arXiv subjects

Yuhao Wang

Publications and source records attributed to Yuhao Wang.

At least 91 records · Page 5Linked to original sources

arXiv2026

VReID-XFD: Video-based Person Re-identification at Extreme Far Distance Challenge Results

Person re-identification (ReID) across aerial and ground views at extreme far distances introduces a distinct operating regime where severe resolution degradation, extreme viewpoint changes, unstable motion cues, and clothing variation jointly undermine the appearance-based assumptions of existing ReID systems. To study this regime, we introduce VReID-XFD, a video-based benchmark and community challenge for extreme far-distance (XFD) aerial-to-ground person re-identification. VReID-XFD is derived from the DetReIDX dataset and comprises 371 identities, 11,288 tracklets, and 11.75 million frames, captured across altitudes from 5.8 m to 120 m, viewing angles from oblique (30 degrees) to nadir (90 degrees), and horizontal distances up to 120 m. The benchmark supports aerial-to-aerial, aerial-to-ground, and ground-to-aerial evaluation under strict identity-disjoint splits, with rich physical metadata. The VReID-XFD-25 Challenge attracted 10 teams with hundreds of submissions. Systematic analysis reveals monotonic performance degradation with altitude and distance, a universal disadvantage of nadir views, and a trade-off between peak performance and robustness. Even the best-performing SAS-PReID method achieves only 43.93 percent mAP in the aerial-to-ground setting. The dataset, annotations, and official evaluation protocols are publicly available at https://www.it.ubi.pt/DetReIDX/ .

cs.CV↗

arXiv2026

A multivariate extension of Azadkia-Chatterjee's rank coefficient

The Azadkia-Chatterjee coefficient is a rank-based measure of dependence between a random variable $Y \in \mathbb{R}$ and a random vector ${\boldsymbol Z} \in \mathbb{R}^{d_Z}$. In this paper, we propose a multivariate extension that measures the dependence between random vectors ${\boldsymbol Y} \in \mathbb{R}^{d_Y}$ and ${\boldsymbol Z} \in \mathbb{R}^{d_Z}$, based on $n$ i.i.d. samples. The proposed coefficient converges almost surely to a limit with the following properties: i) it lies in $[0, 1]$; ii) it is equal to zero if and only if ${\boldsymbol Y}$ and ${\boldsymbol Z}$ are independent; and iii) it is equal to one if and only if ${\boldsymbol Y}$ is almost surely a function of ${\boldsymbol Z}$. Remarkably, the only assumption required by this convergence is that ${\boldsymbol Y}$ is not almost surely a constant vector. We further prove that under the same mild condition and after a proper scaling, this coefficient converges in distribution to a standard normal random variable when ${\boldsymbol Y}$ and ${\boldsymbol Z}$ are independent. This asymptotic normality result allows us to construct a Wald-type hypothesis test of independence based on this coefficient. To compute this coefficient, we propose a merge sort based algorithm that runs in $O(n (\log n)^{d_Y})$. Finally, we show that it can be used to measure the conditional dependence between ${\boldsymbol Y}$ and ${\boldsymbol Z}$ conditional on a third random vector ${\boldsymbol X}$, and prove that the measure is monotonic with respect to the deviation from an independence distribution under certain model restrictions.

math.ST↗

arXiv2026

W2S-AlignTree: Weak-to-Strong Inference-Time Alignment for Large Language Models via Monte Carlo Tree Search

Large Language Models (LLMs) demonstrate impressive capabilities, yet their outputs often suffer from misalignment with human preferences due to the inadequacy of weak supervision and a lack of fine-grained control. Training-time alignment methods like Reinforcement Learning from Human Feedback (RLHF) face prohibitive costs in expert supervision and inherent scalability limitations, offering limited dynamic control during inference. Consequently, there is an urgent need for scalable and adaptable alignment mechanisms. To address this, we propose W2S-AlignTree, a pioneering plug-and-play inference-time alignment framework that synergistically combines Monte Carlo Tree Search (MCTS) with the Weak-to-Strong Generalization paradigm for the first time. W2S-AlignTree formulates LLM alignment as an optimal heuristic search problem within a generative search tree. By leveraging weak model's real-time, step-level signals as alignment proxies and introducing an Entropy-Aware exploration mechanism, W2S-AlignTree enables fine-grained guidance during strong model's generation without modifying its parameters. The approach dynamically balances exploration and exploitation in high-dimensional generation search trees. Experiments across controlled sentiment generation, summarization, and instruction-following show that W2S-AlignTree consistently outperforms strong baselines. Notably, W2S-AlignTree raises the performance of Llama3-8B from 1.89 to 2.19, a relative improvement of 15.9 on the summarization task.

cs.CL↗

arXiv2025

On the physics of nested Markov models: a generalized probabilistic theory perspective

Determining potential probability distributions with a given causal graph is vital for causality studies. To bypass the difficulty in characterizing latent variables in a Bayesian network, the nested Markov model provides an elegant algebraic approach by listing exactly all the equality constraints on the observed variables. However, this algebraically motivated causal model comprises distributions outside Bayesian networks, and its physical interpretation remains vague. In this work, we inspect the nested Markov model through the lens of generalized probabilistic theory, an axiomatic framework to describe general physical theories. We prove that all the equality constraints defining the nested Markov model are valid theory-independently. At the same time, not every distribution within the nested Markov model is implementable, not even via exotic physical theories associated with generalized probability theories (GPTs). To interpret the origin of such a gap, we study three causal models standing between the nested Markov model and the set of all distributions admitting some GPT realization. Each of the successive three models gives a strictly tighter characterization of the physically implementable distribution set; that is, each successive model manifests new types of GPT-inviolable constraints. We further demonstrate each gap through a specially chosen illustrative causal structure. We anticipate our results will enlighten further explorations on the unification of algebraic and physical perspectives of causality.

quant-ph↗

arXiv2025

Bridging Relevance and Reasoning: Rationale Distillation in Retrieval-Augmented Generation

The reranker and generator are two critical components in the Retrieval-Augmented Generation (i.e., RAG) pipeline, responsible for ranking relevant documents and generating responses. However, due to differences in pre-training data and objectives, there is an inevitable gap between the documents ranked as relevant by the reranker and those required by the generator to support answering the query. To address this gap, we propose RADIO, a novel and practical preference alignment framework with RAtionale DIstillatiOn. Specifically, we first propose a rationale extraction method that leverages the reasoning capabilities of Large Language Models (LLMs) to extract the rationales necessary for answering the query. Subsequently, a rationale-based alignment process is designed to rerank the documents based on the extracted rationales, and fine-tune the reranker to align the preferences. We conduct extensive experiments on two tasks across three datasets to demonstrate the effectiveness of our approach compared to baseline methods. Our code is released online to ease reproduction.

cs.CL↗

arXiv2025

Infrared and Visible Image Fusion with Language-Driven Loss in CLIP Embedding Space

Infrared-visible image fusion (IVIF) has attracted much attention owing to the highly-complementary properties of the two image modalities. Due to the lack of ground-truth fused images, the fusion output of current deep-learning based methods heavily depends on the loss functions defined mathematically. As it is hard to well mathematically define the fused image without ground truth, the performance of existing fusion methods is limited. In this paper, we propose to use natural language to express the objective of IVIF, which can avoid the explicit mathematical modeling of fusion output in current losses, and make full use of the advantage of language expression to improve the fusion performance. For this purpose, we present a comprehensive language-expressed fusion objective, and encode relevant texts into the multi-modal embedding space using CLIP. A language-driven fusion model is then constructed in the embedding space, by establishing the relationship among the embedded vectors representing the fusion objective and input image modalities. Finally, a language-driven loss is derived to make the actual IVIF aligned with the embedded language-driven fusion model via supervised training. Experiments show that our method can obtain much better fusion results than existing techniques. The code is available at https://github.com/wyhlaowang/LDFusion.

cs.CV↗

arXiv2025

Histomorphology-Guided Prototypical Multi-Instance Learning for Breast Cancer WSI Classification

Histomorphology is crucial in cancer diagnosis. However, existing whole slide image (WSI) classification methods struggle to effectively incorporate histomorphology information, limiting their ability to capture key pathological features. Particularly when the number of instances within a bag is large and their features are complex, it becomes challenging to accurately identify instances decisive for the bag label, making these methods prone to interference from ambiguous instances. To address this limitation, we propose a novel Histomorphology-Guided Prototypical Multi-Instance Learning (HGPMIL) framework that explicitly learns histomorphology-guided prototypical representations by incorporating tumor cellularity, cellular morphology, and tissue architecture. Specifically, our approach consists of three key components: (1) estimating the importance of tumor-related histomorphology information at patch-level based on medical prior knowledge; (2) generating representative prototypes through histomorphology-prototypical clustering; and (3) enabling WSI classification through histomorphology-guided prototypical aggregation. HGPMIL adjusts the decision boundary by incorporating histomorphological importance to reduce instance label uncertainty, thereby reversely optimizing the bag-level boundary. Experimental results demonstrate its effectiveness, achieving high diagnostic accuracy for molecular subtyping, cancer subtyping and survival analysis. The code will be made available at https://github.com/Badgewho/HMDMIL.

cs.CV↗

arXiv2025

Signal: Selective Interaction and Global-local Alignment for Multi-Modal Object Re-Identification

Multi-modal object Re-IDentification (ReID) is devoted to retrieving specific objects through the exploitation of complementary multi-modal image information. Existing methods mainly concentrate on the fusion of multi-modal features, yet neglecting the background interference. Besides, current multi-modal fusion methods often focus on aligning modality pairs but suffer from multi-modal consistency alignment. To address these issues, we propose a novel selective interaction and global-local alignment framework called Signal for multi-modal object ReID. Specifically, we first propose a Selective Interaction Module (SIM) to select important patch tokens with intra-modal and inter-modal information. These important patch tokens engage in the interaction with class tokens, thereby yielding more discriminative features. Then, we propose a Global Alignment Module (GAM) to simultaneously align multi-modal features by minimizing the volume of 3D polyhedra in the gramian space. Meanwhile, we propose a Local Alignment Module (LAM) to align local features in a shift-aware manner. With these modules, our proposed framework could extract more discriminative features for object ReID. Extensive experiments on three multi-modal object ReID benchmarks (i.e., RGBNT201, RGBNT100, MSVR310) validate the effectiveness of our method. The source code is available at https://github.com/010129/Signal.

cs.CV↗

arXiv2025

CADTrack: Learning Contextual Aggregation with Deformable Alignment for Robust RGBT Tracking

RGB-Thermal (RGBT) tracking aims to exploit visible and thermal infrared modalities for robust all-weather object tracking. However, existing RGBT trackers struggle to resolve modality discrepancies, which poses great challenges for robust feature representation. This limitation hinders effective cross-modal information propagation and fusion, which significantly reduces the tracking accuracy. To address this limitation, we propose a novel Contextual Aggregation with Deformable Alignment framework called CADTrack for RGBT Tracking. To be specific, we first deploy the Mamba-based Feature Interaction (MFI) that establishes efficient feature interaction via state space models. This interaction module can operate with linear complexity, reducing computational cost and improving feature discrimination. Then, we propose the Contextual Aggregation Module (CAM) that dynamically activates backbone layers through sparse gating based on the Mixture-of-Experts (MoE). This module can encode complementary contextual information from cross-layer features. Finally, we propose the Deformable Alignment Module (DAM) to integrate deformable sampling and temporal propagation, mitigating spatial misalignment and localization drift. With the above components, our CADTrack achieves robust and accurate tracking in complex scenarios. Extensive experiments on five RGBT tracking benchmarks verify the effectiveness of our proposed method. The source code is released at https://github.com/IdolLab/CADTrack.

cs.CV↗

arXiv2025

A Respiratory Motion Analysis for Guiding Stereotactic Arrhythmia Radiotherapy Motion Management

Stereotactic Arrhythmia Radiotherapy (STAR) treats ventricular tachycardia (VT) but requires internal target volume (ITV) expansions to compensate for cardiorespiratory motion. Current clinical r4DCT imaging methods are limited, and the reconstructed r4DCTs suffer from unmanaged cardiac motion artifacts that affect the quantitative assessment of respiratory motion. A groupwise surface-to-surface deformable image registration (DIR) algorithm, named gCGF, was developed. A novel principal component filtering (PCF) mechanism and a spatial smoothing mechanism were developed and incorporated into gCGF to iteratively register heart contours from an average respiratory-phase CT to ten r4DCT phases while removing random cardiac motion from the cyclic respiratory motion. The performance of the groupwise DIR was quantitatively validated using 8 digital phantoms with simulated cardiac artifacts. An ablation study was conducted to compare gCGF to another comparable state-of-the-art groupwise DIR method. gCGF was applied to r4DCTs of 20 STAR patients to analyze the respiratory motion of the heart. Validation on digital phantoms showed that gCGF achieved a mean target registration error of 0.63+-0.51 mm while successfully achieving phase smoothness and reducing cardiac motion artifacts. Among all STAR patients, the heart's maximum and mean respiratory motion magnitudes ranged from 3.6 to 7.9 mm and 1.0 mm to 2.6 mm. The peak-to-peak motion range was from 6.2 to 14.7 mm. For VT targets, the max and mean motion magnitude ranges were 3.0 to 6.7 mm and 0.8 to 2.9 mm, respectively. The peak-to-peak range was from 4.7 to 11.8 mm. Significant dominance of the first principal component of the motion direction was observed (p = 0).

physics.med-ph↗

arXiv2025

VocalBench-zh: Decomposing and Benchmarking the Speech Conversational Abilities in Mandarin Context

The development of multi-modal large language models (LLMs) leads to intelligent approaches capable of speech interactions. As one of the most widely spoken languages globally, Mandarin is supported by most models to enhance their applicability and reach. However, the scarcity of comprehensive speech-to-speech (S2S) benchmarks in Mandarin contexts impedes systematic evaluation for developers and hinders fair model comparison for users. In this work, we propose VocalBench-zh, an ability-level divided evaluation suite adapted to Mandarin context consisting of 10 well-crafted subsets and over 10K high-quality instances, covering 12 user-oriented characters. The evaluation experiment on 14 mainstream models reveals the common challenges for current routes, and highlights the need for new insights into next-generation speech interactive systems. The evaluation codes and datasets will be available at https://github.com/SJTU-OmniAgent/VocalBench-zh.

cs.CL↗

arXiv2025

Emotion and Intention Guided Multi-Modal Learning for Sticker Response Selection

Stickers are widely used in online communication to convey emotions and implicit intentions. The Sticker Response Selection (SRS) task aims to select the most contextually appropriate sticker based on the dialogue. However, existing methods typically rely on semantic matching and model emotional and intentional cues separately, which can lead to mismatches when emotions and intentions are misaligned. To address this issue, we propose Emotion and Intention Guided Multi-Modal Learning (EIGML). This framework is the first to jointly model emotion and intention, effectively reducing the bias caused by isolated modeling and significantly improving selection accuracy. Specifically, we introduce Dual-Level Contrastive Framework to perform both intra-modality and inter-modality alignment, ensuring consistent representation of emotional and intentional features within and across modalities. In addition, we design an Intention-Emotion Guided Multi-Modal Fusion module that integrates emotional and intentional information progressively through three components: Emotion-Guided Intention Knowledge Selection, Intention-Emotion Guided Attention Fusion, and Similarity-Adjusted Matching Mechanism. This design injects rich, effective information into the model and enables a deeper understanding of the dialogue, ultimately enhancing sticker selection performance. Experimental results on two public SRS datasets show that EIGML consistently outperforms state-of-the-art baselines, achieving higher accuracy and a better understanding of emotional and intentional features. Code is provided in the supplementary materials.

cs.LG↗

arXiv2025

VocalNet-M2: Advancing Low-Latency Spoken Language Modeling via Integrated Multi-Codebook Tokenization and Multi-Token Prediction

Current end-to-end spoken language models (SLMs) have made notable progress, yet they still encounter considerable response latency. This delay primarily arises from the autoregressive generation of speech tokens and the reliance on complex flow-matching models for speech synthesis. To overcome this, we introduce VocalNet-M2, a novel low-latency SLM that integrates a multi-codebook tokenizer and a multi-token prediction (MTP) strategy. Our model directly generates multi-codebook speech tokens, thus eliminating the need for a latency-inducing flow-matching model. Furthermore, our MTP strategy enhances generation efficiency and improves overall performance. Extensive experiments demonstrate that VocalNet-M2 achieves a substantial reduction in first chunk latency (from approximately 725ms to 350ms) while maintaining competitive performance across mainstream SLMs. This work also provides a comprehensive comparison of single-codebook and multi-codebook strategies, offering valuable insights for developing efficient and high-performance SLMs for real-time interactive applications.

cs.CL↗

arXiv2025

Efficient Reasoning via Reward Model

Reinforcement learning with verifiable rewards (RLVR) has been shown to enhance the reasoning capabilities of large language models (LLMs), enabling the development of large reasoning models (LRMs). However, LRMs such as DeepSeek-R1 and OpenAI o1 often generate verbose responses containing redundant or irrelevant reasoning step-a phenomenon known as overthinking-which substantially increases computational costs. Prior efforts to mitigate this issue commonly incorporate length penalties into the reward function, but we find they frequently suffer from two critical issues: length collapse and training collapse, resulting in sub-optimal performance. To address them, we propose a pipeline for training a Conciseness Reward Model (CRM) that scores the conciseness of reasoning path. Additionally, we introduce a novel reward formulation named Conciseness Reward Function (CRF) with explicit dependency between the outcome reward and conciseness score, thereby fostering both more effective and more efficient reasoning. From a theoretical standpoint, we demonstrate the superiority of the new reward from the perspective of variance reduction and improved convergence properties. Besides, on the practical side, extensive experiments on five mathematical benchmark datasets demonstrate the method's effectiveness and token efficiency, which achieves an 8.1% accuracy improvement and a 19.9% reduction in response token length on Qwen2.5-7B. Furthermore, the method generalizes well to other LLMs including Llama and Mistral. The implementation code and datasets are publicly available for reproduction: https://anonymous.4open.science/r/CRM.

cs.AI↗

arXiv2025

PEOD: A Pixel-Aligned Event-RGB Benchmark for Object Detection under Challenging Conditions

Robust object detection for challenging scenarios increasingly relies on event cameras, yet existing Event-RGB datasets remain constrained by sparse coverage of extreme conditions and low spatial resolution (<= 640 x 480), which prevents comprehensive evaluation of detectors under challenging scenarios. To address these limitations, we propose PEOD, the first large-scale, pixel-aligned and high-resolution (1280 x 720) Event-RGB dataset for object detection under challenge conditions. PEOD contains 130+ spatiotemporal-aligned sequences and 340k manual bounding boxes, with 57% of data captured under low-light, overexposure, and high-speed motion. Furthermore, we benchmark 14 methods across three input configurations (Event-based, RGB-based, and Event-RGB fusion) on PEOD. On the full test set and normal subset, fusion-based models achieve the excellent performance. However, in illumination challenge subset, the top event-based model outperforms all fusion models, while fusion models still outperform their RGB-based counterparts, indicating limits of existing fusion methods when the frame modality is severely degraded. PEOD establishes a realistic, high-quality benchmark for multimodal perception and facilitates future research.

cs.CV↗

arXiv2025

BEE-RAG: Balanced Entropy Engineering for Retrieval-Augmented Generation

With the rapid advancement of large language models (LLMs), retrieval-augmented generation (RAG) has emerged as a critical approach to supplement the inherent knowledge limitations of LLMs. However, due to the typically large volume of retrieved information, RAG tends to operate with long context lengths. From the perspective of entropy engineering, we identify unconstrained entropy growth and attention dilution due to long retrieval context as significant factors affecting RAG performance. In this paper, we propose the balanced entropy-engineered RAG (BEE-RAG) framework, which improves the adaptability of RAG systems to varying context lengths through the principle of entropy invariance. By leveraging balanced context entropy to reformulate attention dynamics, BEE-RAG separates attention sensitivity from context length, ensuring a stable entropy level. Building upon this, we introduce a zero-shot inference strategy for multi-importance estimation and a parameter-efficient adaptive fine-tuning mechanism to obtain the optimal balancing factor for different settings. Extensive experiments across multiple RAG tasks demonstrate the effectiveness of BEE-RAG.

cs.CL↗

arXiv2025

Uncovering Pretraining Code in LLMs: A Syntax-Aware Attribution Approach

As large language models (LLMs) become increasingly capable, concerns over the unauthorized use of copyrighted and licensed content in their training data have grown, especially in the context of code. Open-source code, often protected by open source licenses (e.g, GPL), poses legal and ethical challenges when used in pretraining. Detecting whether specific code samples were included in LLM training data is thus critical for transparency, accountability, and copyright compliance. We propose SynPrune, a syntax-pruned membership inference attack method tailored for code. Unlike prior MIA approaches that treat code as plain text, SynPrune leverages the structured and rule-governed nature of programming languages. Specifically, it identifies and excludes consequent tokens that are syntactically required and not reflective of authorship, from attribution when computing membership scores. Experimental results show that SynPrune consistently outperforms the state-of-the-arts. Our method is also robust across varying function lengths and syntax categories.

cs.CR↗

arXiv2025

LATex: Leveraging Attribute-based Text Knowledge for Aerial-Ground Person Re-Identification

As an important task in intelligent transportation systems, Aerial-Ground person Re-IDentification (AG-ReID) aims to retrieve specific persons across heterogeneous cameras in different viewpoints. Previous methods typically adopt deep learning-based models, focusing on extracting view-invariant features. However, they usually overlook the semantic information in person attributes. In addition, existing training strategies often rely on full fine-tuning large-scale models, which significantly increases training costs. To address these issues, we propose a novel framework named LATex for AG-ReID, which adopts prompt-tuning strategies to leverage attribute-based text knowledge. Specifically, with the Contrastive Language-Image Pre-training (CLIP) model, we first propose an Attribute-aware Image Encoder (AIE) to extract both global semantic features and attribute-aware features from input images. Then, with these features, we propose a Prompted Attribute Classifier Group (PACG) to predict person attributes and obtain attribute representations. Finally, we design a Coupled Prompt Template (CPT) to transform attribute representations and view information into structured sentences. These sentences are processed by the text encoder of CLIP to generate more discriminative features. As a result, our framework can fully leverage attribute-based text knowledge to improve AG-ReID performance. Extensive experiments on three AG-ReID benchmarks demonstrate the effectiveness of our proposed methods. The source code is available at https://github.com/kevinhu314/LATex.

cs.CV↗