arXiv subjects

Yan Liu

Publications and source records attributed to Yan Liu.

At least 145 records · Page 8Linked to original sources

arXiv2025

Energy transport in holographic non-conformal interfaces

We study energy transport in a system of two dimensional conformal field theories exchanging energy across a non-conformal interface involving a localised scalar operator, using holographic duality. By imposing the sourceless boundary condition, or equivalently, enforcing energy conservation at the interface, we show that the sum of the transmission and reflection coefficients is equal to one. Unlike conformal interfaces, we find that both the energy transmission and reflection coefficients are generally complex and frequency dependent. When the interface brane connects two distinct AdS$_2$ geometries, the transmission coefficient approaches the value expected for a conformal interface in the UV regime at high frequencies and in the IR regime at low frequencies. In the intermediate frequency range, the transmission coefficient may exhibit oscillatory behavior. Moreover, we present a nontrivial example of a fully transmissive interface, which exhibits similarities to a topological interface.

hep-th↗

arXiv2025

MusicScaffold: Bridging Machine Efficiency and Human Growth in Adolescent Creative Education through Generative AI

Adolescence is marked by strong creative impulses but limited strategies for structured expression, often leading to frustration or disengagement. While generative AI lowers technical barriers and delivers efficient outputs, its role in fostering adolescents' expressive growth has been overlooked. We propose MusicScaffold, the first adolescent-centered framework that repositions AI as a guide, coach, and partner, making expressive strategies transparent and learnable, and supporting autonomy. In a four-week study with middle school students (ages 12--14), MusicScaffold enhanced cognitive specificity, behavioral self-regulation, and affective confidence in music creation. By reframing generative AI as a scaffold rather than a generator, this work bridges the machine efficiency of generative systems with human growth in adolescent creative education.

cs.HC↗

arXiv2025

Who Decides How Knowing Becomes Doing? Redistributing Authority in Human-AI Music Co-Creation

In the era of human-AI co-creation, the maxim "knowing is easy, doing is hard" is redefined. AI has the potential to ease execution, yet the essence of "hard" lies in who governs the translation from knowing to doing. Mainstream tools often centralize interpretive authority and homogenize expression, suppressing marginal voices. To address these challenges, we introduce the first systematic framework for redistributing authority in the knowing-doing cycle, built on three principles, namely contestability, agency, and plurality. Through interactive studies with 180 music practitioners, complemented by in-depth interviews, we demonstrate that these principles reshape human-AI authority relations and reactivate human creative expression. The findings establish a new paradigm for critical computing and human-AI co-creation that advances from critique to practice.

cs.HC↗

arXiv2025

Thermalization dynamics of finite-size quantum critical systems

Using holographic duality, we investigate thermalization process when two finite-size quantum critical systems are brought into thermal contact along a perfectly transmitting interface. Through real-time simulations of gravitational dynamics, which are spatially inhomogeneous and anisotropic and are confined within two dynamical bulk branes, we identify three distinct thermalization patterns governed by the energy imbalance (temperature difference) and system size. For systems with large size and small energy imbalance, we observe recurrent cycles of formation and collapse of non-equilibrium steady states (NESS). Under large energy imbalance, shock waves persist for a prolonged period with sustained boundary reflections, while rarefaction waves rapidly homogenize. When the system size is sufficiently small, dissipation dominates and leads to oscillatory decay without sustained NESS or shock structure. In sharp contrast to diffusive systems, we uncover that wave-propagated energy transfer together with boundary reflections enables nearly complete energy swapping between subsystems during thermalization. Our results reveal rich thermalization dynamics in finite-size quantum critical systems across spatial scales and energy gradient regimes.

hep-th↗

arXiv2025

When LLM Meets Time Series: Can LLMs Perform Multi-Step Time Series Reasoning and Inference

The rapid advancement of Large Language Models (LLMs) has sparked growing interest in their application to time series analysis tasks. However, their ability to perform complex reasoning over temporal data in real-world application domains remains underexplored. To move toward this goal, a first step is to establish a rigorous benchmark dataset for evaluation. In this work, we introduce the TSAIA Benchmark, a first attempt to evaluate LLMs as time-series AI assistants. To ensure both scientific rigor and practical relevance, we surveyed over 20 academic publications and identified 33 real-world task formulations. The benchmark encompasses a broad spectrum of challenges, ranging from constraint-aware forecasting to anomaly detection with threshold calibration: tasks that require compositional reasoning and multi-step time series analysis. The question generator is designed to be dynamic and extensible, supporting continuous expansion as new datasets or task types are introduced. Given the heterogeneous nature of the tasks, we adopt task-specific success criteria and tailored inference-quality metrics to ensure meaningful evaluation for each task. We apply this benchmark to assess eight state-of-the-art LLMs under a unified evaluation protocol. Our analysis reveals limitations in current models' ability to assemble complex time series analysis workflows, underscoring the need for specialized methodologies for domain-specific adaptation. Our benchmark is available at https://huggingface.co/datasets/Melady/TSAIA, and the code is available at https://github.com/USC-Melady/TSAIA.

cs.LG↗

arXiv2025

CompLex: Music Theory Lexicon Constructed by Autonomous Agents for Automatic Music Generation

Generative artificial intelligence in music has made significant strides, yet it still falls short of the substantial achievements seen in natural language processing, primarily due to the limited availability of music data. Knowledge-informed approaches have been shown to enhance the performance of music generation models, even when only a few pieces of musical knowledge are integrated. This paper seeks to leverage comprehensive music theory in AI-driven music generation tasks, such as algorithmic composition and style transfer, which traditionally require significant manual effort with existing techniques. We introduce a novel automatic music lexicon construction model that generates a lexicon, named CompLex, comprising 37,432 items derived from just 9 manually input category keywords and 5 sentence prompt templates. A new multi-agent algorithm is proposed to automatically detect and mitigate hallucinations. CompLex demonstrates impressive performance improvements across three state-of-the-art text-to-music generation models, encompassing both symbolic and audio-based methods. Furthermore, we evaluate CompLex in terms of completeness, accuracy, non-redundancy, and executability, confirming that it possesses the key characteristics of an effective lexicon.

cs.SD↗

arXiv2025

Objective Task-based Evaluation of Quantitative Medical Imaging Methods: Emerging Frameworks and Future Directions

Quantitative imaging (QI) is demonstrating strong promise across multiple clinical applications. For clinical translation of QI methods, objective evaluation on clinically relevant tasks is essential. To address this need, multiple evaluation strategies are being developed. In this paper, based on previous literature, we outline four emerging frameworks to perform evaluation studies of QI methods. We first discuss the use of virtual imaging trials (VITs) to evaluate QI methods. Next, we outline a no-gold-standard evaluation framework to clinically evaluate QI methods without ground truth. Third, a framework to evaluate QI methods for joint detection and quantification tasks is outlined. Finally, we outline a framework to evaluate QI methods that output multi-dimensional parameters, such as radiomic features. We review these frameworks, discussing their utilities and limitations. Further, we examine future research areas in evaluation of QI methods. Given the recent advancements in PET, including long axial field-of-view scanners and the development of artificial-intelligence algorithms, we present these frameworks in the context of PET.

physics.med-ph↗

arXiv2025

Atomistic mechanisms of phase transitions in all-temperature barocaloric material KPF$_6$

Conventional barocaloric materials typically exhibit limited operating temperature ranges. In contrast, KPF$_6$ has recently been reported to achieve an exceptional all-temperature barocaloric effect (BCE) via pressure-driven phase transitions. Here, we elucidate the atomistic mechanisms underlying the phase transitions through first-principles calculations and machine-learning potential accelerated molecular dynamics simulations. We identify four distinct phases: the room-temperature cubic (C) plastic crystal characterized by strong fluorine orientational disorder (FOD) and anharmonicity, the intermediate-temperature monoclinic (M-II) phase with decreasing FOD, the low-temperature monoclinic (M-I) phase with suppressed FOD, and the fully ordered rhombohedral (R) phase under pressure. Phonon calculations confirm the dynamic stability of the M-II, M-I, and R phases at 0 K, whereas the C phase requires thermal fluctuations for stabilization. Under pressure, all the C, M-II, and M-I phases transform to the R phase, which are driven by cooperative PF$_6$ octahedral rotations coupled with lattice modulations. These pressure-induced phase transitions result in persistent isothermal entropy changes across a wide temperature range, thereby explaining the experimentally observed all-temperature BCE in this material. Hybrid functional calculations reveal wide-bandgap insulating behavior across all phases. This work deciphers the interplay between FOD, anharmonicity, and phase transitions in KPF$_6$, providing important insights for the design of BCE materials with broad operational temperature spans.

cond-mat.mtrl-sci↗

arXiv2025

Adaptive sparsening and smoothing of the treatment model for longitudinal causal inference using outcome-adaptive LASSO and marginal fused LASSO

Causal variable selection in time-varying treatment settings is challenging due to evolving confounding effects. Existing methods mainly focus on time-fixed exposures and are not directly applicable to time-varying scenarios. We propose a novel two-step procedure for variable selection when modeling the treatment probability at each time point. We first introduce a novel approach to longitudinal confounder selection using a Longitudinal Outcome Adaptive LASSO (LOAL) that will data-adaptively select covariates with theoretical justification of variance reduction of the estimator of the causal effect. We then propose an Adaptive Fused LASSO that can collapse treatment model parameters over time points with the goal of simplifying the models in order to improve the efficiency of the estimator while minimizing model misspecification bias compared with naive pooled logistic regression models. Our simulation studies highlight the need for and usefulness of the proposed approach in practice. We implemented our method on data from the Nicotine Dependence in Teens study to estimate the effect of the timing of alcohol initiation during adolescence on depressive symptoms in early adulthood.

stat.ME↗

arXiv2025

PTSM: Physiology-aware and Task-invariant Spatio-temporal Modeling for Cross-Subject EEG Decoding

Cross-subject electroencephalography (EEG) decoding remains a fundamental challenge in brain-computer interface (BCI) research due to substantial inter-subject variability and the scarcity of subject-invariant representations. This paper proposed PTSM (Physiology-aware and Task-invariant Spatio-temporal Modeling), a novel framework for interpretable and robust EEG decoding across unseen subjects. PTSM employs a dual-branch masking mechanism that independently learns personalized and shared spatio-temporal patterns, enabling the model to preserve individual-specific neural characteristics while extracting task-relevant, population-shared features. The masks are factorized across temporal and spatial dimensions, allowing fine-grained modulation of dynamic EEG patterns with low computational overhead. To further address representational entanglement, PTSM enforces information-theoretic constraints that decompose latent embeddings into orthogonal task-related and subject-related subspaces. The model is trained end-to-end via a multi-objective loss integrating classification, contrastive, and disentanglement objectives. Extensive experiments on cross-subject motor imagery datasets demonstrate that PTSM achieves strong zero-shot generalization, outperforming state-of-the-art baselines without subject-specific calibration. Results highlight the efficacy of disentangled neural representations for achieving both personalized and transferable decoding in non-stationary neurophysiological settings.

cs.LG↗

arXiv2025

Reducing AoI and Improving Throughput for NOMA-assisted SGF Systems: A Hierarchical Learning Approach

A non-orthogonal multiple access (NOMA) assisted semi-grant-free (SGF) framework is proposed to enable channel access for grant-free users (GFUs) by using residual resources from grant-based users. Under this framework, the problem of joint beamforming design and transmission scheduling is formulated to improve the system throughput and reduce the age-of-information of GFUs. The aforementioned problem is transferred into a Markov Decision Process to model the changing environment with the transmission/ waiting/ retransmission of GFUs. In an effort to solve the pertinent problem, firstly, a deep reinforcement learning (DRL) based transmission scheduling approach is proposed for determining the optimal transmission probability based on the available transmission slots and transmission status of GFUs. Secondly, a hierarchical learning algorithm is proposed to analyze the channel state information of GBUs and the transmission status of GFUs, and to train an upper-level policy based on this analysis for beamforming to achieve efficient grant-based transmission, while a lower-level policy adapts to maximize the utilization of transmission slots allocated by the upper-level agent. The two policies interact to improve channel access and avoid collisions. Numerical results reveal that 1) The DRL based transmission scheduling outperforms existing adaptive and state-dependent baselines in AoI reduction, where an average three-time-slots-earlier-transmission can be obtained compared to the state-dependent choice, and five time slots earlier can be achieved when comparing to the adaptive choice; 2) The hierarchical learning algorithm is able to achieve approximately a 31.82% gain while maintaining the average AoI of GFUs within 1.5 time slots. 3) The effectiveness of the hierarchical learning scheme in NOMA-assisted SGF system is validated across scenarios with GFUs counts from 1-5 times of GBUs.

eess.SP↗

arXiv2025

Foundation Models for Demand Forecasting via Dual-Strategy Ensembling

Accurate demand forecasting is critical for supply chain optimization, yet remains difficult in practice due to hierarchical complexity, domain shifts, and evolving external factors. While recent foundation models offer strong potential for time series forecasting, they often suffer from architectural rigidity and limited robustness under distributional change. In this paper, we propose a unified ensemble framework that enhances the performance of foundation models for sales forecasting in real-world supply chains. Our method combines two complementary strategies: (1) Hierarchical Ensemble (HE), which partitions training and inference by semantic levels (e.g., store, category, department) to capture localized patterns; and (2) Architectural Ensemble (AE), which integrates predictions from diverse model backbones to mitigate bias and improve stability. We conduct extensive experiments on the M5 benchmark and three external sales datasets, covering both in-domain and zero-shot forecasting. Results show that our approach consistently outperforms strong baselines, improves accuracy across hierarchical levels, and provides a simple yet effective mechanism for boosting generalization in complex forecasting environments.

cs.LG↗

arXiv2025

Simultaneous All-Pay Auctions with Budget Constraints

The all-pay auction, a classic competitive model, is widely applied in scenarios such as political elections, sports competitions, and research and development, where all participants pay their bids regardless of winning or losing. However, in the traditional all-pay auction, players have no budget constraints, whereas in real-world scenarios, players typically face budget constraints. This paper studies the Nash equilibrium of two players with budget constraints across multiple heterogeneous items in a complete-information framework. The main contributions are as follows: (1) a comprehensive characterization of the Nash equilibrium in single-item auctions with asymmetric budgets and valuations; (2) the construction of a joint distribution Nash equilibrium for the two-item scenario; and (3) the construction of a joint distribution Nash equilibrium for the three-item scenario. Unlike the unconstrained all-pay auction, which always has a Nash equilibrium, a Nash equilibrium may not exist when players have budget constraints. Our findings highlight the intricate effects of budget constraints on bidding strategies, providing new perspectives and methodologies for theoretical analysis and practical applications of all-pay auctions.

econ.TH↗

arXiv2025

Understanding the Dark Side of LLMs' Intrinsic Self-Correction

Intrinsic self-correction was proposed to improve LLMs' responses via feedback prompts solely based on their inherent capability. However, recent works show that LLMs' intrinsic self-correction fails without oracle labels as feedback prompts. In this paper, we aim to interpret LLMs' intrinsic self-correction for different tasks, especially for those failure cases. By including one simple task and three complex tasks with state-of-the-art (SOTA) LLMs like ChatGPT families (o1, 4o, 3.5-turbo) and Llama families (2-7B, 3-8B, and 3.1-8B), we design three interpretation methods to reveal the dark side of LLMs' intrinsic self-correction. We identify intrinsic self-correction can (1) cause LLMs to waver both intermedia and final answers and lead to prompt bias on simple factual questions; (2) introduce human-like cognitive bias on complex tasks. In light of our findings, we also provide two simple yet effective strategies for alleviation: question repeating and supervised fine-tuning with a few samples. We open-source our work at https://x-isc.info/.

cs.CL↗

arXiv2025

All-optical Fourier neural network using partially coherent light

Optical neural networks present distinct advantages over traditional electrical counterparts, such as accelerated data processing and reduced energy consumption. While coherent light is conventionally employed in optical neural networks, our study proposes harnessing spatially incoherent light in all-optical Fourier neural networks. Contrary to numerical predictions of declining target recognition accuracy with increased incoherence, our experimental results demonstrate a surprising outcome: improved accuracy with incoherent light. We attribute this unexpected enhancement to spatially incoherent light's ability to alleviate experimental errors like diffraction rings, laser speckle, and edge effects. Our controlled experiments introduced spatial incoherence by passing monochromatic light through a spatial light modulator featuring a dynamically changing random phase array. These findings underscore partially coherent light's potential to optimize optical neural networks, delivering dependable and efficient solutions for applications demanding consistent accuracy and robustness across diverse conditions.

physics.optics↗

arXiv2025

InstCache: A Predictive Cache for LLM Serving

The revolutionary capabilities of Large Language Models (LLMs) are attracting rapidly growing popularity and leading to soaring user requests to inference serving systems. Caching techniques, which leverage data reuse to reduce computation, offer opportunities to optimize the performance of LLM inference engines. On the one hand, the low-level key-value (KV) cache working at the token level is widely adopted, albeit it incurs significant overhead as request volume grows. On the other hand, instruction-level caching, which stores full instruction-response pairs, is expected to play an increasingly crucial role. However, the high variability in the content and length of instructions make it rare for identical instructions to recur within a short time window, presenting challenges for effective caching instruction-response pairs. To address this challenge, we propose InstCache, a predictive caching mechanism for LLM serving systems. Leveraging the capability of LLMs, we can effectively reorder the representation space of instruction texts and develop a sufficient level of spatial locality. Such spatial locality enables us to predict potential instructions located in a compact region in the space, resulting in an effective caching system at runtime. Experimental results demonstrate that InstCache achieves a 2.3x higher hit rate compared to the upper bound of traditional caching mechanisms on WildChat dataset and reduces the time per output token of vLLM by up to 42.0% and 50.0% on LMSys and Moss datasets, respectively.

cs.CL↗

arXiv2025

Rattan: An Extensible and Scalable Modular Internet Path Emulator

The rapid growth of Internet paths in heterogeneity, scale, and dynamics has made existing emulators increasingly insufficient in flexibility, scalability, and usability. To address these limitations, we present Rattan, an extensible and scalable software network path emulator for modern Internet conditions. Rattan's core innovation lies in its cell-based architecture: by splitting emulation functions into modular "cells" with well-documented asynchronous interfaces, users are allowed to easily compose different cells by hierarchically linking them and easily construct new cells by using standard cell interfaces. This design enables: (1) scalability, supporting hundreds of concurrent gigabit-level paths on a single machine and cluster-level experiments composed of multiple machines; (2) extensibility, simulating new network conditions by constructing new cells. Rattan empowers developers and researchers to efficiently and confidently evaluate, validate, and diagnose diverse network transport innovations for online services.

cs.NI↗

arXiv2025

Dual-Alignment Knowledge Retention for Continual Medical Image Segmentation

Continual learning in medical image segmentation involves sequential data acquisition across diverse domains (e.g., clinical sites), where task interference between past and current domains often leads to catastrophic forgetting. Existing continual learning methods fail to capture the complex dependencies between tasks. We introduce a novel framework that mitigates forgetting by establishing and enhancing complex dependencies between historical data and the network in the present task. Our framework features a dual-alignment strategy, the cross-network alignment (CNA) module aligns the features extracted from the bottleneck layers of the current and previous networks, respectively, while the cross-representation alignment (CRA) module aligns the features learned by the current network from historical buffered data and current input data, respectively. Implementing both types of alignment is a non-trivial task. To address this, we further analyze the linear and nonlinear forms of the well-established Hilbert-Schmidt Independence Criterion (HSIC) and deliberately design feature mapping and feature pairing blocks within the CRA module. Experiments on medical image segmentation task demonstrate our framework's effectiveness in mitigating catastrophic forgetting under domain shifts.

eess.IV↗