arXiv subjects

Jiandong Zhang

Publications and source records attributed to Jiandong Zhang.

15 recordsLinked to original sources

arXiv2026

The Ubiquity of Three Steady States: Minimal Multistable Zero-One Reaction Networks

Biochemical networks with unit stoichiometry arise naturally in receptor-ligand binding and multisite phosphorylation systems. A central question is to identify which networks admit multistability. Because this property can be inherited from smaller subnetworks to larger ones, it is natural to seek the smallest networks with it. In this paper, we completely determine all minimal quadratic zero-one networks that exhibit multistability. Building on recent progress that has narrowed the search to the zero-one networks with 3 species, 6 reactions, and dimension 3, say (3, 6, 3) family. We develop a computational pipeline to provide a complete characterization of multistability within this class. The primary theoretical contribution is a structural simplification showing that for (3, m, 3) quadratic zero-one networks, the Jacobian determinants at consecutive positive steady states have opposite signs, and exactly half of these states are stable. This reduces multistationarity detection and stability verification to a local sign-checking problem, eliminating the need for boundary or asymptotic analysis. Applying this result, we identify 373 quadratic zero-one networks that exhibit bistability (two stable and one unstable positive steady states). Strikingly, no (3, 6, 3) quadratic zero-one network admits more than 3 positive steady states--a sharp bound that falls well below the theoretical BKK bound of 5 and the Bezout bound of 8. Among the 375 networks that admit exactly 3 positive steady states, 373 are multistable and only 2 are not. These minimal networks provide essential test cases for understanding how bistability emerges in cell signaling without requiring higher-order molecular collisions.

math.DS↗

arXiv2026

Degeneracy of Two-Dimensional Zero-One Reaction Networks with Up to Three Species

Zero-one biochemical reaction networks are widely recognized for their importance in analyzing signal transduction and cellular decision-making processes. Degenerate networks reveal non-standard behaviors and mark the boundary where classical methods fail. Their analysis is key to understanding exceptional dynamical phenomena in biochemical systems. Therefore, we focus on investigating the degeneracy of zero-one reaction networks. It is known that one-dimensional zero-one networks cannot degenerate. In this work, we identify all degenerate two-dimensional zero-one reaction networks with up to three species by an efficient algorithm. By analyzing the structure of these networks, we arrive at the following conclusion: if a two-dimensional zero-one reaction network with three species is degenerate, then its steady-state system is equivalent to a binomial system.

q-bio.MN↗

arXiv2026

Data-dependent Exploration for Online Reinforcement Learning from Human Feedback

Online reinforcement learning from human feedback (RLHF) has emerged as a promising paradigm for aligning large language models (LLMs) by continuously collecting new preference feedback during training. A foundational challenge in this setting is exploration, which requires algorithms that enable the LLMs to generate informative comparisons that improve sample-efficiency in online RLHF. Existing exploration strategies often derive bonuses via on-policy expectations, which are difficult to estimate reliably from the limited historical preference data available during training; as a result, the policy can prematurely down-weight under-explored regions that may contain high-value behaviors. In this paper, we propose data-dependent exploration for preference optimization (DEPO), a simple and scalable method that leverages historical data to construct an extra uncertainty bonus for high-uncertainty regions, encouraging exploration toward potentially high-value data. Theoretically, we provide a data-dependent regret bound for the proposed algorithm, showing that it adapts to the hardness of the learning task itself and can be tighter than worst-case bounds in practice. Empirically, the proposed method consistently outperforms strong baselines across benchmarks, demonstrating improved sample efficiency.

cs.LG↗

arXiv2026

Decoupled Multimodal Fusion for User Interest Modeling in Click-Through Rate Prediction

Modern industrial recommendation systems improve recommendation performance by integrating multimodal representations from pre-trained models into ID-based Click-Through Rate (CTR) prediction frameworks. However, existing approaches typically adopt modality-centric modeling strategies that process ID-based and multimodal embeddings independently, failing to capture fine-grained interactions between content semantics and behavioral signals. In this paper, we propose Decoupled Multimodal Fusion (DMF), which introduces a modality-enriched modeling strategy to enable fine-grained interactions between ID-based collaborative representations and multimodal representations for user interest modeling. Specifically, we construct target-aware features to bridge the semantic gap across different embedding spaces and leverage them as side information to enhance the effectiveness of user interest modeling. Furthermore, we design an inference-optimized attention mechanism that decouples the computation of target-aware features and ID-based embeddings before the attention layer, thereby alleviating the computational bottleneck introduced by incorporating target-aware features. To achieve comprehensive multimodal integration, DMF combines user interest representations learned under the modality-centric and modality-enriched modeling strategies. Offline experiments on public and industrial datasets demonstrate the effectiveness of DMF. Moreover, DMF has been deployed on the product recommendation system of the international e-commerce platform Lazada, achieving relative improvements of 5.30% in CTCVR and 7.43% in GMV with negligible computational overhead.

cs.IR↗

arXiv2025

Superresolution of unequal-brightness thermal sources for stellar interferometry

Resolving high-contrast targets is a fundamental yet highly challenging task in astronomy. Using quantum estimation theory, we demonstrate that the ultimate limit for estimating the separation between two unequal-brightness thermal sources via interferometry remains constant, enabling the potential for superresolution. We give a comparative analysis of two primary stellar interferometric schemes: amplitude interferometry and intensity interferometry. Notably, the nulling strategy employed in amplitude interferometry, a configuration specifically proposed for exoplanet detection by leveraging destructive interference to suppress the brighter source, is quantum optimal for separation estimation. While intensity interferometry is less effective than amplitude interferometry in lossless scenarios and fails to achieve superresolution, it becomes competitive when optical loss in large-scale interferometry is considered. By applying these methodologies to modern stellar interferometry, we highlight the promise of large-scale interferometry for advancing high-resolution astronomical observation.

astro-ph.IM↗

arXiv2025

In-context Demonstration Matters: On Prompt Optimization for Pseudo-Supervision Refinement

Large language models (LLMs) have achieved great success across diverse tasks, and fine-tuning is sometimes needed to further enhance generation quality. Most existing methods rely on human supervision or parameter retraining, both of which are costly in terms of data collection and computational resources. To handle these challenges, a direct solution is to generate ``high-confidence'' data from unsupervised downstream tasks and use them for in-context prompting or prompt optimization to refine the pseudo-supervision. However, relying solely on such data may lead to overfitting. In this paper, we leverage the in-context learning (ICL) abilities of LLMs and propose a novel approach, pseudo-supervised demonstrations aligned prompt optimization (PAPO) algorithm, which jointly refines both the prompt and the overall pseudo-supervision. The proposed learning objective ensures that the optimized prompt guides the LLM to generate consistent responses for a given input when pseudo-supervised data from the downstream task are used as demonstrations, enabling refinement over the entire pseudo-supervision. The prompt is optimized by translating gradient signals into textual critiques, which serve as feedback to iteratively refine the prompt and model responses. Theoretical analysis in a simplified classification setting shows that the refined pseudo-supervision exhibits a geometric clustering structure, helping to mitigate overfitting. Experiments on question answering, natural language inference benchmarks, and a real-world molecule optimization task, show the effectiveness of the proposed algorithm.

cs.CL↗

arXiv2023

Analyzing Journal Category Assignment Using a Paper-level Classification System: Multidisciplinary Sciences Journals

In the field of scientometrics, the subject classification system of academic journals holds great importance. Accurate identification and classification of "multidisciplinary" journals are crucial in revealing the scientific structure and evaluating journals. Based on data from the Web of Science database from 2016 to 2020, we calculated the disciplinary diversity of journals using the paper-level subject classification system, then conducted a systematic analysis of JCR multidisciplinary journals. Studies showed that most multidisciplinary journals have high disciplinary diversity, while non-multidisciplinary journals tend to have relatively lower diversity. Some multidisciplinary journals with low disciplinary diversities may misclassify disciplines. In addition, there are inconsistencies in the diversity of journal disciplines at different granularities. Our study also visually analyzed the four types of diversity distribution tendencies of multidisciplinary journals. Moreover, ten potential multidisciplinary journals were found in non-multidisciplinary categories.

cs.DL↗

arXiv2023

Collective-motion-enhanced acceleration sensing via an optically levitated microsphere array

Optically levitated microspheres are an excellent candidate for force and acceleration sensing. Here, we propose an acceleration sensing protocol based on an optically levitated microsphere array (MSA). The system consists of an $N$-microsphere array levitated in a driven optical cavity via holographic optical tweezers. By positioning the microspheres suitably relative to the cavity, only one of the collective modes of the MSA is coupled to the cavity mode. The optomechanical interaction encodes the information of acceleration acting on the MSA onto the intracavity photons, which can then be detected directly at the output of the cavity. The optically levitated MSA forms an effective large mass-distributed particle, which not only circumvents the problem of levitating a large mass microsphere but also results in a significant improvement of sensitivity. Compared with the traditional single-microsphere measurement scheme, our method presents an improvement in sensitivity by a factor of $\sqrt{N}$.

physics.optics↗

arXiv2022

TransVCL: Attention-enhanced Video Copy Localization Network with Flexible Supervision

Video copy localization aims to precisely localize all the copied segments within a pair of untrimmed videos in video retrieval applications. Previous methods typically start from frame-to-frame similarity matrix generated by cosine similarity between frame-level features of the input video pair, and then detect and refine the boundaries of copied segments on similarity matrix under temporal constraints. In this paper, we propose TransVCL: an attention-enhanced video copy localization network, which is optimized directly from initial frame-level features and trained end-to-end with three main components: a customized Transformer for feature enhancement, a correlation and softmax layer for similarity matrix generation, and a temporal alignment module for copied segments localization. In contrast to previous methods demanding the handcrafted similarity matrix, TransVCL incorporates long-range temporal information between feature sequence pair using self- and cross- attention layers. With the joint design and optimization of three components, the similarity matrix can be learned to present more discriminative copied patterns, leading to significant improvements over previous methods on segment-level labeled datasets (VCSL and VCDB). Besides the state-of-the-art performance in fully supervised setting, the attention architecture facilitates TransVCL to further exploit unlabeled or simply video-level labeled data. Additional experiments of supplementing video-level labeled datasets including SVD and FIVR reveal the high flexibility of TransVCL from full supervision to semi-supervision (with or without video-level annotation). Code is publicly available at https://github.com/transvcl/TransVCL.

cs.CV↗

arXiv2022

A Method to Judge the Style of Classical Poetry Based on Pre-trained Model

One of the important topics in the research field of Chinese classical poetry is to analyze the poetic style. By examining the relevant works of previous dynasties, researchers judge a poetic style mostly by their subjective feelings, and refer to the previous evaluations that have become a certain conclusion. Although this judgment method is often effective, there may be some errors. This paper builds the most perfect data set of Chinese classical poetry at present, trains a BART-poem pre -trained model on this data set, and puts forward a generally applicable poetry style judgment method based on this BART-poem model, innovatively introduces in-depth learning into the field of computational stylistics, and provides a new research method for the study of classical poetry. This paper attempts to use this method to solve the problem of poetry style identification in the Tang and Song Dynasties, and takes the poetry schools that are considered to have a relatively clear and consistent poetic style, such as the Hongzheng Qizi and Jiajing Qizi, Jiangxi poetic school and Tongguang poetic school, as the research object, and takes the poems of their representative poets for testing. Experiments show that the judgment results of the tested poetry work made by the model are basically consistent with the conclusions given by critics of previous dynasties, verify some avant-garde judgments of Mr. Qian Zhongshu, and better solve the task of poetry style recognition in the Tang and Song dynasties.

cs.CL↗

arXiv2022

CAEN: A Hierarchically Attentive Evolution Network for Item-Attribute-Change-Aware Recommendation in the Growing E-commerce Environment

Traditional recommendation systems mainly focus on modeling user interests. However, the dynamics of recommended items caused by attribute modifications (e.g. changes in prices) are also of great importance in real systems, especially in the fast-growing e-commerce environment, which may cause the users' demands to emerge, shift and disappear. Recent studies that make efforts on dynamic item representations treat the item attributes as side information but ignore its temporal dependency, or model the item evolution with a sequence of related users but do not consider item attributes. In this paper, we propose Core Attribute Evolution Network (CAEN), which partitions the user sequence according to the attribute value and thus models the item evolution over attribute dynamics with these users. Under this framework, we further devise a hierarchical attention mechanism that applies attribute-aware attention for user aggregation under each attribute, as well as personalized attention for activating similar users in assessing the matching degree between target user and item. Results from the extensive experiments over actual e-commerce datasets show that our approach outperforms the state-of-art methods and achieves significant improvements on the items with rapid changes over attributes, therefore helping the item recommendation to adapt to the growth of the e-commerce platform.

cs.IR↗

arXiv2021

SU(1,1) interferometry with parity measurement

We present a new operator method in the Heisenberg representation to obtain the signal of parity measurement within a lossless SU(1,1) interferometer. Based on this method, it is convenient to derive the parity signal directly in terms of input states, including general Gaussian or non-Gaussian state. As applications, we revisit the signal of parity measurement within an SU(1,1) interferometer when a coherent or thermal state and a squeezed vacuum state are considered as input states. In addition, we also obtain the parity signal of a Fock state when it passes through an SU(1,1) interferometer, which is also a new result. Therefore, the operator method proposed in this work may bring convenience to the study of quantum metrology, particularly the phase estimation based on an SU(1,1) interferometer.

quant-ph↗

arXiv2020

Hybrid Interest Modeling for Long-tailed Users

User behavior modeling is a key technique for recommender systems. However, most methods focus on head users with large-scale interactions and hence suffer from data sparsity issues. Several solutions integrate side information such as demographic features and product reviews, another is to transfer knowledge from other rich data sources. We argue that current methods are limited by the strict privacy policy and have low scalability in real-world applications and few works consider the behavioral characteristics behind long-tailed users. In this work, we propose the Hybrid Interest Modeling (HIM) network to hybrid both personalized interest and semi-personalized interest in learning long-tailed users' preferences in the recommendation. To achieve this, we first design the User Behavior Pyramid (UBP) module to capture the fine-grained personalized interest of high confidence from sparse even noisy positive feedbacks. Moreover, the individual interaction is too sparse and not enough for modeling user interest adequately, we design the User Behavior Clustering (UBC) module to learn latent user interest groups with self-supervised learning mechanism novelly, which capture coarse-grained semi-personalized interest from group-item interaction data. Extensive experiments on both public and industrial datasets verify the superiority of HIM compared with the state-of-the-art baselines.

cs.IR↗

arXiv2020

An Accurate Model for Predicting the (Graded) Effect of Context in Word Similarity Based on Bert

Natural Language Processing (NLP) has been widely used in the semantic analysis in recent years. Our paper mainly discusses a methodology to analyze the effect that context has on human perception of similar words, which is the third task of SemEval 2020. We apply several methods in calculating the distance between two embedding vector generated by Bidirectional Encoder Representation from Transformer (BERT). Our team will_go won the 1st place in Finnish language track of subtask1, the second place in English track of subtask1.

cs.CL↗

arXiv2018

Orbital-angular-momentum-enhanced estimation of sub-Heisenberg-limited angular displacement with two-mode squeezed vacuum and parity detection

We report on an orbital-angular-momentum-enhanced scheme for angular displacement estimation based on two-mode squeezed vacuum and parity detection. The sub-Heisenberg-limited sensitivity for angular displacement estimation is obtained in an ideal situation. Several realistic factors are also considered, including photon loss, dark counts, response-time delay, and thermal photon noise. Our results indicate that the effects of the realistic factors on the sensitivity can be offset by raising orbital angular momentum quantum number $\ell$. This reflects that the robustness and the practicability of the system can be improved via raising $\ell$ without changing mean photon number $N$.

quant-ph↗