasoning benchmarks such as CLadder, but it is unclear whether this reflects structural reasoning or lexical pattern matching. We introduce Caliper, a controlled perturbation that replaces semantic variable names with placeholder tokens while preserving the causal graph and probabilistic specification of each question. Across nine instruction-tuned LLMs from 3.8B to 671B and three causal reasoning benchmarks, lexical anonymization yields robust accuracy drops of +7.6, +27.0, and +11.1 pp on a local 3.8B-14B set, rising to +29.6 and +18.0 pp on CRASS and e-CARE across nine frontier models spanning the 2024-2026 generations. Of 40 engaged model-by-benchmark cells, 39 show a positive gap, and the gap collapses by 17x on CLadder's pseudoword subset. Structured scaffolding and few-shot in-context learning each narrow the gap, but mainly by lowering P0 accuracy on smaller models rather than recovering P1. Current instruction-tuned LLMs, evaluated zero-shot, show little evidence of structural causal reasoning once lexical anchors are removed.

Yangchen Zeng, Zhenyu Yu, Dongming Jiang et al.

Transformer-based detectors have advanced small-object detection, but they often remain inefficient and vulnerable to background-induced query noise, which motivates deep decoders to refine low-quality queries. We present HELP (Heatmap-guided Embedding Learning Paradigm), a noise-aware positional-semantic fusion framework that studies where to embed positional information by selectively preserving positional encodings in foreground-salient regions while suppressing background clutter. Within HELP, we introduce Heatmap-guided Positional Embedding (HPE) as the core embedding mechanism and visualize it with a heatbar for interpretable diagnosis and fine-tuning. HPE is integrated into both the encoder and decoder: it guides noise-suppressed feature encoding by injecting heatmap-aware positional encoding, and it enables high-quality query retrieval by filtering background-dominant embeddings via a gradient-based mask filter before decoding. To address feature sparsity in complex small targets, we integrate Linear-Snake Convolution to enrich retrieval-relevant representations. The gradient-based heatmap supervision is used during training only, incurring no additional gradient computation at inference. As a result, our design reduces decoder layers from eight to three and achieves a 59.4% parameter reduction (66.3M vs. 163M) while maintaining consistent accuracy gains under a reduced compute budget across benchmarks. Code Repository: https://github.com/yidimopozhibai/Noise-Suppressed-Query-Retrieval

Zhenyu Yu, Chunlei Meng, Yangchen Zeng et al.

Next Point-of-Interest (POI) recommendation plays a crucial role in location-based services by predicting users' future mobility patterns. Existing methods typically compute a single user representation from historical trajectories and use it to score all candidate POIs uniformly. However, this candidate-agnostic paradigm overlooks that the relevance of historical visits inherently depends on which candidate is being evaluated. In this paper, we propose CaST-POI, a candidate-conditioned spatiotemporal model for next POI recommendation. Our key insight is that the same user history should be interpreted differently when evaluating different candidate POIs. CaST-POI employs a candidate-conditioned sequence reader that uses candidates as queries to dynamically attend to user history. In addition, we introduce candidate-relative temporal and spatial biases to capture fine-grained mobility patterns based on the relationships between historical visits and each candidate POI. Extensive experiments on three benchmark datasets demonstrate that CaST-POI consistently outperforms state-of-the-art methods, yielding substantial improvements across multiple evaluation metrics, with particularly strong advantages under large candidate pools. Code is available at https://github.com/YuZhenyuLindy/CaST-POI.git.

Zhenyu Yu, Chunlei Meng, Yangchen Zeng et al.

Next point-of-interest (POI) recommendation requires modeling user mobility as a spatiotemporal sequence, where different behavioral factors may evolve at different temporal and spatial scales. Most existing methods compress a user's history into a single latent representation, which tends to entangle heterogeneous signals such as routine mobility patterns, short-term intent, and temporal regularities. This entanglement limits the flexibility of state evolution and reduces the model's ability to adapt to diverse decision contexts. We propose ADS-POI, a spatiotemporal state decomposition framework for next POI recommendation. ADS-POI represents a user with multiple parallel evolving latent sub-states, each governed by its own spatiotemporal transition dynamics. These sub-states are selectively aggregated through a context-conditioned mechanism to form the decision state used for prediction. This design enables different behavioral components to evolve at different rates while remaining coordinated under the current spatiotemporal context. Extensive experiments on three real-world benchmark datasets from Foursquare and Gowalla demonstrate that ADS-POI consistently outperforms strong state-of-the-art baselines under a full-ranking evaluation protocol. The results show that decomposing user behavior into multiple spatiotemporally aware states leads to more effective and robust next POI recommendation. Our code is available at https://github.com/YuZhenyuLindy/ADS-POI.git.

Zhenyu Yu, Yangchen Zeng, Chunlei Meng et al.

Machine unlearning in Vertical Federated Learning (VFL) has attracted growing interest, yet existing methods certify forgetting solely using output-level metrics. We challenge these claims by introducing Mirage, a representation-level auditing framework comprising four complementary diagnostics: Linear Probe Recovery (LPR), Centered Kernel Alignment (CKA), Feature Separability Scoring, and Layer-Wise Recovery Analysis. Through experiments across seven datasets and seven baseline methods following recent VFL unlearning protocols, Mirage reveals three key findings: (i) Forgetting gap: methods that pass output-level certification still retain substantial class structure in their representations, with LPR exceeding the retrained baseline by up to 15.4 points; CKA shows these models remain structurally closer to the original than to the retrained reference, while separability scores indicate persistent geometric discrimination. (ii) Unlearning trilemma: no existing method simultaneously achieves high utility, output-level forgetting, and representation-level forgetting. (iii) Class-sample asymmetry: class-level forgetting leaves strong representational traces (LPR up to 97%), whereas sample-level forgetting is indistinguishable from chance (LPR approx. 50%); layer-wise analysis further shows residual class information persists across network depths. These findings call for representation-aware evaluation standards in federated unlearning research.

Chunlei Meng, Jiabin Luo, Zhenglin Yan et al.

Multimodal Sentiment Analysis (MSA) integrates language, visual, and acoustic modalities to infer human sentiment. Most existing methods either focus on globally shared representations or modality-specific features, while overlooking signals that are shared only by certain modality pairs. This limits the expressiveness and discriminative power of multimodal representations. To address this limitation, we propose a Tri-Subspace Disentanglement (TSD) framework that explicitly factorizes features into three complementary subspaces: a common subspace capturing global consistency, submodally-shared subspaces modeling pairwise cross-modal synergies, and private subspaces preserving modality-specific cues. To keep these subspaces pure and independent, we introduce a decoupling supervisor together with structured regularization losses. We further design a Subspace-Aware Cross-Attention (SACA) fusion module that adaptively models and integrates information from the three subspaces to obtain richer and more robust representations. Experiments on CMU-MOSI and CMU-MOSEI demonstrate that TSD achieves state-of-the-art performance across all key metrics, reaching 0.691 MAE on CMU-MOSI and 54.9% ACC-7 on CMU-MOSEI, and also transfers well to multimodal intent recognition tasks. Ablation studies confirm that tri-subspace disentanglement and SACA jointly enhance the modeling of multi-granular cross-modal sentiment cues.

Zhenyu Yu, Mohd Yamani Idna Idris, Hua Wang et al.

We present DanceText, a training-free framework for multilingual text editing in images, designed to support complex geometric transformations and achieve seamless foreground-background integration. While diffusion-based generative models have shown promise in text-guided image synthesis, they often lack controllability and fail to preserve layout consistency under non-trivial manipulations such as rotation, translation, scaling, and warping. To address these limitations, DanceText introduces a layered editing strategy that separates text from the background, allowing geometric transformations to be performed in a modular and controllable manner. A depth-aware module is further proposed to align appearance and perspective between the transformed text and the reconstructed background, enhancing photorealism and spatial consistency. Importantly, DanceText adopts a fully training-free design by integrating pretrained modules, allowing flexible deployment without task-specific fine-tuning. Extensive experiments on the AnyWord-3M benchmark demonstrate that our method achieves superior performance in visual quality, especially under large-scale and complex transformation scenarios. Code is avaible at https://github.com/YuZhenyuLindy/DanceText.git.

Ayushman Sarkar, Zhenyu Yu, Mohd Yamani Idna Idris

Maintaining visual and semantic consistency across frames is a key challenge in text-to-image storytelling. Existing training-free methods, such as One-Prompt-One-Story, concatenate all prompts into a single sequence, which often induces strong embedding correlation and leads to color leakage, background blending, and identity drift. We propose DeCorStory, a training-free inference-time framework that explicitly reduces inter-frame semantic interference. DeCorStory applies Gram-Schmidt prompt embedding decorrelation to orthogonalize frame-level semantics, followed by singular value reweighting to strengthen prompt-specific information and identity-preserving cross-attention to stabilize character identity during diffusion. The method requires no model modification or fine-tuning and can be seamlessly integrated into existing diffusion pipelines. Experiments demonstrate consistent improvements in prompt-image alignment, identity consistency, and visual diversity, achieving state-of-the-art performance among training-free baselines. Code is available at: https://github.com/YuZhenyuLindy/DeCorStory

Ayushman Sarkar, Zhenyu Yu, Wei Tang et al.

Large multimodal models have enabled one-click storybook generation, where users provide a short description and receive a multi-page illustrated story. However, the underlying story state, such as characters, world settings, and page-level objects, remains implicit, making edits coarse-grained and often breaking visual consistency. We present StoryState, an agent-based orchestration layer that introduces an explicit and editable story state on top of training-free text-to-image generation. StoryState represents each story as a structured object composed of a character sheet, global settings, and per-page scene constraints, and employs a small set of LLM agents to maintain this state and derive 1Prompt1Story-style prompts for generation and editing. Operating purely through prompts, StoryState is model-agnostic and compatible with diverse generation backends. System-level experiments on multi-page editing tasks show that StoryState enables localized page edits, improves cross-page consistency, and reduces unintended changes, interaction turns, and editing time compared to 1Prompt1Story, while approaching the one-shot consistency of Gemini Storybook. Code is available at https://github.com/YuZhenyuLindy/StoryState

Ayushman Sarkar, Zhenyu Yu, Chu Chen et al.

Generating coherent visual stories requires maintaining subject identity across multiple images while preserving frame-specific semantics. Recent training-free methods concatenate identity and frame prompts into a unified representation, but this often introduces inter-frame semantic interference that weakens identity preservation in complex stories. We propose ReDiStory, a training-free framework that improves multi-frame story generation via inference-time prompt embedding reorganization. ReDiStory explicitly decomposes text embeddings into identity-related and frame-specific components, then decorrelates frame embeddings by suppressing shared directions across frames. This reduces cross-frame interference without modifying diffusion parameters or requiring additional supervision. Under identical diffusion backbones and inference settings, ReDiStory improves identity consistency while maintaining prompt fidelity. Experiments on the ConsiStory+ benchmark show consistent gains over 1Prompt1Story on multiple identity consistency metrics. Code is available at: https://github.com/YuZhenyuLindy/ReDiStory

Yangchen Zeng, Hao Peng, Rongfeng Guo et al.

We introduce TriAlignGR, a unified multitask-multimodal framework for generative recommendation that establishes two-stage multimodal semantic propagation: (i) encoding visual semantics directly into SIDs via multimodal embeddings, and (ii) enabling the model to decode these semantics through visual description tasks. Existing Semantic ID (SID) pipelines suffer from two fundamental but underexplored problems: \textbf{SID Content Degradation (SCD)}, where cascaded encoding and residual quantization discard critical multimodal and interest-level semantics; and \textbf{SID Semantic Opacity (SSO)}, where models autoregressively generate SID sequences without truly comprehending their underlying meaning, leading to hallucination and poor generalization. Prior work addresses at most text-SID alignment, leaving visual semantics and latent user interests entirely unexploited. TriAlignGR resolves both problems through three tightly integrated components: (1)~\textbf{Cross-Modal Semantic Alignment (CMSA)} integrates visual content into SID construction through both VLM-generated textual descriptions and a multimodal embedding model that directly encodes image features alongside text, ensuring that SIDs inherently carry multimodal semantics; (2)~\textbf{Multimodal Deep Interest Mining (MDIM)} leverages LLM Chain-of-Thought reasoning to extract latent user intents (\eg ``productivity-focused lifestyle'' from noise-canceling headphones) beyond surface attributes, enriching SID semantics before discretization; and (3)~\textbf{Triangular Multitask (TMT)} jointly trains on eight complementary generation tasks under a single autoregressive loss -- including two novel visual-semantic tasks (VisDesc$\to$SID, VisDesc$\to$Title) that map VLM-generated image descriptions to SIDs and titles, completing the SID-Text-Image triangle -- without requiring task-specific towers or complex loss weighting.

Zhenyu Yu, Chee Seng Chan

Generative AI presents transformative potential across various domains, from creative arts to scientific visualization. However, the utility of AI-generated imagery is often compromised by visual flaws, including anatomical inaccuracies, improper object placements, and misplaced textual elements. These imperfections pose significant challenges for practical applications. To overcome these limitations, we introduce \textit{Yuan}, a novel framework that autonomously corrects visual imperfections in text-to-image synthesis. \textit{Yuan} uniquely conditions on both the textual prompt and the segmented image, generating precise masks that identify areas in need of refinement without requiring manual intervention -- a common constraint in previous methodologies. Following the automated masking process, an advanced inpainting module seamlessly integrates contextually coherent content into the identified regions, preserving the integrity and fidelity of the original image and associated text prompts. Through extensive experimentation on publicly available datasets such as ImageNet100 and Stanford Dogs, along with a custom-generated dataset, \textit{Yuan} demonstrated superior performance in eliminating visual imperfections. Our approach consistently achieved higher scores in quantitative metrics, including NIQE, BRISQUE, and PI, alongside favorable qualitative evaluations. These results underscore \textit{Yuan}'s potential to significantly enhance the quality and applicability of AI-generated images across diverse fields.

Zhenyu Yu, Mohd Yamani Idna Idris, Hua Wang et al.

Quantitative remote sensing inversion aims to estimate continuous surface variables-such as biomass, vegetation indices, and evapotranspiration-from satellite observations, supporting applications in ecosystem monitoring, carbon accounting, and land management. With the evolution of remote sensing systems and artificial intelligence, traditional physics-based paradigms are giving way to data-driven and foundation model (FM)-based approaches. This paper systematically reviews the methodological evolution of inversion techniques, from physical models (e.g., PROSPECT, SCOPE, DART) to machine learning methods (e.g., deep learning, multimodal fusion), and further to foundation models (e.g., SatMAE, GFM, mmEarth). We compare the modeling assumptions, application scenarios, and limitations of each paradigm, with emphasis on recent FM advances in self-supervised pretraining, multi-modal integration, and cross-task adaptation. We also highlight persistent challenges in physical interpretability, domain generalization, limited supervision, and uncertainty quantification. Finally, we envision the development of next-generation foundation models for remote sensing inversion, emphasizing unified modeling capacity, cross-domain generalization, and physical interpretability.

Zhenyu Yu, Mohd Yamani Inda Idris, Pei Wang

Remote sensing imagery is essential for environmental monitoring, agricultural management, and disaster response. However, data loss due to cloud cover, sensor failures, or incomplete acquisition-especially in high-resolution and high-frequency tasks-severely limits satellite imagery's effectiveness. Traditional interpolation methods struggle with large missing areas and complex structures. Remote sensing imagery consists of multiple bands, each with distinct meanings, and ensuring consistency across bands is critical to avoid anomalies in the combined images. This paper proposes SatelliteMaker, a diffusion-based method that reconstructs missing data across varying levels of data loss while maintaining spatial, spectral, and temporal consistency. We also propose Digital Elevation Model (DEM) as a conditioning input and use tailored prompts to generate realistic images, making diffusion models applicable to quantitative remote sensing tasks. Additionally, we propose a VGG-Adapter module based on Distribution Loss, which reduces distribution discrepancy and ensures style consistency. Extensive experiments show that SatelliteMaker achieves state-of-the-art performance across multiple tasks.

Zhenyu Yu, Mohd. Yamani Idna Idris, Pei Wang

Quantitative remote sensing inversion plays a critical role in environmental monitoring, enabling the estimation of key ecological variables such as vegetation indices, canopy structure, and carbon stock. Although vision foundation models have achieved remarkable progress in classification and segmentation tasks, their application to physically interpretable regression remains largely unexplored. Furthermore, the multi-spectral nature and geospatial heterogeneity of remote sensing data pose significant challenges for generalization and transferability. To address these issues, we introduce SatelliteCalculator, the first vision foundation model tailored for quantitative remote sensing inversion. By leveraging physically defined index formulas, we automatically construct a large-scale dataset of over one million paired samples across eight core ecological indicators. The model integrates a frozen Swin Transformer backbone with a prompt-guided architecture, featuring cross-attentive adapters and lightweight task-specific MLP decoders. Experiments on the Open-Canopy benchmark demonstrate that SatelliteCalculator achieves competitive accuracy across all tasks while significantly reducing inference cost. Our results validate the feasibility of applying foundation models to quantitative inversion, and provide a scalable framework for task-adaptive remote sensing estimation.

Ayushman Sarkar, Mohd Yamani Idna Idris, Zhenyu Yu

Visual reasoning is critical for a wide range of computer vision tasks that go beyond surface-level object detection and classification. Despite notable advances in relational, symbolic, temporal, causal, and commonsense reasoning, existing surveys often address these directions in isolation, lacking a unified analysis and comparison across reasoning types, methodologies, and evaluation protocols. This survey aims to address this gap by categorizing visual reasoning into five major types (relational, symbolic, temporal, causal, and commonsense) and systematically examining their implementation through architectures such as graph-based models, memory networks, attention mechanisms, and neuro-symbolic systems. We review evaluation protocols designed to assess functional correctness, structural consistency, and causal validity, and critically analyze their limitations in terms of generalizability, reproducibility, and explanatory power. Beyond evaluation, we identify key open challenges in visual reasoning, including scalability to complex scenes, deeper integration of symbolic and neural paradigms, the lack of comprehensive benchmark datasets, and reasoning under weak supervision. Finally, we outline a forward-looking research agenda for next-generation vision systems, emphasizing that bridging perception and reasoning is essential for building transparent, trustworthy, and cross-domain adaptive AI systems, particularly in critical domains such as autonomous driving and medical diagnostics.

Zhenyu Yu, Jinnian Wang, Mohd Yamani Idna Idris

The forest serves as the most significant terrestrial carbon stock mechanism, effectively reducing atmospheric CO2 concentrations and mitigating climate change. Remote sensing provides high data accuracy and enables large-scale observations. Optical images facilitate long-term monitoring, which is crucial for future carbon stock estimation studies. This study focuses on Huize County, Qujing City, Yunnan Province, China, utilizing GF-1 WFV satellite imagery. The KD-VGG and KD-UNet modules were introduced for initial feature extraction, and the improved implicit diffusion model (IIDM) was proposed. The results showed: (1) The VGG module improved initial feature extraction, improving accuracy, and reducing inference time with optimized model parameters. (2) The Cross-attention + MLPs module enabled effective feature fusion, establishing critical relationships between global and local features, achieving high-accuracy estimation. (3) The IIDM model, a novel contribution, demonstrated the highest estimation accuracy with an RMSE of 12.17%, significantly improving by 41.69% to 42.33% compared to the regression model. In carbon stock estimation, the generative model excelled in extracting deeper features, significantly outperforming other models, demonstrating the feasibility of AI-generated content in quantitative remote sensing. The 16-meter resolution estimates provide a robust basis for tailoring forest carbon sink regulations, enhancing regional carbon stock management.

Zhenyu Yu, Hanqing Chen, Mohd Yamani Idna Idris et al.

Precipitation plays a critical role in the Earth's hydrological cycle, directly affecting ecosystems, agriculture, and water resource management. Accurate precipitation estimation and prediction are crucial for understanding climate dynamics, disaster preparedness, and environmental monitoring. In recent years, artificial intelligence (AI) has gained increasing attention in quantitative remote sensing (QRS), enabling more advanced data analysis and improving precipitation estimation accuracy. Although traditional methods have been widely used for precipitation estimation, they face limitations due to the difficulty of data acquisition and the challenge of capturing complex feature relationships. Furthermore, the lack of standardized multi-source satellite datasets, and in most cases, the exclusive reliance on station data, significantly hinders the effective application of advanced AI models. To address these challenges, we propose the Rainy dataset, a multi-source spatio-temporal dataset that integrates pure satellite data with station data, and propose Taper Loss, designed to fill the gap in tasks where only in-situ data is available without area-wide support. The Rainy dataset supports five main tasks: (1) satellite calibration, (2) precipitation event prediction, (3) precipitation level prediction, (4) spatiotemporal prediction, and (5) precipitation downscaling. For each task, we selected benchmark models and evaluation metrics to provide valuable references for researchers. Using precipitation as an example, the Rainy dataset and Taper Loss demonstrate the seamless collaboration between QRS and computer vision, offering data support for AI for Science in the field of QRS and providing valuable insights for interdisciplinary collaboration and integration.

Zhenyu Yu, Mohd Yamani Idna Idris, Pei Wang

Cloud occlusion significantly hinders remote sensing applications by obstructing surface information and complicating analysis. To address this, we propose DC4CR (Diffusion Control for Cloud Removal), a novel multimodal diffusion-based framework for cloud removal in remote sensing imagery. Our method introduces prompt-driven control, allowing selective removal of thin and thick clouds without relying on pre-generated cloud masks, thereby enhancing preprocessing efficiency and model adaptability. Additionally, we integrate low-rank adaptation for computational efficiency, subject-driven generation for improved generalization, and grouped learning to enhance performance on small datasets. Designed as a plug-and-play module, DC4CR seamlessly integrates into existing cloud removal models, providing a scalable and robust solution. Extensive experiments on the RICE and CUHK-CR datasets demonstrate state-of-the-art performance, achieving superior cloud removal across diverse conditions. This work presents a practical and efficient approach for remote sensing image processing with broad real-world applications.

Zhenyu Yu, Mohd Yamani Inda Idris, Pei Wang

The widespread adoption of diffusion models in image generation has increased the demand for privacy-compliant unlearning. However, due to the high-dimensional nature and complex feature representations of diffusion models, achieving selective unlearning remains challenging, as existing methods struggle to remove sensitive information while preserving the consistency of non-sensitive regions. To address this, we propose an Automatic Dataset Creation Framework based on prompt-based layered editing and training-free local feature removal, constructing the ForgetMe dataset and introducing the Entangled evaluation metric. The Entangled metric quantifies unlearning effectiveness by assessing the similarity and consistency between the target and background regions and supports both paired (Entangled-D) and unpaired (Entangled-S) image data, enabling unsupervised evaluation. The ForgetMe dataset encompasses a diverse set of real and synthetic scenarios, including CUB-200-2011 (Birds), Stanford-Dogs, ImageNet, and a synthetic cat dataset. We apply LoRA fine-tuning on Stable Diffusion to achieve selective unlearning on this dataset and validate the effectiveness of both the ForgetMe dataset and the Entangled metric, establishing them as benchmarks for selective unlearning. Our work provides a scalable and adaptable solution for advancing privacy-preserving generative AI.

Avash Palikhe, Zhenyu Yu, Zichong Wang et al.

Large Language Models (LLMs) have played a pivotal role in advancing Artificial Intelligence (AI). However, despite their achievements, LLMs often struggle to explain their decision-making processes, making them a 'black box' and presenting a substantial challenge to explainability. This lack of transparency poses a significant obstacle to the adoption of LLMs in high-stakes domain applications, where interpretability is particularly essential. To overcome these limitations, researchers have developed various explainable artificial intelligence (XAI) methods that provide human-interpretable explanations for LLMs. However, a systematic understanding of these methods remains limited. To address this gap, this survey provides a comprehensive review of explainability techniques by categorizing XAI methods based on the underlying transformer architectures of LLMs: encoder-only, decoder-only, and encoder-decoder models. Then these techniques are examined in terms of their evaluation for assessing explainability, and the survey further explores how these explanations are leveraged in practical applications. Finally, it discusses available resources, ongoing research challenges, and future directions, aiming to guide continued efforts toward developing transparent and responsible LLMs.

Zhenyu Yu, Jinnian Wang

Forest is the most significant land-based carbon storage mechanism. The forest carbon sink can effectively decrease the atmospheric CO2 concentration and mitigate climate change. Remote sensing estimation not only ensures high accuracy of data, but also enables large-scale area observation. Optical images provide the possibility for long-term monitoring, which is a potential issue in the future carbon storage estimation research. We chose Huize County, Qujing City, Yunnan Province, China as the study area, took GF-1 WFV satellite image as the data, introduced the KD-VGG module to extract the initial features, and proposed the improved implicit diffusion model (IIDM). The results showed that: (1) The VGG-19 module after knowledge distillation can realize the initial feature extraction, reduce the inference time and improve the accuracy in the case of reducing the number of model parameters. (2) The Attention + MLP module was added for feature fusion to obtain the relationship between global and local features and realized the restoration of high-fidelity images in the continuous scale range. (3) The IIDM model proposed in this paper had the highest estimation accuracy, with RMSE of 28.68, which was 13.16 higher than that of the regression model, about 31.45%. In the estimation of carbon storage, the generative model can extract deeper features, and its performance was significantly better than other models. It demonstrated the feasibility of artificial intelligence-generated content (AIGC) in the field of quantitative remote sensing and provided valuable insights for the study of carbon neutralization effect. By combining the actual characteristics of the forest, the regional carbon storage estimation with a resolution of 16-meter was utilized to provide a significant theoretical basis for the formulation of forest carbon sink regulation.

Shengkai Xu, Hsiang Lun Kao, Tianxiang Xu et al.

Polyp detectors trained on clean datasets often underperform in real-world endoscopy, where illumination changes, motion blur, and occlusions degrade image quality. Existing approaches struggle with the domain gap between controlled laboratory conditions and clinical practice, where adverse imaging conditions are prevalent. In this work, we propose AdaptiveDetector, a novel two-stage detector-verifier framework comprising a YOLOv11 detector with a vision-language model (VLM) verifier. The detector adaptively adjusts per-frame confidence thresholds under VLM guidance, while the verifier is fine-tuned with Group Relative Policy Optimization (GRPO) using an asymmetric, cost-sensitive reward function specifically designed to discourage missed detections -- a critical clinical requirement. To enable realistic assessment under challenging conditions, we construct a comprehensive synthetic testbed by systematically degrading clean datasets with adverse conditions commonly encountered in clinical practice, providing a rigorous benchmark for zero-shot evaluation. Extensive zero-shot evaluation on synthetically degraded CVC-ClinicDB and Kvasir-SEG images demonstrates that our approach improves recall by 14 to 22 percentage points over YOLO alone, while precision remains within 0.7 points below to 1.7 points above the baseline. This combination of adaptive thresholding and cost-sensitive reinforcement learning achieves clinically aligned, open-world polyp detection with substantially fewer false negatives, thereby reducing the risk of missed precancerous polyps and improving patient outcomes.

Tarique Dahri, Zulfiqar Ali Memon, Zhenyu Yu et al.

We introduce Layered Self-Supervised Knowledge Distillation (LSSKD) framework for training compact deep learning models. Unlike traditional methods that rely on pre-trained teacher networks, our approach appends auxiliary classifiers to intermediate feature maps, generating diverse self-supervised knowledge and enabling one-to-one transfer across different network stages. Our method achieves an average improvement of 4.54\% over the state-of-the-art PS-KD method and a 1.14% gain over SSKD on CIFAR-100, with a 0.32% improvement on ImageNet compared to HASSKD. Experiments on Tiny ImageNet and CIFAR-100 under few-shot learning scenarios also achieve state-of-the-art results. These findings demonstrate the effectiveness of our approach in enhancing model generalization and performance without the need for large over-parameterized teacher networks. Importantly, at the inference stage, all auxiliary classifiers can be removed, yielding no extra computational cost. This makes our model suitable for deploying small language models on affordable low-computing devices. Owing to its lightweight design and adaptability, our framework is particularly suitable for multimodal sensing and cyber-physical environments that require efficient and responsive inference. LSSKD facilitates the development of intelligent agents capable of learning from limited sensory data under weak supervision.

Zhenyu Yu, Jinnian Wang

Forests function as crucial carbon reservoirs on land, and their carbon sinks can efficiently reduce atmospheric CO2 concentrations and mitigate climate change. Currently, the overall trend for monitoring and assessing forest carbon stocks is to integrate ground monitoring sample data with satellite remote sensing imagery. This style of analysis facilitates large-scale observation. However, these techniques require improvement in accuracy. We used GF-1 WFV and Landsat TM images to analyze Huize County, Qujing City, Yunnan Province in China. Using the style transfer method, we introduced Swin Transformer to extract global features through attention mechanisms, converting the carbon stock estimation into an image translation.

Shaina Raza, Caesar Saleh, Azib Farooq et al.

Detecting bias in multimodal news requires models that reason over text--image pairs, not just classify text. In response, we present ViLBias, a VQA-style benchmark and framework for detecting and reasoning about bias in multimodal news. The dataset comprises 40,945 text--image pairs from diverse outlets, each annotated with a bias label and concise rationale using a two-stage LLM-as-annotator pipeline with hierarchical majority voting and human-in-the-loop validation. We evaluate Small Language Models (SLMs), Large Language Models (LLMs), and Vision--Language Models (VLMs) across closed-ended classification and open-ended reasoning (oVQA), and compare parameter-efficient tuning strategies. Results show that incorporating images alongside text improves detection accuracy by 3--5\%, and that LLMs/VLMs better capture subtle framing and text--image inconsistencies than SLMs. Parameter-efficient methods (LoRA/QLoRA/Adapters) recover 97--99\% of full fine-tuning performance with $<5\%$ trainable parameters. For oVQA, reasoning accuracy spans 52--79\% and faithfulness 68--89\%, both improved by instruction tuning; closed accuracy correlates strongly with reasoning ($r = 0.91$). ViLBias offers a scalable benchmark and strong baselines for multimodal bias detection and rationale quality.