arXiv subjects

Zecheng Zhao

Publications and source records attributed to Zecheng Zhao.

9 recordsLinked to original sources

arXiv2026

What Would You Click? Personalized Video Thumbnail Generation with Preference-aware Highlight Retrieval

Video thumbnails are a key factor for attracting user clicks on video platforms, and are increasingly supported by automation. However, existing thumbnail generation methods typically produce generic results shared across users, overlooking the diversity of individual preferences. We therefore introduce personalized video thumbnail generation, a novel task that aims to create thumbnails tailored to user-specific preferences. It is challenging in two aspects: (i) identifying visual anchors (i.e., key frames) from each video to guide the generation, which requires a balance between personalization and informativeness that existing highlight detection methods fail to achieve; and (ii) generating personalized thumbnails that are both visually coherent and faithful to the original video. As a response, we propose a two-stage framework that tightly couples preference-aware retrieval with controllable generation. In the first stage, a personalized highlight retriever captures fine-grained user-video interactions and incorporates video semantics through summarization, enabling the selection of diverse visual anchors aligned with both user preferences and video contexts. In the second stage, a VLM-guided diffusion pipeline transforms these anchors into thumbnails by extracting and injecting semantically grounded visual cues, improving personalization while preserving visual coherence and fidelity. Experiments on two public datasets show our method delivers state-of-the-art performance compared with both retrieval-based and generative baselines. A user study further demonstrates improved click preference, highlighting its effectiveness in enhancing user engagement. The code is available at https://github.com/hezy18/PVTG.

cs.MM↗

arXiv2026

Gauge-Invariant Non-Hermitian Quantum Theory: Foundation and Applications to Dynamical Phase Transitions

The description of states and dynamics in non-Hermitian systems is fundamentally linked to the choice of an appropriate theoretical framework -- a point of ongoing debate in the field. This work addresses this issue by proposing a consistent formulation that reconciles existing controversies and establishes a unified theoretical understanding. Our approach rests on two foundational premises: (i) the dynamics of both left and right-vectors of a non-Hermitian system must satisfy the Schrödinger equation; (ii) the theoretical framework must preserve gauge invariance, ensuring that physical quantities are independent of unobservable phase choices. Building on these physically motivated assumptions, we refine the biorthogonal framework, leading to a gauge-invariant non-Hermitian quantum theory. Our framework naturally encompasses the open-system effective non-Hermitian evolution as a special case, and can naturally reduce to standard quantum mechanics in the Hermitian limit. As a concrete application, we analyze the dynamical phase transition in a one-dimensional Su-Schrieffer-Heeger (SSH) model within this gauge-invariant non-Hermitian quantum theory. Notably, our formulation naturally generalizes the known condition for such transitions in Hermitian two-band systems, namely, $\mathbf{d}_{k}^i\cdot\mathbf{d}_{k}^f=0$, to the non-Hermitian case, where it takes the form $\mathrm{Re}\Bigl[\frac{\mathbf{d}_{k}^i}{d_{k}^i}\cdot\frac{\mathbf{d}_{k}^f}{d_{k}^f}\Bigr]=0$. Furthermore, we identify entirely new dynamical phase transitions that cannot be characterized by the winding number. We hope that this gauge-invariant non-Hermitian quantum theory will find broad applications in the study of non-Hermitian systems.

quant-ph↗

arXiv2026

Survival of Hermitian Criticality in the Non-Hermitian Framework

In this work, we investigate many-body phase transitions in a one-dimensional anisotropic XY model subject to a complex-valued transverse field. Within the biorthogonal framework, we calculate the ground-state correlation functions and entanglement entropy, confirming that their scaling behavior remains identical to that in the Hermitian XY model. The preservation of Hermitian phase transition features in the non-Hermitian setting is rooted in the persistence and emergence of symmetries and their breaking. Specifically, the ferromagnetic (FM) phase arises from the breaking of a $Z_2$ symmetry, while the Luttinger liquid (LL) phase is enabled by the emergence of a $U(1)$ symmetry together with the degeneracy of the real part of the energy spectrum. The nontrivial topology of the LL phase are characterized by the winding number around the exceptional point (EP). Given that non-Hermitian systems are inherently open, this research opens a new avenue for exploring conventional quantum phase transitions that are typically vulnerable to decoherence and environmental disruption in open quantum systems.

quant-ph↗

arXiv2026

Generative Recall, Dense Reranking: Learning Multi-View Semantic IDs for Efficient Text-to-Video Retrieval

Text-to-Video Retrieval (TVR) is essential in video platforms. Dense retrieval with dual-modality encoders leads in accuracy, but its computation and storage scale poorly with corpus size. Thus, real-time large-scale applications adopt two-stage retrieval, where a fast recall model gathers a small candidate pool, which is reranked by an advanced dense retriever. Due to hugely reduced candidates, the reranking model can use any off-the-shelf dense retriever without hurting efficiency, meaning the recall model bounds two-stage TVR performance. Recently, generative retrieval (GR) replaces dense video embeddings with discrete semantic IDs and retrieves by decoding text queries into ID tokens. GR offers near-constant inference and storage complexity, and its semantic IDs capture high-level video features via quantization, making it ideal for quickly eliminating irrelevant candidates during recall. However, as a recall model in two-stage TVR, GR suffers from (i) semantic ambiguity, where each video satisfies diverse queries but is forced into one semantic ID; and (ii) cross-modal misalignment, as semantic IDs are solely derived from visual features without text supervision. We propose Generative Recall and Dense Reranking (GRDR), designing a novel GR method to uplift recalled candidate quality. GRDR assigns multiple semantic IDs to each video using a query-guided multi-view tokenizer exposing diverse semantic access paths, and jointly trains the tokenizer and generative retriever via a shared codebook to cast semantic IDs as the semantic bridge between texts and videos. At inference, trie-constrained decoding generates a compact candidate set reranked by a dense model for fine-grained matching. Experiments on TVR benchmarks show GRDR matches strong dense retrievers in accuracy while reducing index storage by an order of magnitude and accelerating up to 300$\times$ in full-corpus retrieval.

cs.CV↗

arXiv2025

SVIP: Semantically Contextualized Visual Patches for Zero-Shot Learning

Zero-shot learning (ZSL) aims to recognize unseen classes without labeled training examples by leveraging class-level semantic descriptors such as attributes. A fundamental challenge in ZSL is semantic misalignment, where semantic-unrelated information involved in visual features introduce ambiguity to visual-semantic interaction. Unlike existing methods that suppress semantic-unrelated information post hoc either in the feature space or the model space, we propose addressing this issue at the input stage, preventing semantic-unrelated patches from propagating through the network. To this end, we introduce Semantically contextualized VIsual Patches (SVIP) for ZSL, a transformer-based framework designed to enhance visual-semantic alignment. Specifically, we propose a self-supervised patch selection mechanism that preemptively learns to identify semantic-unrelated patches in the input space. This is trained with the supervision from aggregated attention scores across all transformer layers, which estimate each patch's semantic score. As removing semantic-unrelated patches from the input sequence may disrupt object structure, we replace them with learnable patch embeddings. With initialization from word embeddings, we can ensure they remain semantically meaningful throughout feature extraction. Extensive experiments on ZSL benchmarks demonstrate that SVIP achieves state-of-the-art performance results while providing more interpretable and semantically rich feature representations. Code is available at https://github.com/uqzhichen/SVIP.

cs.CV↗

arXiv2025

Are Synthetic Videos Useful? A Benchmark for Retrieval-Centric Evaluation of Synthetic Videos

Text-to-video (T2V) synthesis has advanced rapidly, yet current evaluation metrics primarily capture visual quality and temporal consistency, offering limited insight into how synthetic videos perform in downstream tasks such as text-to-video retrieval (TVR). In this work, we introduce SynTVA, a new dataset and benchmark designed to evaluate the utility of synthetic videos for building retrieval models. Based on 800 diverse user queries derived from MSRVTT training split, we generate synthetic videos using state-of-the-art T2V models and annotate each video-text pair along four key semantic alignment dimensions: Object \& Scene, Action, Attribute, and Prompt Fidelity. Our evaluation framework correlates general video quality assessment (VQA) metrics with these alignment scores, and examines their predictive power for downstream TVR performance. To explore pathways of scaling up, we further develop an Auto-Evaluator to estimate alignment quality from existing metrics. Beyond benchmarking, our results show that SynTVA is a valuable asset for dataset augmentation, enabling the selection of high-utility synthetic samples that measurably improve TVR outcomes. Project page and dataset can be found at https://jasoncodemaker.github.io/SynTVA/.

cs.CV↗

arXiv2025

Continual Text-to-Video Retrieval with Frame Fusion and Task-Aware Routing

Text-to-Video Retrieval (TVR) aims to retrieve relevant videos based on textual queries. However, as video content evolves continuously, adapting TVR systems to new data remains a critical yet under-explored challenge. In this paper, we introduce the first benchmark for Continual Text-to-Video Retrieval (CTVR) to address the limitations of existing approaches. Current Pre-Trained Model (PTM)-based TVR methods struggle with maintaining model plasticity when adapting to new tasks, while existing Continual Learning (CL) methods suffer from catastrophic forgetting, leading to semantic misalignment between historical queries and stored video features. To address these two challenges, we propose FrameFusionMoE, a novel CTVR framework that comprises two key components: (1) the Frame Fusion Adapter (FFA), which captures temporal video dynamics while preserving model plasticity, and (2) the Task-Aware Mixture-of-Experts (TAME), which ensures consistent semantic alignment between queries across tasks and the stored video features. Thus, FrameFusionMoE enables effective adaptation to new video content while preserving historical text-video relevance to mitigate catastrophic forgetting. We comprehensively evaluate FrameFusionMoE on two benchmark datasets under various task settings. Results demonstrate that FrameFusionMoE outperforms existing CL and TVR methods, achieving superior retrieval performance with minimal degradation on earlier tasks when handling continuous video streams. Our code is available at: https://github.com/JasonCodeMaker/CTVR.

cs.CV↗

arXiv2024

CF-PRNet: Coarse-to-Fine Prototype Refining Network for Point Cloud Completion and Reconstruction

In modern agriculture, precise monitoring of plants and fruits is crucial for tasks such as high-throughput phenotyping and automated harvesting. This paper addresses the challenge of reconstructing accurate 3D shapes of fruits from partial views, which is common in agricultural settings. We introduce CF-PRNet, a coarse-to-fine prototype refining network, leverages high-resolution 3D data during the training phase but requires only a single RGB-D image for real-time inference. Our approach begins by extracting the incomplete point cloud data that constructed from a partial view of a fruit with a series of convolutional blocks. The extracted features inform the generation of scaling vectors that refine two sequentially constructed 3D mesh prototypes - one coarse and one fine-grained. This progressive refinement facilitates the detailed completion of the final point clouds, achieving detailed and accurate reconstructions. CF-PRNet demonstrates excellent performance metrics with a Chamfer Distance of 3.78, an F1 Score of 66.76%, a Precision of 56.56%, and a Recall of 85.31%, and win the first place in the Shape Completion and Reconstruction of Sweet Peppers Challenge.

cs.CV↗

arXiv2024

FastEdit: Fast Text-Guided Single-Image Editing via Semantic-Aware Diffusion Fine-Tuning

Conventional Text-guided single-image editing approaches require a two-step process, including fine-tuning the target text embedding for over 1K iterations and the generative model for another 1.5K iterations. Although it ensures that the resulting image closely aligns with both the input image and the target text, this process often requires 7 minutes per image, posing a challenge for practical application due to its time-intensive nature. To address this bottleneck, we introduce FastEdit, a fast text-guided single-image editing method with semantic-aware diffusion fine-tuning, dramatically accelerating the editing process to only 17 seconds. FastEdit streamlines the generative model's fine-tuning phase, reducing it from 1.5K to a mere 50 iterations. For diffusion fine-tuning, we adopt certain time step values based on the semantic discrepancy between the input image and target text. Furthermore, FastEdit circumvents the initial fine-tuning step by utilizing an image-to-image model that conditions on the feature space, rather than the text embedding space. It can effectively align the target text prompt and input image within the same feature space and save substantial processing time. Additionally, we apply the parameter-efficient fine-tuning technique LoRA to U-net. With LoRA, FastEdit minimizes the model's trainable parameters to only 0.37\% of the original size. At the same time, we can achieve comparable editing outcomes with significantly reduced computational overhead. We conduct extensive experiments to validate the editing performance of our approach and show promising editing capabilities, including content addition, style transfer, background replacement, and posture manipulation, etc.

cs.CV↗