arXiv subjects

Xuguang Liu

Publications and source records attributed to Xuguang Liu.

4 recordsLinked to original sources

arXiv2026

ERNIE-Image Technical Report

We introduce ERNIE-Image, an open-source text-to-image generation model built upon an 8B single-stream DiT architecture. ERNIE-Image aims to bridge the gap between current open-source models and leading closed-source systems through more effective mining of large-scale pre-training data and improved supervision quality throughout training. During pre-training, we adopt a bottom-up data construction pipeline that combines fine-grained image categorization, rich caption annotation, aesthetic assessment, and hierarchical sampling. This strategy reduces data noise while preserving long-tail concepts and detailed real-world knowledge, providing a stronger foundation for complex generation tasks. In the post-training stage, we use a top-down data construction pipeline for high-demand scenarios, diversify prompt annotations to better match real user inputs, and apply a stabilized DPO strategy to align the model with human aesthetic preferences. We further train ERNIE-Image-Turbo for efficient 8-NFE generation and propose MT-DMD to mitigate capability drift during distillation. To make the model easier to use in practical scenarios, we equip it with a lightweight Prompt Enhancer that expands concise user intents into structured visual descriptions. In addition, we develop ERNIE-Image-Aes, an industrial-grade aesthetic model, together with ERNIE-Image-Aes-1K, a human-annotated benchmark for realistic aesthetic evaluation. Extensive qualitative and quantitative experiments show that ERNIE-Image achieves leading performance among open-source models and approaches top-tier commercial models in instruction following, text rendering, and aesthetic quality. We release the trained models and aesthetic resources to facilitate further academic research and technical progress in the AIGC community.

cs.CV↗

arXiv2024

Advancements in Recommender Systems: A Comprehensive Analysis Based on Data, Algorithms, and Evaluation

Using 286 research papers collected from Web of Science, ScienceDirect, SpringerLink, arXiv, and Google Scholar databases, a systematic review methodology was adopted to review and summarize the current challenges and potential future developments in data, algorithms, and evaluation aspects of RSs. It was found that RSs involve five major research topics, namely algorithmic improvement, domain applications, user behavior & cognition, data processing & modeling, and social impact & ethics. Collaborative filtering and hybrid recommendation techniques are mainstream. The performance of RSs is jointly limited by four types of eight data issues, two types of twelve algorithmic issues, and two evaluation issues. Notably, data-related issues such as cold start, data sparsity, and data poisoning, algorithmic issues like interest drift, device-cloud collaboration, non-causal driven, and multitask conflicts, along with evaluation issues such as offline data leakage and multi-objective balancing, have prominent impacts. Fusing physiological signals for multimodal modeling, defending against data poisoning through user information behavior, evaluating generative recommendations via social experiments, fine-tuning pre-trained large models to schedule device-cloud resource, enhancing causal inference with deep reinforcement learning, training multi-task models based on probability distributions, using cross-temporal dataset partitioning, and evaluating recommendation objectives across the full lifecycle are feasible solutions to address the aforementioned prominent challenges and unlock the power and value of RSs.The collected literature is mainly based on major international databases, and future research will further expand upon it.

cs.IR↗

arXiv2023

Differential Antisymmetric Infinitesimal Bialgebras, Coherent Derivations and Poisson Bialgebras

We establish a bialgebra theory for differential algebras, called differential antisymmetric infinitesimal (ASI) bialgebras by generalizing the study of ASI bialgebras to the context of differential algebras, in which the derivations play an important role. They are characterized by double constructions of differential Frobenius algebras as well as matched pairs of differential algebras. Antisymmetric solutions of an analogue of associative Yang-Baxter equation in differential algebras provide differential ASI bialgebras, whereas in turn the notions of $\mathcal{O}$-operators of differential algebras and differential dendriform algebras are also introduced to produce the former. On the other hand, the notion of a coherent derivation on an ASI bialgebra is introduced as an equivalent structure of a differential ASI bialgebra. They include derivations on ASI bialgebras and the set of coherent derivations on an ASI bialgebra composes a Lie algebra which is the Lie algebra of the Lie group consisting of coherent automorphisms on this ASI bialgebra. Finally, we apply the study of differential ASI bialgebras to Poisson bialgebras, extending the construction of Poisson algebras from commutative differential algebras with two commuting derivations to the context of bialgebras, which is consistent with the well constructed theory of Poisson bialgebras. In particular, we construct Poisson bialgebras from differential Zinbiel algebras.

math.QA↗

arXiv2016

Low-temperature heat transport of spin-gapped quantum magnets

This article reviews low-temperature heat transport studies of spin-gapped quantum magnets in the last few decades. Quantum magnets with small spins and low dimensionality exhibit a variety of novel phenomena. Among them, some systems are char-acteristic of having quantum-mechanism spin gap in their magnetic excitation spectra, including spin-Peierls systems, S = 1 Haldane chains, S = 1/2 spin ladders, and spin dimmers. In some particular spin-gapped systems, the XY-type antiferromagnetic state induced by magnetic field that closes the spin gap can be described as a magnon Bose-Einstein condensation (BEC). Heat transport is effective in probing the magnetic excitations and magnetic phase transitions, and has been extensively studied for the spin-gapped systems. A large and ballistic spin thermal conductivity was observed in the two-leg Heisenberg S = 1/2 ladder compounds. The characteristic of magnetic thermal transport of the Haldane chain systems is quite controversial on both the theoretical and experimental results. For the spin-Peierls system, the spin excitations can also act as heat carriers. In spin-dimer compounds, the magnetic excitations mainly play a role of scattering phonons. The magnetic excitations in the magnon BEC systems displayed dual roles, carrying heat or scattering phonons, in different materials.

cond-mat.str-el↗