arXiv subjects

Chen Liang

Publications and source records attributed to Chen Liang.

At least 73 records · Page 4Linked to original sources

arXiv2025

Atomistic mechanisms of viscosity in 2D liquid-like fluids

Shear viscosity plays a fundamental role in fluid dynamics from heavy-ion collisions to biological processes. Still, its microscopic mechanisms at the individual particle kinetic level remain a subject of ongoing research, specially in dense systems. In this work, we systematically investigate the shear viscosity ($η$) of two-dimensional (2D) simple fluids using computer simulations of Lennard-Jones, Yukawa, and one-component plasma systems. By combining Frenkel's liquid description, consisting of solid-like quasi-harmonic vibrations interrupted by thermally activated hops, with the concept of lifetime of local atomic connectivity $τ_{LC}$, we find a surprisingly simple formula for the kinematic viscosity that is solely determined by $τ_{LC}$ and the average kinetic particle speed $\bar{v}_p$. The derived analytical expression provides a direct link between macroscopic and microscopic dynamics, which shows excellent agreement with the simulation data in the dense liquid-like regime in all the 2D fluids considered. Moreover, it is discovered that, $τ_{LC}$ in 2D fluids is universally determined by the effective potential difference between the first peak and valley of the pair correlation function, implying a direct connection between macroscopic shear transport and microscopic structure. Finally, we demonstrate that the characteristic length scale $l_p= \bar{v}_p τ_{LC}$, which governs the macroscopic shear viscosity, aligns with the elastic length-scale that defines the propagation limit of collective shear waves in liquids. These findings establish that shear viscosity in 2D fluids arises from the diffusive transport of average particle momentum across the elastic length scale. Moreover, they highlight that shear dynamics are fundamentally governed by localized configurational excitations within the atomic connectivity network.

cond-mat.soft↗

arXiv2025

Molecule Graph Networks with Many-body Equivariant Interactions

Message passing neural networks have demonstrated significant efficacy in predicting molecular interactions. Introducing equivariant vectorial representations augments expressivity by capturing geometric data symmetries, thereby improving model accuracy. However, two-body bond vectors in opposition may cancel each other out during message passing, leading to the loss of directional information on their shared node. In this study, we develop Equivariant N-body Interaction Networks (ENINet) that explicitly integrates l = 1 equivariant many-body interactions to enhance directional symmetric information in the message passing scheme. We provided a mathematical analysis demonstrating the necessity of incorporating many-body equivariant interactions and generalized the formulation to $N$-body interactions. Experiments indicate that integrating many-body equivariant representations enhances prediction accuracy across diverse scalar and tensorial quantum chemical properties.

cs.LG↗

arXiv2025

HandProxy: Expanding the Affordances of Speech Interfaces in Immersive Environments with a Virtual Proxy Hand

Hand interactions are increasingly used as the primary input modality in immersive environments, but they are not always feasible due to situational impairments, motor limitations, and environmental constraints. Speech interfaces have been explored as an alternative to hand input in research and commercial solutions, but are limited to initiating basic hand gestures and system controls. We introduce HandProxy, a system that expands the affordances of speech interfaces to support expressive hand interactions. Instead of relying on predefined speech commands directly mapped to possible interactions, HandProxy enables users to control the movement of a virtual hand as an interaction proxy, allowing them to describe the intended interactions naturally while the system translates speech into a sequence of hand controls for real-time execution. A user study with 20 participants demonstrated that HandProxy effectively enabled diverse hand interactions in virtual environments, achieving a 100% task completion rate with an average of 1.09 attempts per speech command and 91.8% command execution accuracy, while supporting flexible, natural speech input with varying levels of control and granularity.

cs.HC↗

arXiv2025

Subharmonic Shapiro steps in depinning dynamics of a 2D solid dusty plasma modulated by 1D nonlinear deformed periodic substrates

Langevin dynamical simulations are performed to investigate the depinning dynamics of a two-dimensional solid dusty plasma, which is modulated by one-dimensional nonlinear deformed periodic substrates, and also driven by the combination of the DC and AC forces. As the DC driving force increases gradually, pronounced subharmonic and harmonic Shapiro steps are discovered under the combined modulation of the deformed substrate and the external AC drive. These observed subharmonic and harmonic Shapiro steps are attributed to the dynamic mode locking. The data analysis results indicate that the nonlinear deformed substrate strongly influences these subharmonic Shapiro steps, which can be accurately diagnosed using the fraction of sixfold coordinated particles. Furthermore, the diagnostic of the kinetic temperature clearly indicates the difference between harmonic and subharmonic Shapiro steps, i.e., the particle motion is slightly less synchronized at subharmonic Shapiro steps, caused by the unstable locations of the deformed substrate.

physics.plasm-ph↗

arXiv2025

SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation

The Mixture of Experts (MoE) architecture has emerged as a powerful paradigm for scaling large language models (LLMs) while maintaining inference efficiency. However, their enormous memory requirements make them prohibitively expensive to fine-tune or deploy in resource-constrained environments. To address this challenge, we introduce SlimMoE, a multi-stage compression framework for transforming large MoE models into much smaller, efficient variants without incurring the prohibitive costs of training from scratch. Our method systematically reduces parameter counts by slimming experts and transferring knowledge through intermediate stages, effectively mitigating the performance degradation common in one-shot pruning approaches. Using this framework, we compress Phi 3.5-MoE (41.9B total/6.6B activated parameters) to create Phi-mini-MoE (7.6B total/2.4B activated parameters) and Phi-tiny-MoE (3.8B total/1.1B activated parameters) using only 400B tokens--less than 10% of the original model's training data. These compressed models can be fine-tuned on a single GPU (A100 for Phi-mini-MoE, A6000 for Phi-tiny-MoE), making them highly suitable for academic and resource-limited settings. Our experiments demonstrate that these compressed models outperform others of similar size and remain competitive with larger models. For instance, Phi-mini-MoE achieves similar or better performance to Phi-3-mini using only 2/3 of the activated parameters and yields comparable MMLU scores to Llama 3.1 8B despite having significantly lower latency. Our findings demonstrate that structured pruning combined with staged distillation offers an effective path to creating high-quality, compact MoE models, paving the way for broader adoption of MoE architectures. We make our models publicly available at https://huggingface.co/microsoft/Phi-mini-MoE-instruct and https://huggingface.co/microsoft/Phi-tiny-MoE-instruct .

cs.LG↗

arXiv2025

Physics-informed mixture of experts network for interpretable battery degradation trajectory computation amid second-life complexities

Retired electric vehicle batteries offer immense potential to support low-carbon energy systems, but uncertainties in their degradation behavior and data inaccessibilities under second-life use pose major barriers to safe and scalable deployment. This work proposes a Physics-Informed Mixture of Experts (PIMOE) network that computes battery degradation trajectories using partial, field-accessible signals in a single cycle. PIMOE leverages an adaptive multi-degradation prediction module to classify degradation modes using expert weight synthesis underpinned by capacity-voltage and relaxation data, producing latent degradation trend embeddings. These are input to a use-dependent recurrent network for long-term trajectory prediction. Validated on 207 batteries across 77 use conditions and 67,902 cycles, PIMOE achieves an average mean absolute percentage (MAPE) errors of 0.88% with a 0.43 ms inference time. Compared to the state-of-the-art Informer and PatchTST, it reduces computational time and MAPE by 50%, respectively. Compatible with random state of charge region sampling, PIMOE supports 150-cycle forecasts with 1.50% average and 6.26% maximum MAPE, and operates effectively even with pruned 5MB training data. Broadly, PIMOE framework offers a deployable, history-free solution for battery degradation trajectory computation, redefining how second-life energy storage systems are assessed, optimized, and integrated into the sustainable energy landscape.

cs.LG↗

arXiv2025

Multiscale transform based seismic reflectivity inversion using convolutional neural network

The Multiscale Fourier Transform of a seismic trace performs time-frequency analyses over a range of window lengths. The variation in window length captures local and global relative amplitudes between events, thereby allowing reflectivity inversion that is independent of the amplitude spectrum of the seismic wavelet. As the temporal and spatial variation of the actual seismic wavelet in seismic reflection data is poorly known, this approach has many advantages over conventional seismic reflectivity inversion. No wavelet extraction is performed. Thus, the inversion for reflectivity can be conducted without well control, seismic ties, or time-depth functions. The inversion is sparse, so no starting model is needed. Furthermore, as no wavelet is required, the inversion can be applied directly to depth migrated data. The phase of the wavelet is constrained by the assumption of sparse reflectivity and thus works best when earth impedance structure is blocky. Trace integration of the inverted reflectivity provides bandlimited impedance which compares very favorably to well-log bandlimited impedance for both synthetic and real data cases.

physics.geo-ph↗

arXiv2025

Unsupervised Multi-modal Feature Alignment for Time Series Representation Learning

In recent times, the field of unsupervised representation learning (URL) for time series data has garnered significant interest due to its remarkable adaptability across diverse downstream applications. Unsupervised learning goals differ from downstream tasks, making it tricky to ensure downstream task utility by focusing only on temporal feature characterization. Researchers have proposed multiple transformations to extract discriminative patterns implied in informative time series, trying to fill the gap. Despite the introduction of a variety of feature engineering techniques, e.g. spectral domain, wavelet transformed features, features in image form and symbolic features etc. the utilization of intricate feature fusion methods and dependence on heterogeneous features during inference hampers the scalability of the solutions. To address this, our study introduces an innovative approach that focuses on aligning and binding time series representations encoded from different modalities, inspired by spectral graph theory, thereby guiding the neural encoder to uncover latent pattern associations among these multi-modal features. In contrast to conventional methods that fuse features from multiple modalities, our proposed approach simplifies the neural architecture by retaining a single time series encoder, consequently leading to preserved scalability. We further demonstrate and prove mechanisms for the encoder to maintain better inductive bias. In our experimental evaluation, we validated the proposed method on a diverse set of time series datasets from various domains. Our approach outperforms existing state-of-the-art URL methods across diverse downstream tasks.

cs.LG↗

arXiv2025

Coherence-based Approximate Derivatives via Web of Affine Spaces Optimization

Computing derivatives is a crucial subroutine in computer science and related fields as it provides a local characterization of a function's steepest directions of ascent or descent. In this work, we recognize that derivatives are often not computed in isolation; conversely, it is quite common to compute a \textit{sequence} of derivatives, each one somewhat related to the last. Thus, we propose accelerating derivative computation by reusing information from previous, related calculations-a general strategy known as \textit{coherence}. We introduce the first instantiation of this strategy through a novel approach called the Web of Affine Spaces (WASP) Optimization. This approach provides an accurate approximation of a function's derivative object (i.e. gradient, Jacobian matrix, etc.) at the current input within a sequence. Each derivative within the sequence only requires a small number of forward passes through the function (typically two), regardless of the number of function inputs and outputs. We demonstrate the efficacy of our approach through several numerical experiments, comparing it with alternative derivative computation methods on benchmark functions. We show that our method significantly improves the performance of derivative computation on small to medium-sized functions, i.e., functions with approximately fewer than 500 combined inputs and outputs. Furthermore, we show that this method can be effectively applied in a robotics optimization context. We conclude with a discussion of the limitations and implications of our work. Open-source code, visual explanations, and videos are located at the paper website: \href{https://apollo-lab-yale.github.io/25-RSS-WASP-website/}{https://apollo-lab-yale.github.io/25-RSS-WASP-website/}.

cs.RO↗

arXiv2025

ad-trait: A Fast and Flexible Automatic Differentiation Library in Rust

The Rust programming language is an attractive choice for robotics and related fields, offering highly efficient and memory-safe code. However, a key limitation preventing its broader adoption in these domains is the lack of high-quality, well-supported Automatic Differentiation (AD)-a fundamental technique that enables convenient derivative computation by systematically accumulating data during function evaluation. In this work, we introduce ad-trait, a new Rust-based AD library. Our implementation overloads Rust's standard floating-point type with a flexible trait that can efficiently accumulate necessary information for derivative computation. The library supports both forward-mode and reverse-mode automatic differentiation, making it the first operator-overloading AD implementation in Rust to offer both options. Additionally, ad-trait leverages Rust's performance-oriented features, such as Single Instruction, Multiple Data acceleration in forward-mode AD, to enhance efficiency. Through benchmarking experiments, we show that our library is among the fastest AD implementations across several programming languages for computing derivatives. Moreover, it is already integrated into a Rust-based robotics library, where we showcase its ability to facilitate fast optimization procedures. We conclude with a discussion of the limitations and broader implications of our work.

cs.RO↗

arXiv2025

LLMs Can Generate a Better Answer by Aggregating Their Own Responses

Large Language Models (LLMs) have shown remarkable capabilities across tasks, yet they often require additional prompting techniques when facing complex problems. While approaches like self-correction and response selection have emerged as popular solutions, recent studies have shown these methods perform poorly when relying on the LLM itself to provide feedback or selection criteria. We argue this limitation stems from the fact that common LLM post-training procedures lack explicit supervision for discriminative judgment tasks. In this paper, we propose Generative Self-Aggregation (GSA), a novel prompting method that improves answer quality without requiring the model's discriminative capabilities. GSA first samples multiple diverse responses from the LLM, then aggregates them to obtain an improved solution. Unlike previous approaches, our method does not require the LLM to correct errors or compare response quality; instead, it leverages the model's generative abilities to synthesize a new response based on the context of multiple samples. While GSA shares similarities with the self-consistency (SC) approach for response aggregation, SC requires specific verifiable tokens to enable majority voting. In contrast, our approach is more general and can be applied to open-ended tasks. Empirical evaluation demonstrates that GSA effectively improves response quality across various tasks, including mathematical reasoning, knowledge-based problems, and open-ended generation tasks such as code synthesis and conversational responses.

cs.CL↗

arXiv2025

CRYSIM: Prediction of Symmetric Structures of Large Crystals with GPU-based Ising Machines

Solving black-box optimization problems with Ising machines is increasingly common in materials science. However, their application to crystal structure prediction (CSP) is still ineffective due to symmetry agnostic encoding of atomic coordinates. We introduce CRYSIM, an algorithm that encodes the space group, the Wyckoff positions combination, and coordinates of independent atomic sites as separate variables. This encoding reduces the search space substantially by exploiting the symmetry in space groups. When CRYSIM is interfaced to Fixstars Amplify, a GPU-based Ising machine, its prediction performance was competitive with CALYPSO and Bayesian optimization for crystals containing more than 150 atoms in a unit cell. Although it is not realistic to interface CRYSIM to current small-scale quantum devices, it has the potential to become the standard CSP algorithm in the coming quantum age.

cond-mat.mtrl-sci↗

arXiv2025

KDSelector: A Knowledge-Enhanced and Data-Efficient Model Selector Learning Framework for Time Series Anomaly Detection

Model selection has been raised as an essential problem in the area of time series anomaly detection (TSAD), because there is no single best TSAD model for the highly heterogeneous time series in real-world applications. However, despite the success of existing model selection solutions that train a classification model (especially neural network, NN) using historical data as a selector to predict the correct TSAD model for each series, the NN-based selector learning methods used by existing solutions do not make full use of the knowledge in the historical data and require iterating over all training samples, which limits the accuracy and training speed of the selector. To address these limitations, we propose KDSelector, a novel knowledge-enhanced and data-efficient framework for learning the NN-based TSAD model selector, of which three key components are specifically designed to integrate available knowledge into the selector and dynamically prune less important and redundant samples during the learning. We develop a TSAD model selection system with KDSelector as the internal, to demonstrate how users improve the accuracy and training speed of their selectors by using KDSelector as a plug-and-play module. Our demonstration video is hosted at https://youtu.be/2uqupDWvTF0.

cs.LG↗

arXiv2025

A high-resolution microresonator-frequency-comb spectrometer

Spectral analysis is one of the most powerful technologies for studying and understanding matter. As the devices for spectral analysis, spectrometers are widely used in material detection, isotope analysis, trace gas detection, and the study of atomic and molecular hyperfine structures. While high resolution, wide bandwidth and fast speed are essential factors, they are always trade-offs for conventional spectrometers. Here, we present a soliton-microcomb-based spectrometer that overcomes these challenges by integrating dissipative Kerr solitons (DKSs) with double-sideband modulation and parallelized detection. Leveraging a high-quality silicon nitride microresonator, we generate a broadband, fully stabilized soliton microcomb and employ radio-frequency-modulated double sidebands to scan the optical spectrum with the resolution constrained only by the comb-line linewidth. By projecting the comb lines onto a two-dimensional charge-coupled device (CCD) via a virtually imaged phased array (VIPA)-grating system, we enable parallel processing of all spectral components, circumventing sequential scanning delays. The resulting spectrometer achieves 200-kHz resolution across a 4-THz bandwidth with minutes-level processing time while maintaining robustness against environmental fluctuations. Being promising for miniaturization, this work bridges the gap between laboratory-grade performance and field-deployable practicality, unlocking new possibilities for spectroscopy in astronomy, metrology, and integrated photonics.

physics.optics↗

arXiv2025

Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs

We introduce Phi-4-Mini and Phi-4-Multimodal, compact yet highly capable language and multimodal models. Phi-4-Mini is a 3.8-billion-parameter language model trained on high-quality web and synthetic data, significantly outperforming recent open-source models of similar size and matching the performance of models twice its size on math and coding tasks requiring complex reasoning. This achievement is driven by a carefully curated synthetic data recipe emphasizing high-quality math and coding datasets. Compared to its predecessor, Phi-3.5-Mini, Phi-4-Mini features an expanded vocabulary size of 200K tokens to better support multilingual applications, as well as group query attention for more efficient long-sequence generation. Phi-4-Multimodal is a multimodal model that integrates text, vision, and speech/audio input modalities into a single model. Its novel modality extension approach leverages LoRA adapters and modality-specific routers to allow multiple inference modes combining various modalities without interference. For example, it now ranks first in the OpenASR leaderboard to date, although the LoRA component of the speech/audio modality has just 460 million parameters. Phi-4-Multimodal supports scenarios involving (vision + language), (vision + speech), and (speech/audio) inputs, outperforming larger vision-language and speech-language models on a wide range of tasks. Additionally, we experiment to further train Phi-4-Mini to enhance its reasoning capabilities. Despite its compact 3.8-billion-parameter size, this experimental version achieves reasoning performance on par with or surpassing significantly larger models, including DeepSeek-R1-Distill-Qwen-7B and DeepSeek-R1-Distill-Llama-8B.

cs.CL↗

arXiv2025

Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling

Efficiently modeling sequences with infinite context length has long been a challenging problem. Previous approaches have either suffered from quadratic computational complexity or limited extrapolation ability in length generalization. In this work, we present Samba, a simple hybrid architecture that layer-wise combines Mamba, a selective State Space Model (SSM), with Sliding Window Attention (SWA). Samba selectively compresses a given sequence into recurrent hidden states while still maintaining the ability to precisely recall recent memories with the attention mechanism. We scale Samba up to 3.8B parameters with 3.2T training tokens and demonstrate that it significantly outperforms state-of-the-art models across a variety of benchmarks. Pretrained on sequences of 4K length, Samba shows improved perplexity in context lengths of up to 1M in zero-shot. When finetuned on 4K-length sequences, Samba efficiently extrapolates to a 256K context length with perfect memory recall on the Passkey Retrieval task, and exhibits superior retrieval extrapolation on the challenging Phonebook task compared to full-attention models. As a linear-time sequence model, Samba achieves a 3.73x higher throughput compared to Transformers with grouped-query attention for user prompts of 128K length, and a 3.64x speedup when generating 64K tokens with unlimited streaming. Our code for training on open source data is publicly available at https://github.com/microsoft/Samba.

cs.CL↗

arXiv2025

InteRecon: Towards Reconstructing Interactivity of Personal Memorable Items in Mixed Reality

Digital capturing of memorable personal items is a key way to archive personal memories. Although current digitization methods (e.g., photos, videos, 3D scanning) can replicate the physical appearance of an item, they often cannot preserve its real-world interactivity. We present Interactive Digital Item (IDI), a concept of reconstructing both the physical appearance and, more importantly, the interactivity of an item. We first conducted a formative study to understand users' expectations of IDI, identifying key physical interactivity features, including geometry, interfaces, and embedded content of items. Informed by these findings, we developed InteRecon, an AR prototype enabling personal reconstruction functions for IDI creation. An exploratory study was conducted to assess the feasibility of using InteRecon and explore the potential of IDI to enrich personal memory archives. Results show that InteRecon is feasible for IDI creation, and the concept of IDI brings new opportunities for augmenting personal memory archives.

cs.HC↗

arXiv2025

Bifurcations in Bosonic Stars: chains and rings from spherical solutions

We study the bifurcation phenomena between spherical and axisymmetric bosonic stars. By numerically solving for the zero-modes of spherical bosonic stars under specific axially symmetric perturbations, we discover that excited state spherical bosonic stars bifurcate into two types of axisymmetric bosonic stars under $\ell=2$ perturbations, with matter distributions resembling chains and rings, respectively. Meanwhile, $\ell=4$ axisymmetric perturbations lead spherical scalar bosonic stars to bifurcate into a new type of axisymmetric bosonic stars, exhibiting a mixed chain-like and ring-like matter distribution, which we refer to as gyroscope-like. Additionally, for the first time, we have constructed chains of scalar bosonic stars with 7 constituents and their corresponding ring-like scalar bosonic stars. Our results provide an explanation for the bifurcations in bosonic stars from the perspective of perturbations, and by analyzing physical quantities such as quadrupoles and energy densities we systematically discuss the impact of axisymmetric perturbations on spherical bosonic stars.

gr-qc↗