arXiv subjects

Siyu Wu

Publications and source records attributed to Siyu Wu.

33 records · Page 2Linked to original sources

arXiv2025

Arrow: Adaptive Scheduling Mechanisms for Disaggregated LLM Inference Architecture

Existing large language model (LLM) serving systems typically employ Prefill-Decode disaggregated architecture to prevent computational interference between the prefill and decode phases. However, in real-world LLM serving scenarios, significant fluctuations in request input/output lengths lead to imbalanced computational loads between prefill and decode nodes under traditional static node allocation strategies, consequently preventing efficient utilization of computing resources to improve the system's goodput. To address this challenge, we design and implement Arrow, an adaptive scheduler that leverages stateless instances and latency characteristics of prefill and decode tasks to achieve efficient adaptive request and instance scheduling. Arrow dynamically adjusts the number of instances handling prefill and decode tasks based on real-time cluster performance metrics, substantially enhancing the system's capability to handle traffic spikes and load variations. Our evaluation under diverse real-world workloads shows that Arrow achieves up to $2.55 \times$ higher request serving rates compared to state-of-the-art Prefill-Decode disaggregated serving systems.

cs.DC↗

arXiv2025

Engineering Nonporous Polymer Hybrids with Suppressed Heat Conduction and Enhanced Flame Retardancy via Molecular and Filler Design

This study presents a new strategy for achieving ultralow thermal conductivity in nonporous polymer/organic filler hybrids by suppressing heat capacity through tailored atomic vibrations to enhance thermal insulation. Unlike conventional polymer/inorganic filler hybrids, these hybrids exhibit interfacial thermal resistance one to three orders of magnitude lower. Combined experiments and simulations uncover thermal transport mechanisms. These hybrids demonstrate enhanced flame retardancy. Please see the abstract in the attached PDF.

physics.app-ph↗

arXiv2025

The Unanticipated Asymmetry Between Perceptual Optimization and Assessment

Perceptual optimization is primarily driven by the fidelity objective, which enforces both semantic consistency and overall visual realism, while the adversarial objective provides complementary refinement by enhancing perceptual sharpness and fine-grained detail. Despite their central role, the correlation between their effectiveness as optimization objectives and their capability as image quality assessment (IQA) metrics remains underexplored. In this work, we conduct a systematic analysis and reveal an unanticipated asymmetry between perceptual optimization and assessment: fidelity metrics that excel in IQA are not necessarily effective for perceptual optimization, with this misalignment emerging more distinctly under adversarial training. In addition, while discriminators effectively suppress artifacts during optimization, their learned representations offer only limited benefits when reused as backbone initializations for IQA models. Beyond this asymmetry, our findings further demonstrate that discriminator design plays a decisive role in shaping optimization, with patch-level and convolutional architectures providing more faithful detail reconstruction than vanilla or Transformer-based alternatives. These insights advance the understanding of loss function design and its connection to IQA transferability, paving the way for more principled approaches to perceptual optimization.

cs.CV↗

arXiv2025

AudioCodecBench: A Comprehensive Benchmark for Audio Codec Evaluation

Multimodal Large Language Models (MLLMs) have been widely applied in speech and music. This tendency has led to a focus on audio tokenization for Large Models (LMs). Unlike semantic-only text tokens, audio tokens must both capture global semantic content and preserve fine-grained acoustic details. Moreover, they provide a discrete method for speech and music that can be effectively integrated into MLLMs. However, existing research is unsuitable in the definitions of semantic tokens and acoustic tokens. In addition, the evaluation of different codecs typically concentrates on specific domains or tasks, such as reconstruction or Automatic Speech Recognition (ASR) task, which prevents fair and comprehensive comparisons. To address these problems, this paper provides suitable definitions for semantic and acoustic tokens and introduces a systematic evaluation framework. This framework allows for a comprehensive assessment of codecs' capabilities which evaluate across four dimensions: audio reconstruction metric, codebook index (ID) stability, decoder-only transformer perplexity, and performance on downstream probe tasks. Our results show the correctness of the provided suitable definitions and the correlation among reconstruction metrics, codebook ID stability, downstream probe tasks and perplexity.

cs.SD↗

arXiv2025

Past-Future Scheduler for LLM Serving under SLA Guarantees

The exploration and application of Large Language Models (LLMs) is thriving. To reduce deployment costs, continuous batching has become an essential feature in current service frameworks. The effectiveness of continuous batching relies on an accurate estimate of the memory requirements of requests. However, due to the diversity in request output lengths, existing frameworks tend to adopt aggressive or conservative schedulers, which often result in significant overestimation or underestimation of memory consumption. Consequently, they suffer from harmful request evictions or prolonged queuing times, failing to achieve satisfactory throughput under strict Service Level Agreement (SLA) guarantees (a.k.a. goodput), across various LLM application scenarios with differing input-output length distributions. To address this issue, we propose a novel Past-Future scheduler that precisely estimates the peak memory resources required by the running batch via considering the historical distribution of request output lengths and calculating memory occupancy at each future time point. It adapts to applications with all types of input-output length distributions, balancing the trade-off between request queuing and harmful evictions, thereby consistently achieving better goodput. Furthermore, to validate the effectiveness of the proposed scheduler, we developed a high-performance LLM serving framework, LightLLM, that implements the Past-Future scheduler. Compared to existing aggressive or conservative schedulers, LightLLM demonstrates superior goodput, achieving up to 2-3$\times$ higher goodput than other schedulers under heavy loads. LightLLM is open source to boost the research in such direction (https://github.com/ModelTC/lightllm).

cs.DC↗

arXiv2025

Pre$^3$: Enabling Deterministic Pushdown Automata for Faster Structured LLM Generation

Extensive LLM applications demand efficient structured generations, particularly for LR(1) grammars, to produce outputs in specified formats (e.g., JSON). Existing methods primarily parse LR(1) grammars into a pushdown automaton (PDA), leading to runtime execution overhead for context-dependent token processing, especially inefficient under large inference batches. To address these issues, we propose Pre$^3$ that exploits deterministic pushdown automata (DPDA) to optimize the constrained LLM decoding efficiency. First, by precomputing prefix-conditioned edges during the preprocessing, Pre$^3$ enables ahead-of-time edge analysis and thus makes parallel transition processing possible. Second, by leveraging the prefix-conditioned edges, Pre$^3$ introduces a novel approach that transforms LR(1) transition graphs into DPDA, eliminating the need for runtime path exploration and achieving edge transitions with minimal overhead. Pre$^3$ can be seamlessly integrated into standard LLM inference frameworks, reducing time per output token (TPOT) by up to 40% and increasing throughput by up to 36% in our experiments. Our code is available at https://github.com/ModelTC/lightllm.

cs.CL↗

arXiv2025

CIKT: A Collaborative and Iterative Knowledge Tracing Framework with Large Language Models

Knowledge Tracing (KT) aims to model a student's learning state over time and predict their future performance. However, traditional KT methods often face challenges in explainability, scalability, and effective modeling of complex knowledge dependencies. While Large Language Models (LLMs) present new avenues for KT, their direct application often struggles with generating structured, explainable student representations and lacks mechanisms for continuous, task-specific refinement. To address these gaps, we propose Collaborative Iterative Knowledge Tracing (CIKT), a framework that harnesses LLMs to enhance both prediction accuracy and explainability. CIKT employs a dual-component architecture: an Analyst generates dynamic, explainable user profiles from student historical responses, and a Predictor utilizes these profiles to forecast future performance. The core of CIKT is a synergistic optimization loop. In this loop, the Analyst is iteratively refined based on the predictive accuracy of the Predictor, which conditions on the generated profiles, and the Predictor is subsequently retrained using these enhanced profiles. Evaluated on multiple educational datasets, CIKT demonstrates significant improvements in prediction accuracy, offers enhanced explainability through its dynamically updated user profiles, and exhibits improved scalability. Our work presents a robust and explainable solution for advancing knowledge tracing systems, effectively bridging the gap between predictive performance and model transparency.

cs.AI↗

arXiv2025

LLM-based MOFs Synthesis Condition Extraction using Few-Shot Demonstrations

The extraction of Metal-Organic Frameworks (MOFs) synthesis route from literature has been crucial for the logical MOFs design with desirable functionality. The recent advent of large language models (LLMs) provides disruptively new solution to this long-standing problem. While the latest researches mostly stick to primitive zero-shot LLMs lacking specialized material knowledge, we introduce in this work the few-shot LLM in-context learning paradigm. First, a human-AI interactive data curation approach is proposed to secure high-quality demonstrations. Second, an information retrieval algorithm is applied to pick and quantify few-shot demonstrations for each extraction. Over three datasets randomly sampled from nearly 90,000 well-defined MOFs, we conduct triple evaluations to validate our method. The synthesis extraction, structure inference, and material design performance of the proposed few-shot LLMs all significantly outplay zero-shot LLM and baseline methods. The lab-synthesized material guided by LLM surpasses 91.1% high-quality MOFs of the same class reported in the literature, on the key physical property of specific surface area.

cs.CL↗

arXiv2024

Pion-production cross sections in neutrino reactions for studying generalized parton distributions of the nucleon

Gravitational form factors of hadrons provide information on mass and pressure distributions in the hadrons, and they have been investigated by generalized parton distributions (GPDs). The GPDs also contain information on internal compositions of hadron spins. Understanding on the origin of hadron masses, pressures, and spins is an important topic not only in hadron physics but also as fundamental physics, and it should be clarified by the GPD studies. The spacelike and timelike GPDs have been measured at charged-lepton accelerator facilities, for example, by the virtual Compton scattering and the two-photon process, respectively. In this work, we show the $π^\pm$ and $π^0$ production cross sections in neutrino (antineutrino) reactions $ν\,(\barν) + N \to \ell + π+ N'$ for measuring the GPDs of the nucleon by using the theoretical formalism of Pire, Szymanowski, and Wagner. The $π^\pm$-production cross sections are useful for determining gluon GPDs, whereas the $π^0$ production probes quark GPDs. In particular, we show the roles of the pion- and rho-pole terms, the contribution from each GPD term ($H$, $E$, $\tilde H$, $\tilde E$), the effects of the gluon GPDs $H_g$ and $E_g$, and the effects of the reaction energy on the neutrino cross sections. The $π^0$-production cross section is sensitive to the quark GPDs including the pion- and rho-pole GPD terms in the Efremov-Radyushkin-Brodsky-Lepage region. These cross sections could be measured at Fermilab in future. The neutrino GPD studies will play a complementary role to the projects of charged-lepton and hadron reactions for determining the accurate GPDs.

hep-ph↗

arXiv2024

VEC-Sim: A Simulation Platform for Evaluating Service Caching and Computation Offloading Policies in Vehicular Edge Networks

Computer simulation platforms offer an alternative solution by emulating complex systems in a controlled manner. However, existing Edge Computing (EC) simulators, as well as general-purpose vehicular network simulators, are not tailored for VEC and lack dedicated support for modeling the distinct access pattern, entity mobility trajectory and other unique characteristics of VEC networks. To fill this gap, this paper proposes VEC-Sim, a versatile simulation platform for in-depth evaluation and analysis of various service caching and computation offloading policies in VEC networks. VEC-Sim incorporates realistic mechanisms to replicate real-world access patterns, including service feature vector, vehicle mobility modeling, evolving service popularity, new service upload and user preference shifts, etc. Moreover, its modular architecture and extensive Application Programming Interfaces (APIs) allow seamless integration of customized scheduling policies and user-defined metrics. A comprehensive evaluation of VEC-Sim's capabilities is undertaken in comparison to real-world ground truths. Results prove it to be accurate in reproducing classical scheduling algorithms and extremely effective in conducting case studies.

cs.NI↗

arXiv2024

Cognitive LLMs: Towards Integrating Cognitive Architectures and Large Language Models for Manufacturing Decision-making

Resolving the dichotomy between the human-like yet constrained reasoning processes of Cognitive Architectures and the broad but often noisy inference behavior of Large Language Models (LLMs) remains a challenging but exciting pursuit, for enabling reliable machine reasoning capabilities in production systems. Because Cognitive Architectures are famously developed for the purpose of modeling the internal mechanisms of human cognitive decision-making at a computational level, new investigations consider the goal of informing LLMs with the knowledge necessary for replicating such processes, e.g., guided perception, memory, goal-setting, and action. Previous approaches that use LLMs for grounded decision-making struggle with complex reasoning tasks that require slower, deliberate cognition over fast and intuitive inference -- reporting issues related to the lack of sufficient grounding, as in hallucination. To resolve these challenges, we introduce LLM-ACTR, a novel neuro-symbolic architecture that provides human-aligned and versatile decision-making by integrating the ACT-R Cognitive Architecture with LLMs. Our framework extracts and embeds knowledge of ACT-R's internal decision-making process as latent neural representations, injects this information into trainable LLM adapter layers, and fine-tunes the LLMs for downstream prediction. Our experiments on novel Design for Manufacturing tasks show both improved task performance as well as improved grounded decision-making capability of our approach, compared to LLM-only baselines that leverage chain-of-thought reasoning strategies.

cs.AI↗

arXiv2023

BARET : Balanced Attention based Real image Editing driven by Target-text Inversion

Image editing approaches with diffusion models have been rapidly developed, yet their applicability are subject to requirements such as specific editing types (e.g., foreground or background object editing, style transfer), multiple conditions (e.g., mask, sketch, caption), and time consuming fine-tuning of diffusion models. For alleviating these limitations and realizing efficient real image editing, we propose a novel editing technique that only requires an input image and target text for various editing types including non-rigid edits without fine-tuning diffusion model. Our method contains three novelties:(I) Target-text Inversion Schedule (TTIS) is designed to fine-tune the input target text embedding to achieve fast image reconstruction without image caption and acceleration of convergence.(II) Progressive Transition Scheme applies progressive linear interpolation between target text embedding and its fine-tuned version to generate transition embedding for maintaining non-rigid editing capability.(III) Balanced Attention Module (BAM) balances the tradeoff between textual description and image semantics.By the means of combining self-attention map from reconstruction process and cross-attention map from transition process, the guidance of target text embeddings in diffusion process is optimized.In order to demonstrate editing capability, effectiveness and efficiency of the proposed BARET, we have conducted extensive qualitative and quantitative experiments. Moreover, results derived from user study and ablation study further prove the superiority over other methods.

cs.CV↗

arXiv2023

Possibility of generating the $^3_{Λ_c}$H in the quark-delocalization color-screening model

We probe the existence of the $^3_{Λ_c}$H where the $NΛ_c$ potentials are derived from the quark-delocalization color-screening model (QDCSM). The $NΛ_c$ system is studied and the $NΛ_c$ scattering length so as the effective range are obtained in the QDCSM. We construct effective Gaussian-type $NΛ_c$ potentials which reproduce the $NΛ_c$ scattering data given by the QDCSM. By solving the $NNΛ_c$ three body Schrödinger equation with the Gaussian expansion method, we calculate the energies of the $^3_{Λ_c}$H with isospin $I=0$, $J^P=1/2^+$ and $I=0$, $J^P=3/2^+$ under different color screening parameter $μ$. The $J^P=1/2^+$ and $J^P=3/2^+$ states are both bound when the color screening parameter $μ$ is set to 1.0 or 1.2, where the $J^P=1/2^+$ state is bound by $0.08\sim0.85$ MeV and the $J^P=3/2^+$ state is bound by $0.15\sim1.31$ MeV with respect to the deuteron-$Λ_c$ threshold.

nucl-th↗

arXiv2023

Structural Hawkes Processes for Learning Causal Structure from Discrete-Time Event Sequences

Learning causal structure among event types from discrete-time event sequences is a particularly important but challenging task. Existing methods, such as the multivariate Hawkes processes based methods, mostly boil down to learning the so-called Granger causality which assumes that the cause event happens strictly prior to its effect event. Such an assumption is often untenable beyond applications, especially when dealing with discrete-time event sequences in low-resolution; and typical discrete Hawkes processes mainly suffer from identifiability issues raised by the instantaneous effect, i.e., the causal relationship that occurred simultaneously due to the low-resolution data will not be captured by Granger causality. In this work, we propose Structure Hawkes Processes (SHPs) that leverage the instantaneous effect for learning the causal structure among events type in discrete-time event sequence. The proposed method is featured with the minorization-maximization of the likelihood function and a sparse optimization scheme. Theoretical results show that the instantaneous effect is a blessing rather than a curse, and the causal structure is identifiable under the existence of the instantaneous effect. Experiments on synthetic and real-world data verify the effectiveness of the proposed method.

cs.LG↗

arXiv2022

THP: Topological Hawkes Processes for Learning Causal Structure on Event Sequences

Learning causal structure among event types on multi-type event sequences is an important but challenging task. Existing methods, such as the Multivariate Hawkes processes, mostly assumed that each sequence is independent and identically distributed. However, in many real-world applications, it is commonplace to encounter a topological network behind the event sequences such that an event is excited or inhibited not only by its history but also by its topological neighbors. Consequently, the failure in describing the topological dependency among the event sequences leads to the error detection of the causal structure. By considering the Hawkes processes from the view of temporal convolution, we propose a Topological Hawkes process (THP) to draw a connection between the graph convolution in the topology domain and the temporal convolution in time domains. We further propose a causal structure learning method on THP in a likelihood framework. The proposed method is featured with the graph convolution-based likelihood function of THP and a sparse optimization scheme with an Expectation-Maximization of the likelihood function. Theoretical analysis and experiments on both synthetic and real-world data demonstrate the effectiveness of the proposed method

cs.LG↗