arXiv subjects

Wei Song

Publications and source records attributed to Wei Song.

At least 73 records · Page 4Linked to original sources

arXiv2026

LLM-Guided Knowledge Distillation for Temporal Knowledge Graph Reasoning

Temporal knowledge graphs (TKGs) support reasoning over time-evolving facts, yet state-of-the-art models are often computationally heavy and costly to deploy. Existing compression and distillation techniques are largely designed for static graphs; directly applying them to temporal settings may overlook time-dependent interactions and lead to performance degradation. We propose an LLM-assisted distillation framework specifically designed for temporal knowledge graph reasoning. Beyond a conventional high-capacity temporal teacher, we incorporate a large language model as an auxiliary instructor to provide enriched supervision. The LLM supplies broad background knowledge and temporally informed signals, enabling a lightweight student to better model event dynamics without increasing inference-time complexity. Training is conducted by jointly optimizing supervised and distillation objectives, using a staged alignment strategy to progressively integrate guidance from both teachers. Extensive experiments on multiple public TKG benchmarks with diverse backbone architectures demonstrate that the proposed approach consistently improves link prediction performance over strong distillation baselines, while maintaining a compact and efficient student model. The results highlight the potential of large language models as effective teachers for transferring temporal reasoning capability to resource-efficient TKG systems.

cs.CL↗

arXiv2026

DeepGen 1.0: A Lightweight Unified Multimodal Model for Advancing Image Generation and Editing

Current unified multimodal models for image generation and editing typically rely on massive parameter scales (e.g., >10B), entailing prohibitive training costs and deployment footprints. In this work, we present DeepGen 1.0, a lightweight 5B unified model that achieves comprehensive capabilities competitive with or surpassing much larger counterparts. To overcome the limitations of compact models in semantic understanding and fine-grained control, we introduce Stacked Channel Bridging (SCB), a deep alignment framework that extracts hierarchical features from multiple VLM layers and fuses them with learnable 'think tokens' to provide the generative backbone with structured, reasoning-rich guidance. We further design a data-centric training strategy spanning three progressive stages: (1) Alignment Pre-training on large-scale image-text pairs and editing triplets to synchronize VLM and DiT representations, (2) Joint Supervised Fine-tuning on a high-quality mixture of generation, editing, and reasoning tasks to foster omni-capabilities, and (3) Reinforcement Learning with MR-GRPO, which leverages a mixture of reward functions and supervision signals, resulting in substantial gains in generation quality and alignment with human preferences, while maintaining stable training progress and avoiding visual artifacts. Despite being trained on only ~50M samples, DeepGen 1.0 achieves leading performance across diverse benchmarks, surpassing the 80B HunyuanImage by 28% on WISE and the 27B Qwen-Image-Edit by 37% on UniREditBench. By open-sourcing our training code, weights, and datasets, we provide an efficient, high-performance alternative to democratize unified multimodal research.

cs.CV↗

arXiv2026

Can Distillation Mitigate Backdoor Attacks in Pre-trained Encoders?

Self-Supervised Learning (SSL) has become a prominent paradigm for pre-training encoders to learning general-purpose representations from unlabeled data and releasing them on third-party platforms for broad downstream deep learning tasks. However, SSL is vulnerable to backdoor attacks, where an adversary may train and distribute poisoned pre-training encoders to contaminate the downstream models. In this paper, we study a defense mechanism based on distillation against poisoned encoders in SSL. Traditionally, distillation transfers knowledge from a pre-trained teacher model to a student model, enabling the student to replicate or refine the teacher's learned representations. We repurpose distillation to extract benign knowledge and remove backdoors from a poisoned pre-trained encoder to produce a clean and reliable pre-trained model. We conduct extensive experiments to evaluate the effectiveness of distillation in mitigating backdoor attacks on pre-trained encoders. Based on two state-of-the-art backdoor attacks and four widely adopted image classification datasets, our results demonstrate that distillation reduces the attack success rate from 80.87% to 27.51%, with only a 6.35% drop in model accuracy. Furthermore, by comparing four teacher architectures, three student models, and six loss functions, we find that the distillation with fine-tuned teacher networks, warm-up-based student training, and attention-based distillation losses yield the best performance.

cs.LG↗

arXiv2026

FAIRT2V: Training-Free Debiasing for Text-to-Video Diffusion Models

Text-to-video (T2V) diffusion models have achieved rapid progress, yet their demographic biases, particularly gender bias, remain largely unexplored. We present FairT2V, a training-free debiasing framework for text-to-video generation that mitigates encoder-induced bias without finetuning. We first analyze demographic bias in T2V models and show that it primarily originates from pretrained text encoders, which encode implicit gender associations even for neutral prompts. We quantify this effect with a gender-leaning score that correlates with bias in generated videos. Based on this insight, FairT2V mitigates demographic bias by neutralizing prompt embeddings via anchor-based spherical geodesic transformations while preserving semantics. To maintain temporal coherence, we apply debiasing only during early identity-forming steps through a dynamic denoising schedule. We further propose a video-level fairness evaluation protocol combining VideoLLM-based reasoning with human verification. Experiments on the modern T2V model Open-Sora show that FairT2V substantially reduces demographic bias across occupations with minimal impact on video quality.

cs.CV↗

arXiv2026

DeMark: A Query-Free Black-Box Attack on Deepfake Watermarking Defenses

The rapid proliferation of realistic deepfakes has raised urgent concerns over their misuse, motivating the use of defensive watermarks in synthetic images for reliable detection and provenance tracking. However, this defense paradigm assumes such watermarks are inherently resistant to removal. We challenge this assumption with DeMark, a query-free black-box attack framework that targets defensive image watermarking schemes for deepfakes. DeMark exploits latent-space vulnerabilities in encoder-decoder watermarking models through a compressive sensing based sparsification process, suppressing watermark signals while preserving perceptual and structural realism appropriate for deepfakes. Across eight state-of-the-art watermarking schemes, DeMark reduces watermark detection accuracy from 100% to 32.9% on average while maintaining natural visual quality, outperforming existing attacks. We further evaluate three defense strategies, including image super resolution, sparse watermarking, and adversarial training, and find them largely ineffective. These results demonstrate that current encoder decoder watermarking schemes remain vulnerable to latent-space manipulations, underscoring the need for more robust watermarking methods to safeguard against deepfakes.

cs.CR↗

arXiv2026

Fuzzychain-edge: A novel Fuzzy logic-based adaptive Access control model for Blockchain in Edge Computing

The rapid integration of IoT with edge computing has revolutionized various domains, particularly healthcare, by enabling real-time data sharing, remote monitoring, and decision-making. However, it introduces critical challenges, including data privacy breaches, security vulnerabilities, especially in environments dealing with sensitive information. Traditional access control mechanisms and centralized security systems do not address these issues, leaving IoT environments exposed to unauthorized access and data misuse. This research proposes Fuzzychain-edge, a novel Fuzzy logic-based adaptive Access control model for Blockchain in Edge Computing framework designed to overcome these limitations by incorporating Zero-Knowledge Proofs (ZKPs), fuzzy logic, and smart contracts. ZKPs secure sensitive data during access control processes by enabling verification without revealing confidential details, thereby ensuring user privacy. Fuzzy logic facilitates adaptive, context-aware decision-making for access control by dynamically evaluating parameters such as data sensitivity, trust levels, and user roles. Blockchain technology, with its decentralized and immutable architecture, ensures transparency, traceability, and accountability using smart contracts that automate access control processes. The proposed framework addresses key challenges by enhancing security, reducing the likelihood of unauthorized access, and providing a transparent audit trail of data transactions. Expected outcomes include improved data privacy, accuracy in access control, and increased user trust in IoT systems. This research contributes significantly to advancing privacy-preserving, secure, and traceable solutions in IoT environments, laying the groundwork for future innovations in decentralized technologies and their applications in critical domains such as healthcare and beyond.

cs.CR↗

arXiv2026

Robust CAPTCHA Using Audio Illusions in the Era of Large Language Models: from Evaluation to Advances

CAPTCHAs are widely used by websites to block bots and spam by presenting challenges that are easy for humans but difficult for automated programs to solve. To improve accessibility, audio CAPTCHAs are designed to complement visual ones. However, the robustness of audio CAPTCHAs against advanced Large Audio Language Models (LALMs) and Automatic Speech Recognition (ASR) models remains unclear. In this paper, we introduce AI-CAPTCHA, a unified framework that offers (i) an evaluation framework, ACEval, which includes advanced LALM- and ASR-based solvers, and (ii) a novel audio CAPTCHA approach, IllusionAudio, leveraging audio illusions. Through extensive evaluations of seven widely deployed audio CAPTCHAs, we show that most existing methods can be solved with high success rates by advanced LALMs and ASR models, exposing critical security weaknesses. To address these vulnerabilities, we design a new audio CAPTCHA approach, IllusionAudio, which exploits perceptual illusion cues rooted in human auditory mechanisms. Extensive experiments demonstrate that our method defeats all tested LALM- and ASR-based attacks while achieving a 100% human pass rate, significantly outperforming existing audio CAPTCHA methods.

cs.SD↗

arXiv2026

Autoregressive Semantic Visual Reconstruction Helps VLMs Understand Better

Typical large vision-language models (LVLMs) apply autoregressive supervision solely to textual sequences, without fully incorporating the visual modality into the learning process. This results in three key limitations: (1) an inability to utilize images without accompanying captions, (2) the risk that captions omit critical visual details, and (3) the challenge that certain vision-centric content cannot be adequately conveyed through text. As a result, current LVLMs often prioritize vision-to-language alignment while potentially overlooking fine-grained visual information. While some prior works have explored autoregressive image generation, effectively leveraging autoregressive visual supervision to enhance image understanding remains an open challenge. In this paper, we introduce Autoregressive Semantic Visual Reconstruction (ASVR), which enables joint learning of visual and textual modalities within a unified autoregressive framework. We show that autoregressively reconstructing the raw visual appearance of images does not enhance and may even impair multimodal understanding. In contrast, autoregressively reconstructing the semantic representation of images consistently improves comprehension. Notably, we find that even when models are given continuous image features as input, they can effectively reconstruct discrete semantic tokens, resulting in stable and consistent improvements across a wide range of multimodal understanding benchmarks. Our approach delivers significant performance gains across varying data scales (556k-2M) and types of LLM bacbones. Specifically, ASVR improves LLaVA-1.5 by 5% in average scores across 14 multimodal benchmarks. The code is available at https://github.com/AlenjandroWang/ASVR.

cs.CV↗

arXiv2026

Knowledge Distillation for Temporal Knowledge Graph Reasoning with Large Language Models

Reasoning over temporal knowledge graphs (TKGs) is fundamental to improving the efficiency and reliability of intelligent decision-making systems and has become a key technological foundation for future artificial intelligence applications. Despite recent progress, existing TKG reasoning models typically rely on large parameter sizes and intensive computation, leading to high hardware costs and energy consumption. These constraints hinder their deployment on resource-constrained, low-power, and distributed platforms that require real-time inference. Moreover, most existing model compression and distillation techniques are designed for static knowledge graphs and fail to adequately capture the temporal dependencies inherent in TKGs, often resulting in degraded reasoning performance. To address these challenges, we propose a distillation framework specifically tailored for temporal knowledge graph reasoning. Our approach leverages large language models as teacher models to guide the distillation process, enabling effective transfer of both structural and temporal reasoning capabilities to lightweight student models. By integrating large-scale public knowledge with task-specific temporal information, the proposed framework enhances the student model's ability to model temporal dynamics while maintaining a compact and efficient architecture. Extensive experiments on multiple publicly available benchmark datasets demonstrate that our method consistently outperforms strong baselines, achieving a favorable trade-off between reasoning accuracy, computational efficiency, and practical deployability.

cs.CL↗

arXiv2025

NGCaptcha: A CAPTCHA Bridging the Past and the Future

CAPTCHAs are widely employed for distinguishing humans from automated bots online. However, current vision based CAPTCHAs face escalating security risks: traditional attacks continue to bypass many deployed CAPTCHA schemes, and recent breakthroughs in AI, particularly large scale vision models, enable machine solvers to significantly outperform humans on many CAPTCHA tasks, undermining their original design assumptions. To address these issues, we introduce NGCAPTCHA, a Next Generation CAPTCHA framework that integrates a lightweight client side proof of work (PoW) mechanism with an AI resistant visual recognition challenge. In NGCAPTCHA, a browser must first complete a small hash based PoW before any challenge is displayed, throttling large scale automated attempts by increasing their computational cost. Once the PoW is solved, the user is presented with a human friendly yet model resistant image selection task that exploits perceptual cues current vision systems still struggle with. This hybrid design combines computational friction with AI robust visual discrimination, substantially raising the barrier for automated bots while keeping the verification process fast and effortless for legitimate users.

cs.CY↗

arXiv2025

Towards Privacy-Preserving Range Queries with Secure Learned Spatial Index over Encrypted Data

With the growing reliance on cloud services for large-scale data management, preserving the security and privacy of outsourced datasets has become increasingly critical. While encrypting data and queries can prevent direct content exposure, recent research reveals that adversaries can still infer sensitive information via access pattern and search path analysis. However, existing solutions that offer strong access pattern privacy often incur substantial performance overhead. In this paper, we propose a novel privacy-preserving range query scheme over encrypted datasets, offering strong security guarantees while maintaining high efficiency. To achieve this, we develop secure learned spatial index (SLS-INDEX), a secure learned index that integrates the Paillier cryptosystem with a hierarchical prediction architecture and noise-injected buckets, enabling data-aware query acceleration in the encrypted domain. To further obfuscate query execution paths, SLS-INDEXbased Range Queries (SLRQ) employs a permutation-based secure bucket prediction protocol. Additionally, we introduce a secure point extraction protocol that generates candidate results to reduce the overhead of secure computation. We provide formal security analysis under realistic leakage functions and implement a prototype to evaluate its practical performance. Extensive experiments on both real-world and synthetic datasets demonstrate that SLRQ significantly outperforms existing solutions in query efficiency while ensuring dataset, query, result, and access pattern privacy.

cs.CR↗

arXiv2025

LLMs4All: A Review of Large Language Models Across Academic Disciplines

Cutting-edge Artificial Intelligence (AI) techniques keep reshaping our view of the world. For example, Large Language Models (LLMs) based applications such as ChatGPT have shown the capability of generating human-like conversation on extensive topics. Due to the impressive performance on a variety of language-related tasks (e.g., open-domain question answering, translation, and document summarization), one can envision the far-reaching impacts that can be brought by the LLMs with broader real-world applications (e.g., customer service, education and accessibility, and scientific discovery). Inspired by their success, this paper will offer an overview of state-of-the-art LLMs and their integration into a wide range of academic disciplines, including: (1) arts, letters, and law (e.g., history, philosophy, political science, arts and architecture, law), (2) economics and business (e.g., finance, economics, accounting, marketing), and (3) science and engineering (e.g., mathematics, physics and mechanical engineering, chemistry and chemical engineering, life sciences and bioengineering, earth sciences and civil engineering, computer science and electrical engineering). Integrating humanity and technology, in this paper, we will explore how LLMs are shaping research and practice in these fields, while also discussing key limitations, open challenges, and future directions in the era of generative AI. The review of how LLMs are engaged across disciplines-along with key observations and insights-can help researchers and practitioners interested in exploiting LLMs to advance their works in diverse real-world applications.

cs.CL↗

arXiv2025

Many-body electronic structure in pyrochlore superconductor CsBi2 and spin liquid Pr2Ir2O7

The pyrochlore lattice materials can exhibit geometrical frustration, while the related many-body electronic states remain elusive. In this work, we performed scanning tunneling microscopy measurements on the pyrochlore superconductor CsBi2 and spin liquid Pr2Ir2O7 at 0.3 K. For the first time, we obtained atomically resolved images of their (111) surfaces, revealing a hexagonal lattice or a kagome lattice. Tunneling spectroscopy in CsBi2 reveals a nearly fully opened superconductivity gap. The ratio of 2Δ/kBTC = 4.7 suggests relatively strong coupling superconductivity, as compared with that in kagome superconductors AV3Sb5 (A = K, Rb, Cs). In contrast to the previous study categorizing CsBi2 as a type-I superconductor, the applied magnetic field induces a hexagonal vortex lattice in which each vortex core exhibits an intriguing three-fold symmetry state. In Pr2Ir2O7, we observed a spatially homogeneous Kondo-lattice resonance, which is compared with that in the kagome Kondo-lattice material CsCr6Sb6. We further discover that the Kondo resonance exhibits a spatial modulation with three-fold symmetry, and the applied magnetic field induces a Zeeman splitting of the Kondo resonance with intriguing atomic site dependence. We discuss the relations of these many-body electronic phenomena with the pyrochlore lattice geometry and its charge or spin frustration. Our systematic observations offer atomic-scale insights into the many-body electronic structures of the geometrically frustrated pyrochlore superconductors and spin liquids.

cond-mat.str-el↗

arXiv2025

Failures to Surface Harmful Contents in Video Large Language Models

Video Large Language Models (VideoLLMs) are increasingly deployed on numerous critical applications, where users rely on auto-generated summaries while casually skimming the video stream. We show that this interaction hides a critical safety gap: if harmful content is embedded in a video, either as full-frame inserts or as small corner patches, state-of-the-art VideoLLMs rarely mention the harmful content in the output, despite its clear visibility to human viewers. A root-cause analysis reveals three compounding design flaws: (1) insufficient temporal coverage resulting from the sparse, uniformly spaced frame sampling used by most leading VideoLLMs, (2) spatial information loss introduced by aggressive token downsampling within sampled frames, and (3) encoder-decoder disconnection, whereby visual cues are only weakly utilized during text generation. Leveraging these insights, we craft three zero-query black-box attacks, aligning with these flaws in the processing pipeline. Our large-scale evaluation across five leading VideoLLMs shows that the harmfulness omission rate exceeds 90% in most cases. Even when harmful content is clearly present in all frames, these models consistently fail to identify it. These results underscore a fundamental vulnerability in current VideoLLMs' designs and highlight the urgent need for sampling strategies, token compression, and decoding mechanisms that guarantee semantic coverage rather than speed alone.

cs.MM↗

arXiv2025

Debiasing LLMs by Masking Unfairness-Driving Attention Heads

Large language models (LLMs) increasingly mediate decisions in domains where unfair treatment of demographic groups is unacceptable. Existing work probes when biased outputs appear, but gives little insight into the mechanisms that generate them, leaving existing mitigations largely fragile. In this paper, we conduct a systematic investigation LLM unfairness and propose DiffHeads, a lightweight debiasing framework for LLMs. We first compare Direct-Answer (DA) prompting to Chain-of-Thought (CoT) prompting across eight representative open- and closed-source LLMs. DA will trigger the nature bias part of LLM and improve measured unfairness by 534.5%-391.9% in both one-turn and two-turn dialogues. Next, we define a token-to-head contribution score that traces each token's influence back to individual attention heads. This reveals a small cluster of bias heads that activate under DA but stay largely dormant with CoT, providing the first causal link between prompting strategy and bias emergence. Finally, building on this insight, we propose DiffHeads that identifies bias heads through differential activation analysis between DA and CoT, and selectively masks only those heads. DiffHeads reduces unfairness by 49.4%, and 40.3% under DA and CoT, respectively, without harming model utility.

cs.CL↗

arXiv2025

Digital Contrast CT Pulmonary Angiography Synthesis from Non-contrast CT for Pulmonary Vascular Disease

Computed Tomography Pulmonary Angiography (CTPA) is the reference standard for diagnosing pulmonary vascular diseases such as Pulmonary Embolism (PE) and Chronic Thromboembolic Pulmonary Hypertension (CTEPH). However, its reliance on iodinated contrast agents poses risks including nephrotoxicity and allergic reactions, particularly in high-risk patients. This study proposes a method to generate Digital Contrast CTPA (DCCTPA) from Non-Contrast CT (NCCT) scans using a cascaded synthesizer based on Cycle-Consistent Generative Adversarial Networks (CycleGAN). Totally retrospective 410 paired CTPA and NCCT scans were obtained from three centers. The model was trained and validated internally on 249 paired images. Extra dataset that comprising 161 paired images was as test set for model generalization evaluation and downstream clinical tasks validation. Compared with state-of-the-art (SOTA) methods, the proposed method achieved the best comprehensive performance by evaluating quantitative metrics (For validation, MAE: 156.28, PSNR: 20.71 and SSIM: 0.98; For test, MAE: 165.12, PSNR: 20.27 and SSIM: 0.98) and qualitative visualization, demonstrating valid vessel enhancement, superior image fidelity and structural preservation. The approach was further applied to downstream tasks of pulmonary vessel segmentation and vascular quantification. On the test set, the average Dice, clDice, and clRecall of artery and vein pulmonary segmentation was 0.70, 0.71, 0.73 and 0.70, 0.72, 0.75 respectively, all markedly improved compared with NCCT inputs.\@ Inter-class Correlation Coefficient (ICC) for vessel volume between DCCTPA and CTPA was significantly better than that between NCCT and CTPA (Average ICC : 0.81 vs 0.70), indicating effective vascular enhancement in DCCTPA, especially for small vessels.

cs.CV↗

arXiv2025

The Early Bird Catches the Leak: Unveiling Timing Side Channels in LLM Serving Systems

The wide deployment of Large Language Models (LLMs) has given rise to strong demands for optimizing their inference performance. Today's techniques serving this purpose primarily focus on reducing latency and improving throughput through algorithmic and hardware enhancements, while largely overlooking their privacy side effects, particularly in a multi-user environment. In our research, for the first time, we discovered a set of new timing side channels in LLM systems, arising from shared caches and GPU memory allocations, which can be exploited to infer both confidential system prompts and those issued by other users. These vulnerabilities echo security challenges observed in traditional computing systems, highlighting an urgent need to address potential information leakage in LLM serving infrastructures. In this paper, we report novel attack strategies designed to exploit such timing side channels inherent in LLM deployments, specifically targeting the Key-Value (KV) cache and semantic cache widely used to enhance LLM inference performance. Our approach leverages timing measurements and classification models to detect cache hits, allowing an adversary to infer private prompts with high accuracy. We also propose a token-by-token search algorithm to efficiently recover shared prompt prefixes in the caches, showing the feasibility of stealing system prompts and those produced by peer users. Our experimental studies on black-box testing of popular online LLM services demonstrate that such privacy risks are completely realistic, with significant consequences. Our findings underscore the need for robust mitigation to protect LLM systems against such emerging threats.

cs.CR↗

arXiv2025

Integrating Various Software Artifacts for Better LLM-based Bug Localization and Program Repair

LLMs have garnered considerable attention for their potential to streamline Automated Program Repair (APR). LLM-based approaches can either insert the correct code or directly generate patches when provided with buggy methods. However, most of LLM-based APR methods rely on a single type of software information, without fully leveraging different software artifacts. Despite this, many LLM-based approaches do not explore which specific types of information best assist in APR. Addressing this gap is crucial for advancing LLM-based APR techniques. We propose DEVLoRe to use issue content (description and message) and stack error traces to localize buggy methods, then rely on debug information in buggy methods and issue content and stack error to localize buggy lines and generate plausible patches which can pass all unit tests. The results show that while issue content is particularly effective in assisting LLMs with fault localization and program repair, different types of software artifacts complement each other. By incorporating different artifacts, DEVLoRe successfully locates 49.3% and 47.6% of single and non-single buggy methods and generates 56.0% and 14.5% plausible patches for the Defects4J v2.0 dataset, respectively. This outperforms current state-of-the-art APR methods. Furthermore, we re-implemented and evaluated our framework, demonstrating its effectiveness in its effectiveness in resolving 9 unique issues compared to other state-of-the-art frameworks using the same or more advanced models on SWE-bench Lite.We also discussed whether a leading framework for Python code can be directly applied to Java code, or vice versa. The source code and experimental results of this work for replication are available at https://github.com/XYZboom/DEVLoRe.

cs.SE↗