arXiv subjects

Ge Gao

Publications and source records attributed to Ge Gao.

At least 37 records · Page 2Linked to original sources

arXiv2026

Zero-Shot Stance Detection in the Wild: Dynamic Target Generation and Multi-Target Adaptation

Current stance detection research typically relies on predicting stance based on given targets and text. However, in real-world social media scenarios, targets are neither predefined nor static but rather complex and dynamic. To address this challenge, we propose a novel task: zero-shot stance detection in the wild with Dynamic Target Generation and Multi-Target Adaptation (DGTA), which aims to automatically identify multiple target-stance pairs from text without prior target knowledge. We construct a Chinese social media stance detection dataset and design multi-dimensional evaluation metrics. We explore both integrated and two-stage fine-tuning strategies for large language models (LLMs) and evaluate various baseline models. Experimental results demonstrate that fine-tuned LLMs achieve superior performance on this task: the two-stage fine-tuned Qwen2.5-7B attains the highest comprehensive target recognition score of 66.99%, while the integrated fine-tuned DeepSeek-R1-Distill-Qwen-7B achieves a stance detection F1 score of 79.26%.

cs.CL↗

arXiv2026

Principled Fine-tuning of LLMs from User-Edits: A Medley of Preference, Supervision, and Reward

We study how to fine-tune LLMs using user-edit deployment data consisting of a set of context, an agent's response, and user edits. This deployment data is naturally generated by users in applications such as LLMs-based writing assistants and coding agents. The _natural_ origin of user edits makes it a desired source for adapting and personalizing LLMs. In this setup, there emerges a unification of various feedback types namely preferences, supervised labels, and cost that are typically studied separately in the literature. In this paper, we initiate the theoretical investigation of learning from user edits. We first derive bounds for learning algorithms that learn from each of these feedback types. We prove that these algorithms have different trade-offs depending upon the user, data distribution, and model class. We then propose a simple ensembling procedure to jointly learn from these feedback types. On two domains adapted from Gao et al. 2024, we show our ensembling procedure outperforms these methods that learn from individual feedback. Further, we show that our proposed procedure can robustly adapt to different user-edit distributions at test time.

cs.LG↗

arXiv2025

Ultra-lightweight Neural Video Representation Compression

Recent works have demonstrated the viability of utilizing over-fitted implicit neural representations (INRs) as alternatives to autoencoder-based models for neural video compression. Among these INR-based video codecs, Neural Video Representation Compression (NVRC) was the first to adopt a fully end-to-end compression framework that compresses INRs, achieving state-of-the-art performance. Moreover, some recently proposed lightweight INRs have shown comparable performance to their baseline codecs with computational complexity lower than 10kMACs/pixel. In this work, we extend NVRC toward lightweight representations, and propose NVRC-Lite, which incorporates two key changes. Firstly, we integrated multi-scale feature grids into our lightweight neural representation, and the use of higher resolution grids significantly improves the performance of INRs at low complexity. Secondly, we address the issue that existing INRs typically leverage autoregressive models for entropy coding: these are effective but impractical due to their slow coding speed. In this work, we propose an octree-based context model for entropy coding high-dimensional feature grids, which accelerates the entropy coding module of the model. Our experimental results demonstrate that NVRC-Lite outperforms C3, one of the best lightweight INR-based video codecs, with up to 21.03% and 23.06% BD-rate savings when measured in PSNR and MS-SSIM, respectively, while achieving 8.4x encoding and 2.5x decoding speedup. The implementation of NVRC-Lite will be made available.

cs.CV↗

arXiv2025

NVRC: Neural Video Representation Compression

Recent advances in implicit neural representation (INR)-based video coding have demonstrated its potential to compete with both conventional and other learning-based approaches. With INR methods, a neural network is trained to overfit a video sequence, with its parameters compressed to obtain a compact representation of the video content. However, although promising results have been achieved, the best INR-based methods are still out-performed by the latest standard codecs, such as VVC VTM, partially due to the simple model compression techniques employed. In this paper, rather than focusing on representation architectures as in many existing works, we propose a novel INR-based video compression framework, Neural Video Representation Compression (NVRC), targeting compression of the representation. Based on the novel entropy coding and quantization models proposed, NVRC, for the first time, is able to optimize an INR-based video codec in a fully end-to-end manner. To further minimize the additional bitrate overhead introduced by the entropy models, we have also proposed a new model compression framework for coding all the network, quantization and entropy model parameters hierarchically. Our experiments show that NVRC outperforms many conventional and learning-based benchmark codecs, with a 24% average coding gain over VVC VTM (Random Access) on the UVG dataset, measured in PSNR. As far as we are aware, this is the first time an INR-based video codec achieving such performance. The implementation of NVRC will be released.

cs.CV↗

arXiv2025

GFix: Perceptually Enhanced Gaussian Splatting Video Compression

3D Gaussian Splatting (3DGS) enhances 3D scene reconstruction through explicit representation and fast rendering, demonstrating potential benefits for various low-level vision tasks, including video compression. However, existing 3DGS-based video codecs generally exhibit more noticeable visual artifacts and relatively low compression ratios. In this paper, we specifically target the perceptual enhancement of 3DGS-based video compression, based on the assumption that artifacts from 3DGS rendering and quantization resemble noisy latents sampled during diffusion training. Building on this premise, we propose a content-adaptive framework, GFix, comprising a streamlined, single-step diffusion model that serves as an off-the-shelf neural enhancer. Moreover, to increase compression efficiency, We propose a modulated LoRA scheme that freezes the low-rank decompositions and modulates the intermediate hidden states, thereby achieving efficient adaptation of the diffusion backbone with highly compressible updates. Experimental results show that GFix delivers strong perceptual quality enhancement, outperforming GSVC with up to 72.1% BD-rate savings in LPIPS and 21.4% in FID.

cs.CV↗

arXiv2025

EPAN: Robust Pedestrian Re-Identification via Enhanced Alignment Network for IoT Surveillance

Person re-identification (ReID) plays a pivotal role in computer vision, particularly in surveillance and security applications within IoT-enabled smart environments. This study introduces the Enhanced Pedestrian Alignment Network (EPAN), tailored for robust ReID across diverse IoT surveillance conditions. EPAN employs a dual-branch architecture to mitigate the impact of perspective and environmental changes, extracting alignment information under varying scales and viewpoints. Here, we demonstrate EPAN's strong feature extraction capabilities, achieving outstanding performance on the Inspection-Personnel dataset with a Rank-1 accuracy of 90.09% and a mean Average Precision (mAP) of 78.82%. This highlights EPAN's potential for real-world IoT applications, enabling effective and reliable person ReID across diverse cameras in surveillance and security systems. The code and data are available at: https://github.com/ggboy2580/EPAN

cs.CV↗

arXiv2025

PCD-ReID: Occluded Person Re-Identification for Base Station Inspection

Occluded pedestrian re-identification (ReID) in base station environments is a critical task in computer vision, particularly for surveillance and security applications. This task faces numerous challenges, as occlusions often obscure key body features, increasing the complexity of identification. Traditional ResNet-based ReID algorithms often fail to address occlusions effectively, necessitating new ReID methods. We propose the PCD-ReID (Pedestrian Component Discrepancy) algorithm to address these issues. The contributions of this work are as follows: To tackle the occlusion problem, we design a Transformer-based PCD network capable of extracting shared component features, such as helmets and uniforms. To mitigate overfitting on public datasets, we collected new real-world patrol surveillance images for model training, covering six months, 10,000 individuals, and over 50,000 images. Comparative experiments with existing ReID algorithms demonstrate that our model achieves a mean Average Precision (mAP) of 79.0% and a Rank-1 accuracy of 82.7%, marking a 15.9% Rank-1 improvement over ResNet50-based methods. Experimental evaluations indicate that PCD-ReID effectively achieves occlusion-aware ReID performance for personnel in tower inspection scenarios, highlighting its potential for practical deployment in surveillance and security applications.

cs.CV↗

arXiv2025

Vision Transformer for Robust Occluded Person Reidentification in Complex Surveillance Scenes

Person re-identification (ReID) in surveillance is challenged by occlusion, viewpoint distortion, and poor image quality. Most existing methods rely on complex modules or perform well only on clear frontal images. We propose Sh-ViT (Shuffling Vision Transformer), a lightweight and robust model for occluded person ReID. Built on ViT-Base, Sh-ViT introduces three components: First, a Shuffle module in the final Transformer layer to break spatial correlations and enhance robustness to occlusion and blur; Second, scenario-adapted augmentation (geometric transforms, erasing, blur, and color adjustment) to simulate surveillance conditions; Third, DeiT-based knowledge distillation to improve learning with limited labels.To support real-world evaluation, we construct the MyTT dataset, containing over 10,000 pedestrians and 30,000+ images from base station inspections, with frequent equipment occlusion and camera variations. Experiments show that Sh-ViT achieves 83.2% Rank-1 and 80.1% mAP on MyTT, outperforming CNN and ViT baselines, and 94.6% Rank-1 and 87.5% mAP on Market1501, surpassing state-of-the-art methods.In summary, Sh-ViT improves robustness to occlusion and blur without external modules, offering a practical solution for surveillance-based personnel monitoring.

cs.CV↗

arXiv2025

From Unaligned to Aligned: Scaling Multilingual LLMs with Multi-Way Parallel Corpora

Continued pretraining and instruction tuning on large-scale multilingual data have proven to be effective in scaling large language models (LLMs) to low-resource languages. However, the unaligned nature of such data limits its ability to effectively capture cross-lingual semantics. In contrast, multi-way parallel data, where identical content is aligned across multiple languages, provides stronger cross-lingual consistency and offers greater potential for improving multilingual performance. In this paper, we introduce a large-scale, high-quality multi-way parallel corpus, TED2025, based on TED Talks. The corpus spans 113 languages, with up to 50 languages aligned in parallel, ensuring extensive multilingual coverage. Using this dataset, we investigate best practices for leveraging multi-way parallel data to enhance LLMs, including strategies for continued pretraining, instruction tuning, and the analysis of key influencing factors. Experiments on six multilingual benchmarks show that models trained on multiway parallel data consistently outperform those trained on unaligned multilingual data.

cs.CL↗

arXiv2025

Toward Machine Translation Literacy: How Lay Users Perceive and Rely on Imperfect Translations

As Machine Translation (MT) becomes increasingly commonplace, understanding how the general public perceives and relies on imperfect MT is crucial for contextualizing MT research in real-world applications. We present a human study conducted in a public museum (n=452), investigating how fluency and adequacy errors impact bilingual and non-bilingual users' reliance on MT during casual use. Our findings reveal that non-bilingual users often over-rely on MT due to a lack of evaluation strategies and alternatives, while experiencing the impact of errors can prompt users to reassess future reliance. This highlights the need for MT evaluation and NLP explanation techniques to promote not only MT quality, but also MT literacy among its users.

cs.CL↗

arXiv2025

Verisimilitude as Boon and Bane: How People Initiate Opportunistic Interactions at Professional Events in Social VR

Opportunistic interactions-the unstructured exchanges that emerge as individuals become aware of each other's presence-are essential for relationship building and information sharing in everyday life. Yet, fostering effective opportunistic interactions has proven challenging, especially at professional events that have increasingly transitioned from in person to online formats. In the current paper, we offer an in-depth qualitative account of how people initiate opportunistic interactions in social VR. Our participants consisted of 16 individuals with ongoing experience attending VR-mediated events in their professional communities. We conducted extensive observations with each participant during one or more events they attended. We also interviewed them after every observed event, obtaining self-reflections on their attempts to navigate opportunistic interactions with others. Our analysis revealed that participants sought to understand the extent to which social VR preserved the real-world meanings of various nonverbal cues, which we refer to as verisimilitude. We detailed the unique connections between a person's perceived verisimilitude and their social behaviors at each of the three steps toward initiating opportunistic interactions: availability recognition, attention capture, and ice-breaking. Across these steps, the VR platform typically replaces complex social mechanisms with feasible technical ones in order to function, thereby altering the preconditions necessary for a nonverbal cue's social meanings to remain intact. We identified a rich set of strategies that participants developed to assess verisimilitude and act upon it, while also confirming a lack of systematic knowledge guiding their practices. Based on these findings, we provide actionable insights for social VR platform design that can best support the initiation of opportunistic interactions for professional purposes.

cs.HC↗

arXiv2025

Ethical Considerations for Observational Research in Social VR

Social VR introduces new ethical challenges for observational research. The current paper presents a narrative literature review of ethical considerations in observational methods, with a focus on work in HCI. We examine how unobtrusive or selectively disclosed observation is implemented in public face-to-face and social VR settings. Our review extends ethical discussions from traditional public research into the context of social VR, highlighting tensions between observer visibility, data traceability, and participant autonomy. Drawing on insights distilled from prior literature, we propose five constructive guidelines for ethical observational research in public social VR environments. Our work offers key implications for future research, addressing anticipated improvements in platform design, the management of researcher presence, and the development of community-informed consent mechanisms.

cs.HC↗

arXiv2025

High-Resolution Quantum Sensing with Rydberg Atomic Receiver: Principles, Experiments and Future Prospects

Quantum sensing using Rydberg atoms offers unprecedented opportunities for next-generation radar systems, transcending classical limitations in miniaturization and spectral agility. Implementing this paradigm for radar sensing, this work proposes a quantum-enhanced radar reception architecture enabled by the emerging Rydberg atomic receiver, replacing conventional antenna-to-mixer chains with a centimeter-scale vapor cell. The proposed approach is based on electromagnetically induced transparency with the Autler-Townes splitting enabling direct RF-to-optical downconversion within the atomic medium via an external co-frequency reference. To circumvent the intrinsic bottleneck on instantaneous bandwidth of atomic receiver, we invoke a non-uniform stepped-frequency synthesis strategy combining coarse laser frequency tuning with fine AC-Stark shift compensation. Additionally, we establish a nonlinear response model of the Rydberg atomic homodyne receiver and propose a customized nonlinear compensation method that extends the linear dynamic range by over 7 dB. We develop a compressive sensing algorithm (CS-Rydberg) to suppress noise and mitigate the undersampling problem. Experimentally, we demonstrate a compact prototype achieving centimeter-level ranging precision (RMSE = 1.06 cm) within 1.6-1.9 m. By synthesizing GHz-bandwidth (2.6-3.6 GHz), resolvable target separations down to 15 cm are observed under controlled sparse scenarios. These results not only validate the feasibility of quantum sensing based on Rydberg atomic receivers but also underscore the architecture's inherent scalability: by harnessing the atoms' ultra-broad spectral response, the synthesized bandwidth can be extended well beyond the current range, enabling sub-centimeter resolution in future radar systems while preserving quantum-traceable calibration and a highly simplified front end.

physics.atom-ph↗

arXiv2025

An Interdisciplinary Approach to Human-Centered Machine Translation

Machine Translation (MT) tools are widely used today, often in contexts where professional translators are not present. Despite progress in MT technology, a gap persists between system development and real-world usage, particularly for non-expert users who may struggle to assess translation reliability. This paper advocates for a human-centered approach to MT, emphasizing the alignment of system design with diverse communicative goals and contexts of use. We survey the literature in Translation Studies and Human-Computer Interaction to recontextualize MT evaluation and design to address the diverse real-world scenarios in which MT is used today.

cs.CL↗

arXiv2025

ViVo: A Dataset for Volumetric Video Reconstruction and Compression

As research on neural volumetric video reconstruction and compression flourishes, there is a need for diverse and realistic datasets, which can be used to develop and validate reconstruction and compression models. However, existing volumetric video datasets lack diverse content in terms of both semantic and low-level features that are commonly present in real-world production pipelines. In this context, we propose a new dataset, ViVo, for VolumetrIc VideO reconstruction and compression. The dataset is faithful to real-world volumetric video production and is the first dataset to extend the definition of diversity to include both human-centric characteristics (skin, hair, etc.) and dynamic visual phenomena (transparent, reflective, liquid, etc.). Each video sequence in this database contains raw data including fourteen multi-view RGB and depth video pairs, synchronized at 30FPS with per-frame calibration and audio data, and their associated 2-D foreground masks and 3-D point clouds. To demonstrate the use of this database, we have benchmarked three state-of-the-art (SotA) 3-D reconstruction methods and two volumetric video compression algorithms. The obtained results evidence the challenging nature of the proposed dataset and the limitations of existing datasets for both volumetric video reconstruction and compression tasks, highlighting the need to develop more effective algorithms for these applications. The database and the associated results are available at https://vivo-bvicr.github.io/

cs.CV↗

arXiv2025

The Hidden Language of Harm: Examining the Role of Emojis in Harmful Online Communication and Content Moderation

Social media platforms have become central to modern communication, yet they also harbor offensive content that challenges platform safety and inclusivity. While prior research has primarily focused on textual indicators of offense, the role of emojis, ubiquitous visual elements in online discourse, remains underexplored. Emojis, despite being rarely offensive in isolation, can acquire harmful meanings through symbolic associations, sarcasm, and contextual misuse. In this work, we systematically examine emoji contributions to offensive Twitter messages, analyzing their distribution across offense categories and how users exploit emoji ambiguity. To address this, we propose an LLM-powered, multi-step moderation pipeline that selectively replaces harmful emojis while preserving the tweet's semantic intent. Human evaluations confirm our approach effectively reduces perceived offensiveness without sacrificing meaning. Our analysis also reveals heterogeneous effects across offense types, offering nuanced insights for online communication and emoji moderation.

cs.CL↗

arXiv2025

Emojis Decoded: Leveraging ChatGPT for Enhanced Understanding in Social Media Communications

Emojis, which encapsulate semantics beyond mere words or phrases, have become prevalent in social network communications. This has spurred increasing scholarly interest in exploring their attributes and functionalities. However, emoji-related research and application face two primary challenges. First, researchers typically rely on crowd-sourcing to annotate emojis in order to understand their sentiments, usage intentions, and semantic meanings. Second, subjective interpretations by users can often lead to misunderstandings of emojis and cause the communication barrier. Large Language Models (LLMs) have achieved significant success in various annotation tasks, with ChatGPT demonstrating expertise across multiple domains. In our study, we assess ChatGPT's effectiveness in handling previously annotated and downstream tasks. Our objective is to validate the hypothesis that ChatGPT can serve as a viable alternative to human annotators in emoji research and that its ability to explain emoji meanings can enhance clarity and transparency in online communications. Our findings indicate that ChatGPT has extensive knowledge of emojis. It is adept at elucidating the meaning of emojis across various application scenarios and demonstrates the potential to replace human annotators in a range of tasks.

cs.CL↗

arXiv2025

GIViC: Generative Implicit Video Compression

While video compression based on implicit neural representations (INRs) has recently demonstrated great potential, existing INR-based video codecs still cannot achieve state-of-the-art (SOTA) performance compared to their conventional or autoencoder-based counterparts given the same coding configuration. In this context, we propose a Generative Implicit Video Compression framework, GIViC, aiming at advancing the performance limits of this type of coding methods. GIViC is inspired by the characteristics that INRs share with large language and diffusion models in exploiting long-term dependencies. Through the newly designed implicit diffusion process, GIViC performs diffusive sampling across coarse-to-fine spatiotemporal decompositions, gradually progressing from coarser-grained full-sequence diffusion to finer-grained per-token diffusion. A novel Hierarchical Gated Linear Attention-based transformer (HGLA), is also integrated into the framework, which dual-factorizes global dependency modeling along scale and sequential axes. The proposed GIViC model has been benchmarked against SOTA conventional and neural codecs using a Random Access (RA) configuration (YUV 4:2:0, GOPSize=32), and yields BD-rate savings of 15.94%, 22.46% and 8.52% over VVC VTM, DCVC-FM and NVRC, respectively. As far as we are aware, GIViC is the first INR-based video codec that outperforms VTM based on the RA coding configuration. The source code will be made available.

eess.IV↗