arXiv subjects

Qiao Yang

Publications and source records attributed to Qiao Yang.

11 recordsLinked to original sources

arXiv2026

LoopFM: Learning frOm HistOrical RePresentations of Foundation Model for Recommendation

Knowledge distillation (KD) transfers a single scalar prediction from a large foundation model (FM) to compact vertical models (VMs), suffering from diminishing transfer ratio -- the fraction of FM improvement captured by the VM -- as a single scalar cannot convey the rich intermediate knowledge that larger FMs learn. To address this bottleneck, we propose LoopFM (Learning frOm HistOrical RePresentations of FM), a framework that opens a high-bandwidth transfer channel by structuring FM intermediate embeddings as input features (e.g., user history sequence) for downstream VMs, without requiring real-time FM inference at serving and architectural coupling between FM and VM. We provide a theoretical framework for LoopFM with a gain decomposition and transfer-ratio analysis. On three public benchmarks, LoopFM demonstrates strong AUC improvements (e.g., 6%+ on TaobaoAd) and complementary knowledge transfer capability with KD. On industrial-scale systems (billions of examples, trillion-parameter FMs), LoopFM approximately doubles the knowledge transfer ratio on top of KD, delivering a +0.5% conversion improvement in the first half after its initial launch, and +1.03% and +1.22% conversion improvement from two individual launches in the subsequent half.

cs.LG↗

arXiv2026

Memento: Personalized RAG-Style Long-Retention Data Scaling for META Ads Recommendation

Modeling of long history data suffers from long-context window attention dilution, system efficiency and catastrophic forgetting problems, where naive linear scaling approach like LastN would fail. We introduce Memento, a personalized retrieval-augmented framework that treats historical user engagements as a document corpus and ad requests as queries, retrieving relevant interactions via Maximal Marginal Relevance (MMR) to balance similarity with diversity. We identify two complementary applications: Representation Memento, which retrieves historical embeddings for feature augmentation, and Data Memento, which retrieves past training examples for multipass training. Through infrastructure co-design -- temporal chunking, INT8 quantization, and asynchronous serving -- Memento achieves 5-10$\times$ resource efficiency over linear scaling. Memento processes daily requests with sub-10ms latency, yielding 0.25-0.3% Normalized Entropy gain on both click-through and conversion prediction. In production, Memento delivers a 1% CTR lift on Facebook Feed and Reels and a 1.2% CVR lift, scaling personalization to 365+ days of history.

cs.IR↗

arXiv2026

Intelligent Elastic Feature Fading: Enabling Model Retrain-Free Feature Efficiency Rollouts at Scale

Large-scale ranking systems depend on thousands of features derived from user behavior across multiple time horizons. Typically requires model retraining -- resulting in long iteration cycles (3--6 months), substantial GPU resource consumption, and limited rollout throughput. We introduce Intelligent Elastic Feature Fading (IEFF), a production infrastructure system that enables retrain-free feature efficiency rollouts by elastically controlling feature coverage and distribution at serving time. IEFF supports incremental feature coverage adjustments while models adapt through recurring training, eliminating dependencies on explicit retraining cycles. The system incorporates strict safety guardrails, reversibility mechanisms, and comprehensive monitoring to ensure stability at scale. Across multiple production use cases, IEFF accelerates efficiency-related rollouts by 5$\times$, eliminates retraining-related GPU overhead, and enables faster capacity recycling. Extensive offline and online experiments demonstrate that gradual feature fading prevents 50--55\% of online performance degradation compared to abrupt feature removal, while maintaining stable model behavior. These results establish elastic, system-level feature fading as a practical and scalable approach for managing feature efficiency in modern industrial ranking systems.

cs.IR↗

arXiv2025

Dominant Excitonic Superconductivity in a Three-component Hubbard Chain

Understanding superconductivity emerging from repulsive fermions remains a major challenge in condensed matter physics. In this paper, we investigate the pairing tendencies in a one-dimensional, three component repulsive Hubbard model, using the density matrix renormalization group method. At half-filling, the system exhibits density wave ground state due to strong Hubbard repulsions. Upon doping, we find that Cooper pairs can emerge, whose fluctuations predominate the long-range physics in the system across a wide parameter range. The effective attractions between Cooper pairs are mediated by the particle-hole fluctuations in the third non-pairing component, resembling an excitonic mechanism of superconductivity. The coexistence of multiple density waves and superconductivity at different fermion fillings is explored. We also present an analytical study of the pairing mechanism in both weak and strong coupling limits. Our results provide a new perspective for understanding and exploring unconventional superconductivities in strongly correlated fermionic systems.

cond-mat.supr-con↗

arXiv2025

EgoTrigger: Toward Audio-Driven Image Capture for Human Memory Enhancement in All-Day Energy-Efficient Smart Glasses

All-day smart glasses are likely to emerge as platforms capable of continuous contextual sensing, uniquely positioning them for unprecedented assistance in our daily lives. Integrating the multi-modal AI agents required for human memory enhancement while performing continuous sensing, however, presents a major energy efficiency challenge for all-day usage. Achieving this balance requires intelligent, context-aware sensor management. Our approach, EgoTrigger, leverages audio cues from the microphone to selectively activate power-intensive cameras, enabling efficient sensing while preserving substantial utility for human memory enhancement. EgoTrigger uses a lightweight audio model (YAMNet) and a custom classification head to trigger image capture from hand-object interaction (HOI) audio cues, such as the sound of a drawer opening or a medication bottle being opened. In addition to evaluating on the QA-Ego4D dataset, we introduce and evaluate on the Human Memory Enhancement Question-Answer (HME-QA) dataset. Our dataset contains 340 human-annotated first-person QA pairs from full-length Ego4D videos that were curated to ensure that they contained audio, focusing on HOI moments critical for contextual understanding and memory. Our results show EgoTrigger can use 54% fewer frames on average, significantly saving energy in both power-hungry sensing components (e.g., cameras) and downstream operations (e.g., wireless transmission), while achieving comparable performance on datasets for an episodic memory task. We believe this context-aware triggering strategy represents a promising direction for enabling energy-efficient, functional smart glasses capable of all-day use -- supporting applications like helping users recall where they placed their keys or information about their routine activities (e.g., taking medications).

cs.CV↗

arXiv2024

Emergence of charge density wave and superconducting phase transitions through Lorentz-invariant interactions in the Haldane-Hubbard model

We derive Lorentz-invariant four-fermion interactions, including Nambu-Jona-Lasinio type and superconducting type, which are widely studied in high-energy physics, from the honeycomb lattice Hamiltonian with Hubbard interaction. We investigate the phase transitions induced by these two interactions and consider the effects of the chemical potential and magnetic flux (Haldane mass term) on these phase transitions. We find that the charge-density-wave and superconductivity generated by the attractive interactions are mainly controlled by the chemical potential, while the magnetic flux delimits the domain of phase transition. Our analysis underscores the influence of the initial topological state on the phase transitions, a facet largely overlooked in prior studies. We present experimental protocols using cold atoms to verify our theoretical results.

cond-mat.str-el↗

arXiv2024

IAIFNet: An Illumination-Aware Infrared and Visible Image Fusion Network

Infrared and visible image fusion (IVIF) is used to generate fusion images with comprehensive features of both images, which is beneficial for downstream vision tasks. However, current methods rarely consider the illumination condition in low-light environments, and the targets in the fused images are often not prominent. To address the above issues, we propose an Illumination-Aware Infrared and Visible Image Fusion Network, named as IAIFNet. In our framework, an illumination enhancement network first estimates the incident illumination maps of input images. Afterwards, with the help of proposed adaptive differential fusion module (ADFM) and salient target aware module (STAM), an image fusion network effectively integrates the salient features of the illumination-enhanced infrared and visible images into a fusion image of high visual quality. Extensive experimental results verify that our method outperforms five state-of-the-art methods of fusing infrared and visible images.

cs.CV↗

arXiv2024

Chiral symmetry breaking in the pseudo-quantum electrodynamics with non-Abelian four-fermion interactions

In the context of 2+1 dimensional Dirac materials, we consider electromagnetic interactions alongside a type of spin-dependent Hubbard interaction. The former is described by PQED theory, while the latter corresponds to an effective theory represented by the $SU(N_c)$ Thirring model. Employing Hubbard-Stratonovich transformation and large N expansion in the model yields a non-local $SU(N_c)$ Yang-Mills action. Subsequently, we solve Schwinger-Dyson equations to obtain the self-energy function of the fermion propagator, from which we determine the critical fermion flavor number $N^c_f$ and critical fine structure constant $α_c$ indicative of chiral symmetry breaking. Our findings suggest that as the non-Abelian color number $N_c$ increases, the minimum value of the critical fermion flavor number monotonically increases, while the maximum value of the critical fine structure constant decreases accordingly, rendering the system more susceptible to chiral symmetry breaking.

hep-th↗

arXiv2024

Adaptive quantization with mixed-precision based on low-cost proxy

It is critical to deploy complicated neural network models on hardware with limited resources. This paper proposes a novel model quantization method, named the Low-Cost Proxy-Based Adaptive Mixed-Precision Model Quantization (LCPAQ), which contains three key modules. The hardware-aware module is designed by considering the hardware limitations, while an adaptive mixed-precision quantization module is developed to evaluate the quantization sensitivity by using the Hessian matrix and Pareto frontier techniques. Integer linear programming is used to fine-tune the quantization across different layers. Then the low-cost proxy neural architecture search module efficiently explores the ideal quantization hyperparameters. Experiments on the ImageNet demonstrate that the proposed LCPAQ achieves comparable or superior quantization accuracy to existing mixed-precision models. Notably, LCPAQ achieves 1/200 of the search time compared with existing methods, which provides a shortcut in practical quantization use for resource-limited devices.

cs.CV↗

arXiv2023

SSPFusion: A Semantic Structure-Preserving Approach for Infrared and Visible Image Fusion

Most existing learning-based multi-modality image fusion (MMIF) methods suffer from significant structure inconsistency due to their inappropriate usage of structural features at the semantic level. To alleviate these issues, we propose a semantic structure-preserving fusion approach for MMIF, namely SSPFusion. At first, we design a structural feature extractor (SFE) to extract the prominent structural features from multiple input images. Concurrently, we introduce a transformation function with Sobel operator to generate self-supervised structural signals in these extracted features. Subsequently, we design a multi-scale structure-preserving fusion (SPF) module, guided by the generated structural signals, to merge the structural features of input images. This process ensures the preservation of semantic structure consistency between the resultant fusion image and the input images. Through the synergy of these two robust modules of SFE and SPF, our method can generate high-quality fusion images and demonstrate good generalization ability. Experimental results, on both infrared-visible image fusion and medical image fusion tasks, demonstrate that our method outperforms nine state-of-the-art methods in terms of both qualitative and quantitative evaluations. The code is publicly available at https://github.com/QiaoYang-CV/SSPFUSION.

cs.CV↗

arXiv2015

Path-based Iterative Reconstruction (PBIR) for X-ray Computed Tomography

Model-based iterative reconstruction (MBIR) techniques have demonstrated many advantages in X-ray CT image reconstruction. The MBIR approach is often modeled as a convex optimization problem including a data fitting function and a penalty function. The tuning parameter value that regulates the strength of the penalty function is critical for achieving good reconstruction results but difficult to choose. In this work, we describe two path seeking algorithms that are capable of efficiently generating a series of MBIR images with different strengths of the penalty function. The root-mean-squared-differences of the proposed path seeking algorithms are below 4 HU throughout the entire reconstruction path. With the efficient path seeking algorithm, we suggest a path-based iterative reconstruction (PBIR) to obtain complete information from the scanned data and reconstruction model.

math.OC↗