arXiv subjects

Meiqi Sun

Publications and source records attributed to Meiqi Sun.

7 recordsLinked to original sources

arXiv2026

E-comIQ-ZH: A Human-Aligned Dataset and Benchmark for Fine-Grained Evaluation of E-commerce Posters with Chain-of-Thought

Generative AI is widely used to create commercial posters. However, rapid advances in generation have outpaced automated quality assessment. Existing models emphasize generic esthetics or low level distortions and lack the functional criteria required for e-commerce design. It is especially challenging for Chinese content, where complex characters often produce subtle but critical textual artifacts that are overlooked by existing methods. To address this, we introduce E-comIQ-ZH, a framework for evaluating Chinese e-commerce posters. We build the first dataset E-comIQ-18k to feature multi dimensional scores and expert calibrated Chain of Thought (CoT) rationales. Using this dataset, we train E-comIQ-M, a specialized evaluation model that aligns with human expert judgment. Our framework enables E-comIQ-Bench, the first automated and scalable benchmark for the generation of Chinese e-commerce posters. Extensive experiments show our E-comIQ-M aligns more closely with expert standards and enables scalable automated assessment of e-commerce posters. All datasets, models, and evaluation tools will be released to support future research in this area.Code will be available at https://github.com/4mm7/E-comIQ-ZH.

cs.CV↗

arXiv2025

SEMDICE: Off-policy State Entropy Maximization via Stationary Distribution Correction Estimation

In the unsupervised pre-training for reinforcement learning, the agent aims to learn a prior policy for downstream tasks without relying on task-specific reward functions. We focus on state entropy maximization (SEM), where the goal is to learn a policy that maximizes the entropy of the state stationary distribution. In this paper, we introduce SEMDICE, a principled off-policy algorithm that computes an SEM policy from an arbitrary off-policy dataset, which optimizes the policy directly within the space of stationary distributions. SEMDICE computes a single, stationary Markov state-entropy-maximizing policy from an arbitrary off-policy dataset. Experimental results demonstrate that SEMDICE outperforms baseline algorithms in maximizing state entropy while achieving the best adaptation efficiency for downstream tasks among SEM-based unsupervised RL pre-training methods.

cs.LG↗

arXiv2025

Cognitive biases shape the evolution of zero-sum norms

Why do maladaptive perceptions and norms, such as zero-sum interpretations of interaction, persist even when they undermine cooperation and investment? We develop a framework where bounded rationality and heterogeneous cognitive biases shape the evolutionary dynamics of norm coordination. Extending evolutionary game theory with quantal response equilibria and prospect-theoretic utility, we show that subjective evaluation of payoffs systematically alters population-level equilibrium selection, generating stable but inefficient attractors. Counterintuitively, our analysis demonstrates that the benefit of rationality and the cost of risk aversion on welfare behave in nonmonotone ways: intermediate precision enhances coordination, while excessive precision or strong loss aversion leads to persistent lock-in at low-payoff and zero-sum equilibria. These dynamics produce an endogenous equity-efficiency trade-off: parameter configurations that raise aggregate welfare also increase inequality, while more equal distributions are associated with lower efficiency. The results highlight how distorted payoff perceptions can anchor societies in divergent institutional trajectories, offering a behavioral-evolutionary explanation for persistent zero-sum norms and inequality.

econ.GN↗

arXiv2025

Investigation of the neutron-proton effective mass splitting via heavy ion collisions: Constraints and Implications

The neutron-proton effective mass splitting ($\Delta m^*_{np}$) is investigated through analyses of heavy-ion collisions using the improved quantum molecular dynamics (ImQMD) model with both standard and extended Skyrme interactions. By uncovering the strong correlation between the slope of the neutron-to-proton yield ratio with respect to the kinetic energy (i.e., $S_{n/p} $) and $\Delta m^*_{np}$, we reveal that the constraints of the neutron-proton effective mass splitting via heavy ion collisions depend on the kinetic energy region of the emitted nucleons. At low kinetic energies, the data favor $m_n^*>m_p^*$ which is consistent with the nucleon-nucleus scattering analysis, while at high kinetic energies, they favor $m_n^*<m_p^*$. Our findings partly resolve the longstanding discrepancy in the constraints of neutron-proton effective mass splitting with heavy ion collisions and nucleon-nucleus scattering, and significantly advance the understanding of nucleon effective mass splitting through heavy ion collisions.

nucl-th↗

arXiv2025

An effective finite-range Gogny-type interaction for the quantum molecular dynamics like model

In this work, we propose an effective finite-range Gogny-type interaction that can be directly used in the quantum molecular dynamics (QMD) like model. Two methods for determining the parameters of the effective interaction are discussed. The first method establishes an approach to connect the conventional Gogny interaction in nuclear structure to that in heavy-ion collisions, the second method allows for the description of the symmetry energy varying from the supersoft to stiff, as well as the momentum-dependent symmetry potential, exhibiting behaviors ranging from monotonic to non-monotonic variations. This effective interaction opens up opportunities for a deeper understanding of finite-range interactions and non-monotonic momentum-dependent symmetry potentials in future studies.

nucl-th↗

arXiv2024

Ego3DT: Tracking Every 3D Object in Ego-centric Videos

The growing interest in embodied intelligence has brought ego-centric perspectives to contemporary research. One significant challenge within this realm is the accurate localization and tracking of objects in ego-centric videos, primarily due to the substantial variability in viewing angles. Addressing this issue, this paper introduces a novel zero-shot approach for the 3D reconstruction and tracking of all objects from the ego-centric video. We present Ego3DT, a novel framework that initially identifies and extracts detection and segmentation information of objects within the ego environment. Utilizing information from adjacent video frames, Ego3DT dynamically constructs a 3D scene of the ego view using a pre-trained 3D scene reconstruction model. Additionally, we have innovated a dynamic hierarchical association mechanism for creating stable 3D tracking trajectories of objects in ego-centric videos. Moreover, the efficacy of our approach is corroborated by extensive experiments on two newly compiled datasets, with 1.04x - 2.90x in HOTA, showcasing the robustness and accuracy of our method in diverse ego-centric scenarios.

cs.CV↗

arXiv2023

UniAP: Towards Universal Animal Perception in Vision via Few-shot Learning

Animal visual perception is an important technique for automatically monitoring animal health, understanding animal behaviors, and assisting animal-related research. However, it is challenging to design a deep learning-based perception model that can freely adapt to different animals across various perception tasks, due to the varying poses of a large diversity of animals, lacking data on rare species, and the semantic inconsistency of different tasks. We introduce UniAP, a novel Universal Animal Perception model that leverages few-shot learning to enable cross-species perception among various visual tasks. Our proposed model takes support images and labels as prompt guidance for a query image. Images and labels are processed through a Transformer-based encoder and a lightweight label encoder, respectively. Then a matching module is designed for aggregating information between prompt guidance and the query image, followed by a multi-head label decoder to generate outputs for various tasks. By capitalizing on the shared visual characteristics among different animals and tasks, UniAP enables the transfer of knowledge from well-studied species to those with limited labeled data or even unseen species. We demonstrate the effectiveness of UniAP through comprehensive experiments in pose estimation, segmentation, and classification tasks on diverse animal species, showcasing its ability to generalize and adapt to new classes with minimal labeled examples.

cs.CV↗