arXiv subjects

Hongbo Xu

Publications and source records attributed to Hongbo Xu.

At least 19 recordsLinked to original sources

arXiv2026

Where to Look and What to Use: Retrieve-Localize-Generate for Long-Term Conversational Memory Question Answering

Retrieval-augmented generation (RAG) enables large language models (LLMs) to answer questions by accessing external knowledge and has been widely adopted for long-term conversational memory question answering. However, existing methods suffer from two key challenges: (1) fragmented evidence scattered across temporally distant sessions, and (2) noisy content within retrieved sessions that triggers the lost-in-the-middle effect. To address these challenges, we propose MemLoc, a unified Retrieve-Localize-Generate framework for long-term conversational memory QA. For retrieval, MemLoc decomposes each session into multi-granularity memory units and performs query routing via an inner-memory graph with entropy-based granularity selection. It further models cross-session semantic and temporal dependencies through a cross-memory graph, enabling coarse-to-fine retrieval of top-K relevant memory candidates. For localization, we introduce a reasoning-based evidence locator trained with Self-reflective Hint Policy Optimization (SHPO), which performs progressive refinement by extracting query-relevant fragments within memory units to suppress noise and reranking across candidates to remove redundancy, producing a compact evidence set with lightweight location IDs. For generation, these IDs act as precise grounding signals that guide the LLM to the correct memory positions, mitigating the lost-in-the-middle effect while preserving original contextual integrity. Extensive experiments on four benchmarks demonstrate that MemLoc achieves state-of-the-art retrieval accuracy and response quality while maintaining efficiency. Our code is available at: https://github.com/Nikol-coder/MemLoc.

cs.CL↗

arXiv2026

HUG-VIS: A Multimodal Benchmark for Human-centered Understanding and Generation in Visual Intelligence

Visual intelligence seeks to perceive, interpret, and synthesize the visual world and is central to modern computer vision. Human-centered visual intelligence is especially demanding because it studies people as expressive, socially situated subjects whose meaning is rarely conveyed by appearance alone. It couples vision with audio and language across four representative tasks: human emotion recognition, human video generation, human voice cloning, and human video matting. Yet existing resources remain task-specific, providing modalities and annotations for individual problems rather than a shared foundation coordinating understanding and generation. This limits multimodal signal use and broader research. We address this gap with HUG-VIS, a unified benchmark for Human-centered Understanding and Generation in Visual Intelligence. It contains 8,400 seated half-body videos of 30 professional actors, each performing the same 280 emotion-action-prompt assignments under a controlled Mandarin studio protocol, with synchronized video, audio, text, and alpha mattes. We evaluate diverse open- and closed-source models across the four tasks under a unified zero-shot protocol using automatic metrics, criterion-specific mean opinion scores, and multiple cross-task analyses. Results show that (i) linguistic content dominates current emotion recognition, while purely visual affect recognition is weakest; (ii) in video generation and voice cloning, automatic metrics and human judgment agree overall but differ in their top rankings, requiring joint reporting; (iii) boundary fidelity under motion is the main remaining obstacle for human matting; and (iv) task difficulty varies across emotions, models, and metrics, with notable cross-task correlations. The dataset and results are available at https://github.com/GML-MMGroup/HUG-VIS.

cs.CV↗

arXiv2026

HOIMask: Towards Generative Masked Modeling for Human Object Interaction Generation

Diffusion-based methods have dominated the HOI generation, as they enable critical contact fusions or signals to guide the diffusion process. However, they often result in high artifacts and unstable interaction quality due to error accumulation during iterative denoising. In this work, we propose HOIMask, the first generative masked framework for modeling HOI motion in discrete space. HOIMask first encodes both motion sequences and contact-aware signals into discrete 2D human and object token maps via HOI Vector Quantization (VQ), preserving fine-grained spatial-temporal structure beyond conventional 1D representations. On this basis, a generative masked modeling framework is employed to jointly capture human-object interaction dynamics, leveraging a transformer architecture designed to model complex spatial-temporal and interaction dependencies. To generate more coherent and physically plausible motions, we further introduce a novel contact-aware reconstruction guidance in discrete space during inference, which fuses contact signals to optimize HOI tokens that forces the generated motion with higher spatio-temporal consistency. With craftily designed motion interaction tokens, dedicated architecture and guidance strategy, HOIMask outperforms state-of-the-art diffusion-based methods, generating more realistic and semantically aligned HOI motions. Please refer to https://jyhflash.github.io/HOIMask/ for more results.

cs.CV↗

arXiv2026

OPD-IAD: From Language Judgment to Industrial Anomaly Detection via On-Policy Self-Distillation

Large vision-language models (LVLMs) have recently shown strong potential for industrial anomaly detection (IAD) by providing image-level anomaly judgments and interpretable defect reasoning. However, current LVLM-based IAD methods still struggle to produce precise pixel-level anomaly maps from generated language judgments. We aim to achieve precise pixel-level localization while using language as guidance rather than letting it dominate the visual response. Specifically, we propose \textbf{OPD-IAD}, an evidence-privileged dense on-policy self-distillation framework for LVLM-based IAD. OPD-IAD distills privileged defect evidence onto the model's own on-policy judgment trajectory, enabling the final generated judgment to be learned under dense supervision rather than treated only as a textual answer. The resulting judgment serves as a semantic condition for dense anomaly perception. To turn this condition into dense visual evidence, we introduce \textbf{Language-guided Visual Anchoring}, which uses a judgment reforward to re-encode the image and question under the final-judgment condition into semantic anchors and contrasts them with dense visual features through a contrastive heatmap head to generate anomaly maps. The language judgment therefore provides compact semantic guidance, while dense visual features remain the basis for pixel-level scoring, allowing language to guide anomaly localization without letting language quality directly dictate the pixel-level response. Extensive experiments show that OPD-IAD achieves the best overall performance among LVLM-based IAD methods, leading on most image-level, pixel-level, and QA metrics.

cs.CV↗

arXiv2026

Sparse Evidence Can Suffice: Agentic Evidence Seeking for Multimodal Video Misinformation Detection

Multimodal video misinformation detection is commonly formulated as a holistic video-understanding task, where the entire video and its associated content are processed and judged in a single pass. However, real-world misinformation often exhibits a sparse and compositional evidence structure: a reliable decision may depend on only a few coupled clues, while most video content contributes limited additional information. Exhaustive multimodal reasoning may therefore introduce substantial redundancy and obscure decisive evidence. This motivates decoupling evidence acquisition from verification: first identifying sparse, decision-relevant clues and then judging veracity based on the acquired evidence. Accordingly, we propose SIEVE, a framework for Sparse Interactive Evidence Verification via Extraction in multimodal video misinformation detection. An evidence-seeking agent actively explores the available multimodal evidence and constructs a compact evidence package, which is then used by a verifier to determine veracity. The agent is trained with supervised evidence-seeking trajectories and an evidence-aware reinforcement learning objective that promotes informative evidence acquisition while discouraging unnecessary or invalid interactions. Experiments on multiple video misinformation benchmarks show that SIEVE consistently outperforms the evaluated baselines and supports reliable verification using compact evidence packages. Moreover, the resulting acquisition process provides an explicit and inspectable evidence trail, improving the transparency and groundedness of multimodal misinformation detection.

cs.CV↗

arXiv2026

Energy Efficiency Optimization for Rotatable Antenna-Enabled Uplink NOMA Systems

This paper investigates a rotatable antenna (RA)-enabled uplink non-orthogonal multiple access (NOMA) system, where a base station equipped with multiple independently RAs serves both ground and aerial users. Specifically, we formulate an energy efficiency (EE) maximization problem by jointly optimizing receive beamforming, user power allocation, and RA rotation. To make the problem tractable, a new block coordinate descent-based algorithm is developed, in which the receive beamforming is updated via the minimum mean square error criterion, while the power allocation and RA rotation are handled by fractional programming and successive convex approximation. Numerical results demonstrate the EE superiority of the proposed RA-NOMA scheme over several benchmarks.

cs.IT↗

arXiv2026

Does Unification Come at a Cost? Uni-SafeBench: A Safety Benchmark for Unified Multimodal Large Models

Unified Multimodal Large Models (UMLMs) integrate understanding and generation capabilities within a single architecture. While unified architectures expand multimodal capabilities, their safety implications remain important yet underexplored. Existing safety benchmarks predominantly focus on isolated understanding or generation tasks, failing to evaluate the holistic safety of UMLMs when handling diverse tasks under a unified framework. To address this, we introduce Uni-SafeBench, a comprehensive benchmark featuring a taxonomy of six major safety categories across seven task types. To ensure rigorous assessment, we develop Uni-Judger, a framework that effectively decouples contextual safety from intrinsic safety. Based on comprehensive evaluations across Uni-SafeBench, we find that the original safety alignment of the underlying LLM is not consistently preserved in current unified models. Moreover, open-source UMLMs exhibit much lower safety performance than multimodal large models specialized for either generation or understanding tasks, particularly on the generation side.

cs.AI↗

arXiv2026

Backdoor Sentinel: Detecting and Detoxifying Backdoors in Diffusion Models via Temporal Noise Consistency

Diffusion models have been widely deployed in AIGC services, but their reliance on opaque training data exposes them to backdoor attacks. In practical auditing scenarios, auditors are typically unable to access model parameters due to intellectual property protection, making white-box or query-intensive detection impractical. After detection, existing detoxification approaches are trapped in a dilemma between detoxification effectiveness and generation quality for service providers. We reveal Temporal Noise Consistency (TNC), a previously unreported phenomenon in which backdoor activation disrupts the consistency of noise predictions between adjacent diffusion timesteps within specific temporal segments, while clean inputs remain stable. Based on this finding, we propose TNC-Defense, a closed-loop framework for gray-box backdoor detection and model repair. Specifically, TNC-Detect (for auditors) uses inference-stage adjacent-noise statistics to detect backdoors and precisely localize anomalous timesteps without model-weight access. TNC-Detox (for service providers) utilizes these locations to perform trigger-agnostic, timestep-aware correction of the generation path, suppressing backdoor behavior while reducing detoxification cost. Across five representative backdoor attacks and state-of-the-art defenses, TNC-Defense improves the average detection accuracy by $11\%$ with negligible additional overhead, and invalidates an average of $98.5\%$ of triggered samples with only a mild degradation in generation quality. Our code is publicly available at: https://github.com/binzhwang/TNC-Defense.

cs.CR↗

arXiv2025

MMSD3.0: A Multi-Image Benchmark for Real-World Multimodal Sarcasm Detection

Despite progress in multimodal sarcasm detection, existing datasets and methods predominantly focus on single-image scenarios, overlooking potential semantic and affective relations across multiple images. This leaves a gap in modeling cases where sarcasm is triggered by multi-image cues in real-world settings. To bridge this gap, we introduce MMSD3.0, a new benchmark composed entirely of multi-image samples curated from tweets and Amazon reviews. We further propose the Cross-Image Reasoning Model (CIRM), which performs targeted cross-image sequence modeling to capture latent inter-image connections. In addition, we introduce a relevance-guided, fine-grained cross-modal fusion mechanism based on text-image correspondence to reduce information loss during integration. We establish a comprehensive suite of strong and representative baselines and conduct extensive experiments, showing that MMSD3.0 is an effective and reliable benchmark that better reflects real-world conditions. Moreover, CIRM demonstrates state-of-the-art performance across MMSD, MMSD2.0 and MMSD3.0, validating its effectiveness in both single-image and multi-image scenarios. Dataset and code are publicly available at https://github.com/ZHCMOONWIND/MMSD3.0.

cs.CV↗

arXiv2025

Channel Estimation for Rydberg Atomic Quantum Receivers: Unrolled Phase Retrieval from Holographic Snapshots

A model-driven deep learning framework is proposed for channel estimation in Rydberg atomic quantum receivers (RAQRs) based on the measurement of holographic snapshots. Specifically, we develop a Transformer-based unrolling architecture, termed URformer, to solve the non-linear biased phase retrieval problem, which is derived by unrolling a stabilized variant of the expectation-maximization Gerchberg-Saxton (EM-GS) algorithm. Each layer of the proposed URformer incorporates three trainable modules: 1) a learnable filter network that replaces the fixed Bessel kernel in the classic EM-GS algorithm; 2) a trainable gating mechanism that adaptively combines classic updates to ensure training stability; and 3) an efficient channel Transformer module that learns to correct residual errors by capturing non-local channel dependencies. Numerical results demonstrate that the proposed URformer significantly outperforms classic iterative algorithms and conventional black-box neural networks with less pilot overhead.

cs.IT↗

arXiv2025

REMOTE: A Unified Multimodal Relation Extraction Framework with Multilevel Optimal Transport and Mixture-of-Experts

Multimodal relation extraction (MRE) is a crucial task in the fields of Knowledge Graph and Multimedia, playing a pivotal role in multimodal knowledge graph construction. However, existing methods are typically limited to extracting a single type of relational triplet, which restricts their ability to extract triplets beyond the specified types. Directly combining these methods fails to capture dynamic cross-modal interactions and introduces significant computational redundancy. Therefore, we propose a novel \textit{unified multimodal Relation Extraction framework with Multilevel Optimal Transport and mixture-of-Experts}, termed REMOTE, which can simultaneously extract intra-modal and inter-modal relations between textual entities and visual objects. To dynamically select optimal interaction features for different types of relational triplets, we introduce mixture-of-experts mechanism, ensuring the most relevant modality information is utilized. Additionally, considering that the inherent property of multilayer sequential encoding in existing encoders often leads to the loss of low-level information, we adopt a multilevel optimal transport fusion module to preserve low-level features while maintaining multilayer encoding, yielding more expressive representations. Correspondingly, we also create a Unified Multimodal Relation Extraction (UMRE) dataset to evaluate the effectiveness of our framework, encompassing diverse cases where the head and tail entities can originate from either text or image. Extensive experiments show that REMOTE effectively extracts various types of relational triplets and achieves state-of-the-art performanc on almost all metrics across two other public MRE datasets. We release our resources at https://github.com/Nikol-coder/REMOTE.

cs.MM↗

arXiv2025

Joint Transmit and Pinching Beamforming Optimization in Pinching Antenna-Assisted Symbiotic Radio Systems

This paper investigates a novel downlink symbiotic radio framework enabled by the pinching antenna system (PASS), designed to enhance both primary and secondary transmissions through reconfigurable antenna positioning. This reconfigurability introduces additional degrees of freedom for adaptive pinching beamforming, thereby enabling constructive signal enhancement and interference suppression tailored to the locations of the backscatter device, the Internet of Things (IoT) receiver, and the primary receivers. To fully exploit these benefits, we formulate a joint transmit and pinching beamforming optimization problem that maximizes the achievable sum rate while satisfying the IoT receiver's detection error probability constraint and feasible deployment constraints for the pinching antennas. The resulting problem is inherently nonconvex and highly coupled. To address this challenge, we develop two complementary solution approaches. The first is a learning-aided gradient descent method, where the constrained optimization is reformulated into a differentiable form and solved through end-to-end learning. In this approach, the pinching antenna position matrix is reparameterized to automatically satisfy minimum spacing constraints, while transmit power and waveguide length limits are enforced via projection and normalization. The second approach is an optimization-based successive convex approximation-particle swarm optimization method, which first determines the transmit beamforming solution using successive convex approximation and subsequently optimizes pinching beamforming via a particle swarm optimization search over candidate pinching antenna placements.

eess.SP↗

arXiv2025

Enhancing Long Video Question Answering with Scene-Localized Frame Grouping

Current Multimodal Large Language Models (MLLMs) often perform poorly in long video understanding, primarily due to resource limitations that prevent them from processing all video frames and their associated information. Efficiently extracting relevant information becomes a challenging task. Existing frameworks and evaluation tasks focus on identifying specific frames containing core objects from a large number of irrelevant frames, which does not align with the practical needs of real-world applications. To address this issue, we propose a new scenario under the video question-answering task, SceneQA, which emphasizes scene-based detail perception and reasoning abilities. And we develop the LVSQA dataset to support the SceneQA task, which is built upon carefully selected videos from LVBench and contains a new collection of question-answer pairs to promote a more fair evaluation of MLLMs' scene perception abilities in long videos. Inspired by human cognition, we introduce a novel method called SLFG. The core idea of SLFG is to combine individual frames into semantically coherent scene frames. By leveraging scene localization methods and dynamic frame reassembly mechanisms, SLFG significantly enhances the understanding capabilities of existing MLLMs in long videos. SLFG requires no modification to the original model architecture and boasts excellent plug-and-play usability. Experimental results show that this method performs exceptionally well in several long video benchmark tests. Code and dataset will be released at http://www.slfg.pkuzwh.cn.

cs.CV↗

arXiv2025

Rotatable RIS Assisted Physical Layer Multicasting

Reconfigurable Intelligent Surfaces (RIS) dynamically control signal propagation to enhance wireless communications. This paper presents a novel framework for rotatable RIS assisted physical-layer multicast systems, aiming to maximize the sum of minimum multicast rates via joint optimization of base station beamforming, RIS phase shifts, and orientation. Unlike unicast or non-rotatable setups, the rotatable RIS adapts orientation to align signals with user groups, improving fairness and rates for weak users. An alternating optimization approach combines convex optimization for beamforming/phase shifts with exhaustive search and particle swarm optimization (PSO) for orientation. Majorization-Minimization-based algorithms solve subproblems iteratively. Simulation results show the framework achieves 24.1% rate improvement via exhaustive search and 20.0% via PSO over the non-rotatable RIS baseline, with PSO performance close to the exhaustive search upper bound, highlighting the benefits of physical-layer multicast and orientation optimization.

cs.IT↗

arXiv2025

Pinching Antenna-Aided Wireless Powered Communication Networks

In this letter, we investigate a novel pinching antenna (PA)-aided wireless powered communication network (WPCN), in which multiple PAs are activated along a waveguide to establish robust line-of-sight links with multiple devices. Both time division multiple access (TDMA) and non-orthogonal multiple access (NOMA) protocols are considered in the PA-WPCN. Moreover, some practical considerations, including a proportional power model for the PAs, a waveguide transmission loss model, and a nonlinear energy harvesting model, are incorporated into the PA-WPCN. Furthermore, we formulate a sum-rate maximization problem by jointly optimizing resource allocation and PAs position. To address the challenging problem of the PAs position optimization, we propose a high-performance element-wise (EW) algorithm and a low-complexity stochastic parameter differential evolution (SPDE) algorithm. Numerical results validate the remarkable performance of the proposed PA-WPCN and the effectiveness of our algorithms, indicating that optimal performance is attained when the PA power distribution ratio of approximately 0.55-0.6.

cs.NI↗

arXiv2025

EmoBench-M: Benchmarking Emotional Intelligence for Multimodal Large Language Models

With the integration of multimodal large language models (MLLMs) into robotic systems and AI applications, embedding emotional intelligence (EI) capabilities is essential for enabling these models to perceive, interpret, and respond to human emotions effectively in real-world scenarios. Existing static, text-based, or text-image benchmarks overlook the multimodal complexities of real interactions and fail to capture the dynamic, context-dependent nature of emotional expressions, rendering them inadequate for evaluating MLLMs' EI capabilities. To address these limitations, we introduce EmoBench-M, a systematic benchmark grounded in established psychological theories, designed to evaluate MLLMs across 13 evaluation scenarios spanning three hierarchical dimensions: foundational emotion recognition (FER), conversational emotion understanding (CEU), and socially complex emotion analysis (SCEA). Evaluation was conducted on 27 state-of-the-art MLLMs, using both objective task-specific metrics and LLM-based evaluation, revealing a substantial performance gap relative to human-level competence. Even the best performing models, Gemini-3.0-Pro and GPT-5.2, achieve the highest scores on EmoBench-M, 70.5 and 66.5 points respectively. Specialized models such as AffectGPT exhibit uneven performance across EmoBench-M, demonstrating strengths in certain scenarios but generally lacking comprehensive emotional intelligence. By providing a comprehensive, multimodal evaluation framework, EmoBench-M captures both the strengths and weaknesses of current MLLMs across diverse emotional contexts. All benchmark resources, including datasets and code, are publicly available at https://emo-gml.github.io/, facilitating further research and advancement in MLLM emotional intelligence.

cs.CL↗

arXiv2024

Ensemble Predicate Decoding for Unbiased Scene Graph Generation

Scene Graph Generation (SGG) aims to generate a comprehensive graphical representation that accurately captures the semantic information of a given scenario. However, the SGG model's performance in predicting more fine-grained predicates is hindered by a significant predicate bias. According to existing works, the long-tail distribution of predicates in training data results in the biased scene graph. However, the semantic overlap between predicate categories makes predicate prediction difficult, and there is a significant difference in the sample size of semantically similar predicates, making the predicate prediction more difficult. Therefore, higher requirements are placed on the discriminative ability of the model. In order to address this problem, this paper proposes Ensemble Predicate Decoding (EPD), which employs multiple decoders to attain unbiased scene graph generation. Two auxiliary decoders trained on lower-frequency predicates are used to improve the discriminative ability of the model. Extensive experiments are conducted on the VG, and the experiment results show that EPD enhances the model's representation capability for predicates. In addition, we find that our approach ensures a relatively superior predictive capability for more frequent predicates compared to previous unbiased SGG methods.

cs.CV↗

arXiv2024

IBMEA: Exploring Variational Information Bottleneck for Multi-modal Entity Alignment

Multi-modal entity alignment (MMEA) aims to identify equivalent entities between multi-modal knowledge graphs (MMKGs), where the entities can be associated with related images. Most existing studies integrate multi-modal information heavily relying on the automatically-learned fusion module, rarely suppressing the redundant information for MMEA explicitly. To this end, we explore variational information bottleneck for multi-modal entity alignment (IBMEA), which emphasizes the alignment-relevant information and suppresses the alignment-irrelevant information in generating entity representations. Specifically, we devise multi-modal variational encoders to generate modal-specific entity representations as probability distributions. Then, we propose four modal-specific information bottleneck regularizers, limiting the misleading clues in refining modal-specific entity representations. Finally, we propose a modal-hybrid information contrastive regularizer to integrate all the refined modal-specific representations, enhancing the entity similarity between MMKGs to achieve MMEA. We conduct extensive experiments on two cross-KG and three bilingual MMEA datasets. Experimental results demonstrate that our model consistently outperforms previous state-of-the-art methods, and also shows promising and robust performance in low-resource and high-noise data scenarios.

cs.CL↗