arXiv subjects

Qi Su

Publications and source records attributed to Qi Su.

At least 37 records · Page 2Linked to original sources

arXiv2025

Behavioral alignment in social networks

The orderly behaviors observed in large-scale groups, such as fish schooling and the organized movement of crowds, are both ubiquitous and essential for the survival and stability of these systems. Understanding how such complex collective behaviors emerge from simple local interactions and behavioral adjustments is a significant scientific challenge. Historically, research has predominantly focused on imitation and social learning, where individuals adopt the strategies of more successful peers to refine their behavior. However, in recent years, an alternative learning approach based on self-exploration and introspective learning has garnered increasing attention. In this paradigm, individuals assess their own circumstances and select strategies that best align with their specific conditions. Two examples are coordination and anti-coordination, where individuals align with and diverge from the local majority, respectively. In this study, we analyze networked systems of coordinating and anti-coordinating individuals, exploring the combined effects of system dynamics, network structure, and behavioral patterns. We address several practical questions, including the number of equilibria, their characteristics, the equilibrium time, and the resilience of the system. We find that the number of equilibrium states can be extremely large, even increasing exponentially with minor alterations to the network structure. Moreover, the network structure has a significant impact on the average equilibrium time. Despite the complexity of these findings, we find that variations can be captured by a single, simple network characteristic (the average path length), which we illustrate in both synthetic and empirical networks.

physics.soc-ph↗

arXiv2025

Evolution of social behaviors in noisy environments

Evolutionary game theory offers a general framework to study how behaviors evolve by social learning in a population. This body of theory can accommodate a range of social dilemmas, or games, as well as real-world complexities such as spatial structure or behaviors conditioned on reputations. Nonetheless, this approach typically assumes a deterministic payoff structure for social interactions. Here, we extend evolutionary game theory to account for random changes in the social environment, so that mutual cooperation may bring different rewards today than it brings tomorrow, for example. Even when such environmental noise is unbiased, we find it can have a qualitative impact on the behaviors that evolve in a population. Noisy payoffs can permit the stable co-existence of cooperators and defectors in the prisoner's dilemma, for example, as well as bistability in snowdrift games and stable limit cycles in rock-paper-scissors games -- dynamical phenomena that cannot occur in the absence of noise. We conclude by discussing the relevance of our framework to scenarios where the nature of social interactions is subject to external perturbations.

physics.soc-ph↗

arXiv2025

Evolutionary dynamics of memory-based strategies in repeated and structured social interactions

Human social life is shaped by repeated interactions, where past experiences guide future behavior. In evolutionary game theory, a key challenge is to identify strategies that harness such memory to succeed in repeated encounters. Decades of research have identified influential one-step memory strategies (such as Tit-for-Tat, Generous Tit-for-Tat, and Win-Stay Lose-Shift) that promote cooperation in iterated pairwise games. However, these strategies occupy only a small corner of the vast strategy space, and performance in isolated pairwise contests does not guarantee evolutionary success. The most effective strategies are those that can spread through a population and stabilize cooperation. We propose a general framework for repeated-interaction strategies that encompasses arbitrary memory lengths, diverse informational inputs (including both one's own and the opponent's past actions), and deterministic or stochastic decision rules. We analyze their evolutionary dynamics and derive general mathematical results for the emergence of cooperation in any network structure. We then introduce a unifying indicator that quantifies the contribution of repeated-interaction strategies to population-level cooperation. Applying this indicator, we show that long-memory strategies evolve to promote cooperation more effectively than short-memory strategies, challenging the traditional view that extended memory offers no advantage. This work expands the study of repeated interactions beyond one-step memory strategies to the full spectrum of memory capacities. It provides a plausible explanation for the high levels of cooperation observed in human societies, which traditional one-step memory models cannot account for.

q-bio.PE↗

arXiv2025

Explaining human cooperation through a dual mechanism of individual and social learning

Cooperation on social networks is crucial for understanding human survival and development. Although network structure has been found to significantly influence cooperation, human experiments have observed different cooperation phenomena under similar conditions. While evidence suggests that these differences arise from human exploration, our understanding of its impact mechanisms and characteristics remains limited. Here, we seek to formalize human exploration as an individual learning process involving trial and reflection, and integrate social learning to examine how their interdependence shapes cooperation. We find that individual learning can alter neighbor imitation tendencies, and the resulting shifts in the local cooperative environment feed back into the experiential cognition that guides individual learning. This coupled dynamic makes the ability of social networks to promote cooperation largely dependent on whether individuals focus on long-term payoffs, and exhibits a series of characteristics that can explain previously unexplained and seemingly contradictory cooperation phenomena. Surprisingly, individual learning can promote cooperation more than social learning when its probability is negatively correlated with payoffs, a mechanism rooted in the psychological tendency to avoid trial-and-error when individuals are satisfied with their current payoffs. These results explain the contradictory cooperation phenomenon by accounting for decision preferences and cognitive processes underlying exploration, bridging the gap between theoretical research and reality.

physics.soc-ph↗

arXiv2025

An Effective Incorporating Heterogeneous Knowledge Curriculum Learning for Sequence Labeling

Sequence labeling models often benefit from incorporating external knowledge. However, this practice introduces data heterogeneity and complicates the model with additional modules, leading to increased expenses for training a high-performing model. To address this challenge, we propose a two-stage curriculum learning (TCL) framework specifically designed for sequence labeling tasks. The TCL framework enhances training by gradually introducing data instances from easy to hard, aiming to improve both performance and training speed. Furthermore, we explore different metrics for assessing the difficulty levels of sequence labeling tasks. Through extensive experimentation on six Chinese word segmentation (CWS) and Part-of-speech tagging (POS) datasets, we demonstrate the effectiveness of our model in enhancing the performance of sequence labeling models. Additionally, our analysis indicates that TCL accelerates training and alleviates the slow training problem associated with complex models.

cs.CL↗

arXiv2025

Direct reciprocity in asynchronous interactions

Cooperation is vital for the survival of living systems but is challenging due to the costs borne by altruistic individuals. Direct reciprocity, where actions are based on past encounters, is a key mechanism fostering cooperation. However, most studies assume synchronous decision-making, whereas real-world interactions are often asynchronous, with individuals acting in sequence. This asynchrony can undermine standard cooperative strategies like Tit-for-Tat and Win-Stay Lose-Shift. To better understand cooperation in real-world contexts, it is crucial to explore the theory of direct reciprocity in asynchronous interactions. To address this, we introduce a framework based on asynchronous stochastic games, incorporating asynchronous decisions and dynamic environmental feedback. We analytically derive the conditions under which strategies form cooperative Nash equilibria. Our results demonstrate that the order of interactions can significantly alter outcomes: interaction asynchrony generally inhibits cooperation, except under specific conditions where environmental feedback effectively mitigates its negative impact. When environmental feedback is incorporated, a variety of stable reciprocal strategies can be sustained. Notably, above a critical environmental threshold, any cooperative strategy can form a Nash equilibrium. Overall, our work underscores the importance of interaction order in long-term evolutionary processes and highlights the pivotal role of environmental feedback in stabilizing cooperation in asynchronous interactions.

physics.soc-ph↗

arXiv2025

Code-as-Monitor: Constraint-aware Visual Programming for Reactive and Proactive Robotic Failure Detection

Automatic detection and prevention of open-set failures are crucial in closed-loop robotic systems. Recent studies often struggle to simultaneously identify unexpected failures reactively after they occur and prevent foreseeable ones proactively. To this end, we propose Code-as-Monitor (CaM), a novel paradigm leveraging the vision-language model (VLM) for both open-set reactive and proactive failure detection. The core of our method is to formulate both tasks as a unified set of spatio-temporal constraint satisfaction problems and use VLM-generated code to evaluate them for real-time monitoring. To enhance the accuracy and efficiency of monitoring, we further introduce constraint elements that abstract constraint-related entities or their parts into compact geometric elements. This approach offers greater generality, simplifies tracking, and facilitates constraint-aware visual programming by leveraging these elements as visual prompts. Experiments show that CaM achieves a 28.7% higher success rate and reduces execution time by 31.8% under severe disturbances compared to baselines across three simulators and a real-world setting. Moreover, CaM can be integrated with open-loop control policies to form closed-loop systems, enabling long-horizon tasks in cluttered scenes with dynamic environments.

cs.RO↗

arXiv2024

BACKRUNNER: Mitigating Smart Contract Attacks in the Real World

Billions of dollars have been lost due to vulnerabilities in smart contracts. To counteract this, researchers have proposed attack frontrunning protections designed to preempt malicious transactions by inserting "whitehat" transactions ahead of them to protect the assets. In this paper, we demonstrate that existing frontrunning protections have become ineffective in real-world scenarios. Specifically, we collected 158 recent real-world attack transactions and discovered that 141 of them can bypass state-of-the-art frontrunning protections. We systematically analyze these attacks and show how inherent limitations of existing frontrunning techniques hinder them from protecting valuable assets in the real world. We then propose a new approach involving 1) preemptive hijack, and 2) attack backrunning, which circumvent the existing limitations and can help protect assets before and after an attack. Our approach adapts the exploit used in the attack to the same or similar contracts before and after the attack to safeguard the assets. We conceptualize adapting exploits as a program repair problem and apply established techniques to implement our approach into a full-fledged framework, BACKRUNNER. Running on previous attacks in 2023, BACKRUNNER can successfully rescue more than \$410M. In the real world, it has helped rescue over \$11.2M worth of assets in 28 separate incidents within two months.

cs.CR↗

arXiv2024

CHisIEC: An Information Extraction Corpus for Ancient Chinese History

Natural Language Processing (NLP) plays a pivotal role in the realm of Digital Humanities (DH) and serves as the cornerstone for advancing the structural analysis of historical and cultural heritage texts. This is particularly true for the domains of named entity recognition (NER) and relation extraction (RE). In our commitment to expediting ancient history and culture, we present the ``Chinese Historical Information Extraction Corpus''(CHisIEC). CHisIEC is a meticulously curated dataset designed to develop and evaluate NER and RE tasks, offering a resource to facilitate research in the field. Spanning a remarkable historical timeline encompassing data from 13 dynasties spanning over 1830 years, CHisIEC epitomizes the extensive temporal range and text heterogeneity inherent in Chinese historical documents. The dataset encompasses four distinct entity types and twelve relation types, resulting in a meticulously labeled dataset comprising 14,194 entities and 8,609 relations. To establish the robustness and versatility of our dataset, we have undertaken comprehensive experimentation involving models of various sizes and paradigms. Additionally, we have evaluated the capabilities of Large Language Models (LLMs) in the context of tasks related to ancient Chinese history. The dataset and code are available at \url{https://github.com/tangxuemei1995/CHisIEC}.

cs.CL↗

arXiv2024

Restoring Ancient Ideograph: A Multimodal Multitask Neural Network Approach

Cultural heritage serves as the enduring record of human thought and history. Despite significant efforts dedicated to the preservation of cultural relics, many ancient artefacts have been ravaged irreversibly by natural deterioration and human actions. Deep learning technology has emerged as a valuable tool for restoring various kinds of cultural heritages, including ancient text restoration. Previous research has approached ancient text restoration from either visual or textual perspectives, often overlooking the potential of synergizing multimodal information. This paper proposes a novel Multimodal Multitask Restoring Model (MMRM) to restore ancient texts, particularly emphasising the ideograph. This model combines context understanding with residual visual information from damaged ancient artefacts, enabling it to predict damaged characters and generate restored images simultaneously. We tested the MMRM model through experiments conducted on both simulated datasets and authentic ancient inscriptions. The results show that the proposed method gives insightful restoration suggestions in both simulation experiments and real-world scenarios. To the best of our knowledge, this work represents the pioneering application of multimodal deep learning in ancient text restoration, which will contribute to the understanding of ancient society and culture in digital humanities fields.

cs.CL↗

arXiv2024

ReFlow-TTS: A Rectified Flow Model for High-fidelity Text-to-Speech

The diffusion models including Denoising Diffusion Probabilistic Models (DDPM) and score-based generative models have demonstrated excellent performance in speech synthesis tasks. However, its effectiveness comes at the cost of numerous sampling steps, resulting in prolonged sampling time required to synthesize high-quality speech. This drawback hinders its practical applicability in real-world scenarios. In this paper, we introduce ReFlow-TTS, a novel rectified flow based method for speech synthesis with high-fidelity. Specifically, our ReFlow-TTS is simply an Ordinary Differential Equation (ODE) model that transports Gaussian distribution to the ground-truth Mel-spectrogram distribution by straight line paths as much as possible. Furthermore, our proposed approach enables high-quality speech synthesis with a single sampling step and eliminates the need for training a teacher model. Our experiments on LJSpeech Dataset show that our ReFlow-TTS method achieves the best performance compared with other diffusion based models. And the ReFlow-TTS with one step sampling achieves competitive performance compared with existing one-step TTS models.

cs.SD↗

arXiv2023

Knowledge-Aware Artifact Image Synthesis with LLM-Enhanced Prompting and Multi-Source Supervision

Ancient artifacts are an important medium for cultural preservation and restoration. However, many physical copies of artifacts are either damaged or lost, leaving a blank space in archaeological and historical studies that calls for artifact image generation techniques. Despite the significant advancements in open-domain text-to-image synthesis, existing approaches fail to capture the important domain knowledge presented in the textual description, resulting in errors in recreated images such as incorrect shapes and patterns. In this paper, we propose a novel knowledge-aware artifact image synthesis approach that brings lost historical objects accurately into their visual forms. We use a pretrained diffusion model as backbone and introduce three key techniques to enhance the text-to-image generation framework: 1) we construct prompts with explicit archaeological knowledge elicited from large language models (LLMs); 2) we incorporate additional textual guidance to correlated historical expertise in a contrastive manner; 3) we introduce further visual-semantic constraints on edge and perceptual features that enable our model to learn more intricate visual details of the artifacts. Compared to existing approaches, our proposed model produces higher-quality artifact images that align better with the implicit details and historical knowledge contained within written documents, thus achieving significant improvements across automatic metrics and in human evaluation. Our code and data are available at https://github.com/danielwusg/artifact_diffusion.

cs.CV↗

arXiv2023

DiffuVST: Narrating Fictional Scenes with Global-History-Guided Denoising Models

Recent advances in image and video creation, especially AI-based image synthesis, have led to the production of numerous visual scenes that exhibit a high level of abstractness and diversity. Consequently, Visual Storytelling (VST), a task that involves generating meaningful and coherent narratives from a collection of images, has become even more challenging and is increasingly desired beyond real-world imagery. While existing VST techniques, which typically use autoregressive decoders, have made significant progress, they suffer from low inference speed and are not well-suited for synthetic scenes. To this end, we propose a novel diffusion-based system DiffuVST, which models the generation of a series of visual descriptions as a single conditional denoising process. The stochastic and non-autoregressive nature of DiffuVST at inference time allows it to generate highly diverse narratives more efficiently. In addition, DiffuVST features a unique design with bi-directional text history guidance and multimodal adapter modules, which effectively improve inter-sentence coherence and image-to-text fidelity. Extensive experiments on the story generation task covering four fictional visual-story datasets demonstrate the superiority of DiffuVST over traditional autoregressive models in terms of both text quality and inference speed.

cs.CL↗

arXiv2023

An Ultra-fast Quantum Random Number Generation Scheme Based on Laser Phase Noise

Based on the intrinsic random property of quantum mechanics, quantum random number generators allow for access of truly unpredictable random sequence and are now heading towards high performance and small miniaturization, among which a popular scheme is based on the laser phase noise. However, this scheme is generally limited in speed and implementation complexity, especially for chip integration. In this work, a general physical model based on wiener process for such schemes is introduced, which provides an approach to clearly explain the limitation on the generation rate and comprehensively optimize the system performance. We present an insight to exploit the potential bandwidth of the quantum entropy source that contains plentiful quantum randomness with a simple spectral filtering method and experimentally boost the bandwidth of the corresponding quantum entropy source to 20 GHz, based on which an ultra-fast generation rate of 218 Gbps is demonstrated, setting a new record for laser phase noise based schemes by one order of magnitude. Our proposal significantly enhances the ceiling speed of such schemes without requiring extra complex hardware, thus effectively benefits the corresponding chip integration with high performance and low implementation cost, which paves the way for its large-scale applications.

quant-ph↗

arXiv2023

Self-Evolved Diverse Data Sampling for Efficient Instruction Tuning

Enhancing the instruction-following ability of Large Language Models (LLMs) primarily demands substantial instruction-tuning datasets. However, the sheer volume of these imposes a considerable computational burden and annotation cost. To investigate a label-efficient instruction tuning method that allows the model itself to actively sample subsets that are equally or even more effective, we introduce a self-evolving mechanism DiverseEvol. In this process, a model iteratively augments its training subset to refine its own performance, without requiring any intervention from humans or more advanced LLMs. The key to our data sampling technique lies in the enhancement of diversity in the chosen subsets, as the model selects new data points most distinct from any existing ones according to its current embedding space. Extensive experiments across three datasets and benchmarks demonstrate the effectiveness of DiverseEvol. Our models, trained on less than 8% of the original dataset, maintain or improve performance compared with finetuning on full data. We also provide empirical evidence to analyze the importance of diversity in instruction data and the iterative scheme as opposed to one-time sampling. Our code is publicly available at https://github.com/OFA-Sys/DiverseEvol.git.

cs.CL↗

arXiv2023

Strategy evolution on dynamic networks

Models of strategy evolution on static networks help us understand how population structure can promote the spread of traits like cooperation. One key mechanism is the formation of altruistic spatial clusters, where neighbors of a cooperative individual are likely to reciprocate, which protects prosocial traits from exploitation. But most real-world interactions are ephemeral and subject to exogenous restructuring, so that social networks change over time. Strategic behavior on dynamic networks is difficult to study, and much less is known about the resulting evolutionary dynamics. Here, we provide an analytical treatment of cooperation on dynamic networks, allowing for arbitrary spatial and temporal heterogeneity. We show that transitions among a large class of network structures can favor the spread of cooperation, even if each individual social network would inhibit cooperation when static. Furthermore, we show that spatial heterogeneity tends to inhibit cooperation, whereas temporal heterogeneity tends to promote it. Dynamic networks can have profound effects on the evolution of prosocial traits, even when individuals have no agency over network structures.

physics.soc-ph↗

arXiv2023

Automatic lobe segmentation using attentive cross entropy and end-to-end fissure generation

The automatic lung lobe segmentation algorithm is of great significance for the diagnosis and treatment of lung diseases, however, which has great challenges due to the incompleteness of pulmonary fissures in lung CT images and the large variability of pathological features. Therefore, we propose a new automatic lung lobe segmentation framework, in which we urge the model to pay attention to the area around the pulmonary fissure during the training process, which is realized by a task-specific loss function. In addition, we introduce an end-to-end pulmonary fissure generation method in the auxiliary pulmonary fissure segmentation task, without any additional network branch. Finally, we propose a registration-based loss function to alleviate the convergence difficulty of the Dice loss supervised pulmonary fissure segmentation task. We achieve 97.83% and 94.75% dice scores on our private dataset STLB and public LUNA16 dataset respectively.

eess.IV↗

arXiv2023

Multi-site, Multi-domain Airway Tree Modeling (ATM'22): A Public Benchmark for Pulmonary Airway Segmentation

Open international challenges are becoming the de facto standard for assessing computer vision and image analysis algorithms. In recent years, new methods have extended the reach of pulmonary airway segmentation that is closer to the limit of image resolution. Since EXACT'09 pulmonary airway segmentation, limited effort has been directed to quantitative comparison of newly emerged algorithms driven by the maturity of deep learning based approaches and clinical drive for resolving finer details of distal airways for early intervention of pulmonary diseases. Thus far, public annotated datasets are extremely limited, hindering the development of data-driven methods and detailed performance evaluation of new algorithms. To provide a benchmark for the medical imaging community, we organized the Multi-site, Multi-domain Airway Tree Modeling (ATM'22), which was held as an official challenge event during the MICCAI 2022 conference. ATM'22 provides large-scale CT scans with detailed pulmonary airway annotation, including 500 CT scans (300 for training, 50 for validation, and 150 for testing). The dataset was collected from different sites and it further included a portion of noisy COVID-19 CTs with ground-glass opacity and consolidation. Twenty-three teams participated in the entire phase of the challenge and the algorithms for the top ten teams are reviewed in this paper. Quantitative and qualitative results revealed that deep learning models embedded with the topological continuity enhancement achieved superior performance in general. ATM'22 challenge holds as an open-call design, the training data and the gold standard evaluation are available upon successful registration via its homepage.

eess.IV↗