arXiv subjects

Zefang Liu

Publications and source records attributed to Zefang Liu.

At least 19 recordsLinked to original sources

arXiv2026

StreamHear: Domain-Adapted Pseudo-Labeling for Semi-Supervised Streaming Speech Recognition

Streaming automatic speech recognition (ASR) underperforms on domain-shifted target audio, where labeled in-domain data is costly to prepare while unlabeled audio is abundant. We present StreamHear, a semi-supervised pipeline that adapts a pretrained streaming student by fine-tuning an offline transducer teacher on the labeled training set, generating pseudo-labels on the unlabeled portion, and fine-tuning the student on the mixture. We further introduce a prior-regularized dynamic-programming realignment step that fixes chunk-level word placement using an ASR-hypothesis anchor. Across four datasets spanning financial calls, prepared read speech, and phone-quality dialogue, StreamHear consistently outperforms supervised student fine-tuning and narrows the gap to the offline teacher.

cs.CL↗

arXiv2026

Ask, Don't Judge: Binary Questions for Interpretable LLM Evaluation and Self-Improvement

Evaluating LLM outputs remains a major bottleneck in NLP: human evaluation is expensive and slow, lexical metrics correlate poorly with human judgments on open-ended generation, and holistic LLM judges often produce opaque scores that are hard to debug. We propose BINEVAL, a framework that decomposes evaluation criteria into atomic binary questions and aggregates the resulting verdicts into interpretable, multi-dimensional scores. Given a task prompt, a meta-prompt generates fine-grained evaluation questions, and an LLM answers them independently for each output, yielding transparent question-level feedback together with calibrated overall scores. This decomposition makes evaluation easier to inspect, easier to diagnose, and directly usable for prompt improvement. Across SummEval, Topical-Chat, and QAGS, BINEVAL matches or outperforms strong baselines including UniEval and G-Eval, with especially strong results on factual consistency benchmarks such as QAGS. Beyond competitive correlation with human judgments, BINEVAL better matches human score distributions and avoids the ceiling effects common in prior LLM judges, leading to better discrimination between borderline and clearly flawed outputs. We further show that the same question-level feedback supports iterative prompt optimization, improving evaluator prompts on summarization and generation prompts on IFBench under both self-update and cross-model update settings. Overall, BINEVAL provides a task-agnostic, training-free, and interpretable evaluation framework that combines strong empirical performance with practical diagnostic and optimization value.

cs.AI↗

arXiv2026

ReHear: Iterative Pseudo-Label Refinement for Semi-Supervised Speech Recognition via Audio Large Language Models

Semi-supervised learning in automatic speech recognition (ASR) typically relies on pseudo-labeling, which often suffers from confirmation bias and error accumulation due to noisy supervision. To address this limitation, we propose ReHear, a framework for iterative pseudo-label refinement that integrates an instruction-tuned, audio-aware large language model (LLM) into the self-training loop. Unlike conventional text-based correctors, our approach conditions the LLM on both the ASR hypothesis and the source audio, allowing it to recover phonetically accurate transcripts even from severe recognition errors. These refined pseudo-labels serve as high-fidelity targets for fine-tuning the ASR model in an iterative cycle. Experimental results across diverse benchmarks demonstrate that ReHear effectively mitigates error propagation, consistently outperforming both supervised and pseudo-labeling baselines.

cs.CL↗

arXiv2025

Temporal Tokenization Strategies for Event Sequence Modeling with Large Language Models

Representing continuous time is a critical and under-explored challenge in modeling temporal event sequences with large language models (LLMs). Various strategies like byte-level representations or calendar tokens have been proposed. However, the optimal approach remains unclear, especially given the diverse statistical distributions of real-world event data, which range from smooth log-normal to discrete, spiky patterns. This paper presents a systematic empirical study of temporal tokenization for modeling event sequences with LLMs, comparing distinct encoding strategies: naive numeric strings, high-precision byte-level representations, human-semantic calendar tokens, classic uniform binning, and adaptive residual scalar quantization. We evaluate these strategies by fine-tuning LLMs on real-world datasets that exemplify these diverse distributions. Our analysis reveals that no single strategy is universally superior; instead, prediction performance depends heavily on aligning the tokenizer with the data's statistical properties, highlighting temporal tokenization as a critical yet often overlooked design dimension in LLM-based event modeling.

cs.CL↗

arXiv2025

AgentBnB: A Browser-Based Cybersecurity Tabletop Exercise with Large Language Model Support and Retrieval-Aligned Scaffolding

Traditional cybersecurity tabletop exercises (TTXs) provide valuable training but are often scripted, resource-intensive, and difficult to scale. We introduce AgentBnB, a browser-based re-imagining of the Backdoors & Breaches game that integrates large language model teammates with a Bloom-aligned, retrieval-augmented copilot (C2D2). The system expands a curated corpus into factual, conceptual, procedural, and metacognitive snippets, delivering on-demand, cognitively targeted hints. Prompt-engineered agents employ a scaffolding ladder that gradually fades as learner confidence grows. In a solo-player pilot with four graduate students, participants reported greater intention to use the agent-based version compared to the physical card deck and viewed it as more scalable, though a ceiling effect emerged on a simple knowledge quiz. Despite limitations of small sample size, single-player focus, and narrow corpus, these early findings suggest that large language model augmented TTXs can provide lightweight, repeatable practice without the logistical burden of traditional exercises. Planned extensions include multi-player modes, telemetry-driven coaching, and comparative studies with larger cohorts.

cs.CL↗

arXiv2025

AutoBnB-RAG: Enhancing Multi-Agent Incident Response with Retrieval-Augmented Generation

Incident response (IR) requires fast, coordinated, and well-informed decision-making to contain and mitigate cyber threats. While large language models (LLMs) have shown promise as autonomous agents in simulated IR settings, their reasoning is often limited by a lack of access to external knowledge. In this work, we present AutoBnB-RAG, an extension of the AutoBnB framework that incorporates retrieval-augmented generation (RAG) into multi-agent incident response simulations. Built on the Backdoors & Breaches (B&B) tabletop game environment, AutoBnB-RAG enables agents to issue retrieval queries and incorporate external evidence during collaborative investigations. We introduce two retrieval settings: one grounded in curated technical documentation (RAG-Wiki), and another using narrative-style incident reports (RAG-News). We evaluate performance across eight team structures, including newly introduced argumentative configurations designed to promote critical reasoning. To validate practical utility, we also simulate real-world cyber incidents based on public breach reports, demonstrating AutoBnB-RAG's ability to reconstruct complex multi-stage attacks. Our results show that retrieval augmentation improves decision quality and success rates across diverse organizational models. This work demonstrates the value of integrating retrieval mechanisms into LLM-based multi-agent systems for cybersecurity decision-making.

cs.CL↗

arXiv2025

A Generalized Multinodal Model for Plasma Particle and Energy Transport

We present a generalized multinodal model for simulating particle and energy transport in toroidal plasma configurations, developed to support burning plasma analysis and reactor-scale modeling. Unlike fixed-node models, this formulation allows an arbitrary number of nodes, offering increased flexibility for coupling with core-edge or core-pedestal simulations. The model derives nodal balance equations for each plasma species by volume-averaging the continuity and energy conservation equations across toroidal shell nodes. Particle and energy transport terms are expressed in terms of internodal fluxes, linked to radial gradients via linear diffusion laws for particle density and temperature, respectively. The resulting transport contributions are characterized through effective particle and energy transport times, derived explicitly in terms of nodal geometry and diffusivities. This generalized framework facilitates efficient, modular implementation of radial transport dynamics in reduced-order or integrated plasma simulations, and is compatible with data-driven approaches such as NeuralPlasmaODE for model calibration and inference from experimental data.

physics.plasm-ph↗

arXiv2025

Sensitivity Analysis of Transport and Radiation in NeuralPlasmaODE for ITER Burning Plasmas

Understanding how key physical parameters influence burning plasma behavior is critical for the reliable operation of ITER. In this work, we extend NeuralPlasmaODE, a multi-region, multi-timescale model based on neural ordinary differential equations, to perform a sensitivity analysis of transport and radiation mechanisms in ITER plasmas. Normalized sensitivities of core and edge temperatures and densities are computed with respect to transport diffusivities, electron cyclotron radiation (ECR) parameters, impurity fractions, and ion orbit loss (IOL) timescales. The analysis focuses on perturbations around a trained nominal model for the ITER inductive scenario. Results highlight the dominant influence of magnetic field strength, safety factor, and impurity content on energy confinement, while also revealing how temperature-dependent transport contributes to self-regulating behavior. These findings demonstrate the utility of NeuralPlasmaODE for predictive modeling and scenario optimization in burning plasma environments.

physics.plasm-ph↗

arXiv2025

Optimizing External Sources for Controlled Burning Plasma in Tokamaks with Neural Ordinary Differential Equations

Achieving controlled burning plasma in tokamaks requires precise regulation of external particle and energy sources to reach and maintain target core densities and temperatures. This work presents an inverse modeling approach using a multinodal plasma dynamics model based on neural ordinary differential equations (Neural ODEs). Given a desired time evolution of nodal quantities such as deuteron density or electron temperature, we compute the external source profiles, such as neutral beam injection (NBI) power, that drive the plasma toward the specified behavior. The approach is implemented within the NeuralPlasmaODE framework, which models multi-region, multi-timescale transport and incorporates physical mechanisms including radiation, auxiliary heating, and internodal energy exchange. By formulating the control task as an optimization problem, we use automatic differentiation through the Neural ODE solver to minimize the discrepancy between simulated and target trajectories. This framework transforms the forward simulation tool into a control-oriented model and provides a practical method for computing external source profiles in both current and future fusion devices.

physics.plasm-ph↗

arXiv2025

EconWebArena: Benchmarking Autonomous Agents on Economic Tasks in Realistic Web Environments

We introduce EconWebArena, a benchmark for evaluating autonomous agents on complex, multimodal economic tasks in realistic web environments. The benchmark comprises 360 curated tasks from 82 authoritative websites spanning domains such as macroeconomics, labor, finance, trade, and public policy. Each task challenges agents to navigate live websites, interpret structured and visual content, interact with real interfaces, and extract precise, time-sensitive data through multi-step workflows. We construct the benchmark by prompting multiple large language models (LLMs) to generate candidate tasks, followed by rigorous human curation to ensure clarity, feasibility, and source reliability. Unlike prior work, EconWebArena emphasizes fidelity to authoritative data sources and the need for grounded web-based economic reasoning. We evaluate a diverse set of state-of-the-art multimodal LLMs as web agents, analyze failure cases, and conduct ablation studies to assess the impact of visual grounding, plan-based reasoning, and interaction design. Our results reveal substantial performance gaps and highlight persistent challenges in grounding, navigation, and multimodal understanding, positioning EconWebArena as a rigorous testbed for economic web intelligence.

cs.CL↗

arXiv2024

Multi-Agent Collaboration in Incident Response with Large Language Models

Incident response (IR) is a critical aspect of cybersecurity, requiring rapid decision-making and coordinated efforts to address cyberattacks effectively. Leveraging large language models (LLMs) as intelligent agents offers a novel approach to enhancing collaboration and efficiency in IR scenarios. This paper explores the application of LLM-based multi-agent collaboration using the Backdoors & Breaches framework, a tabletop game designed for cybersecurity training. We simulate real-world IR dynamics through various team structures, including centralized, decentralized, and hybrid configurations. By analyzing agent interactions and performance across these setups, we provide insights into optimizing multi-agent collaboration for incident response. Our findings highlight the potential of LLMs to enhance decision-making, improve adaptability, and streamline IR processes, paving the way for more effective and coordinated responses to cyber threats.

cs.CL↗

arXiv2024

Retrieval of Temporal Event Sequences from Textual Descriptions

Retrieving temporal event sequences from textual descriptions is crucial for applications such as analyzing e-commerce behavior, monitoring social media activities, and tracking criminal incidents. To advance this task, we introduce TESRBench, a comprehensive benchmark for temporal event sequence retrieval (TESR) from textual descriptions. TESRBench includes diverse real-world datasets with synthesized and reviewed textual descriptions, providing a strong foundation for evaluating retrieval performance and addressing challenges in this domain. Building on this benchmark, we propose TPP-Embedding, a novel model for embedding and retrieving event sequences. The model leverages the TPP-LLM framework, integrating large language models (LLMs) with temporal point processes (TPPs) to encode both event texts and times. By pooling representations and applying a contrastive loss, it unifies temporal dynamics and event semantics in a shared embedding space, aligning sequence-level embeddings of event sequences and their descriptions. TPP-Embedding demonstrates superior performance over baseline models across TESRBench datasets, establishing it as a powerful solution for the temporal event sequence retrieval task.

cs.CL↗

arXiv2024

TPP-LLM: Modeling Temporal Point Processes by Efficiently Fine-Tuning Large Language Models

Temporal point processes (TPPs) are widely used to model the timing and occurrence of events in domains such as social networks, transportation systems, and e-commerce. In this paper, we introduce TPP-LLM, a novel framework that integrates large language models (LLMs) with TPPs to capture both the semantic and temporal aspects of event sequences. Unlike traditional methods that rely on categorical event type representations, TPP-LLM directly utilizes the textual descriptions of event types, enabling the model to capture rich semantic information embedded in the text. While LLMs excel at understanding event semantics, they are less adept at capturing temporal patterns. To address this, TPP-LLM incorporates temporal embeddings and employs parameter-efficient fine-tuning (PEFT) methods to effectively learn temporal dynamics without extensive retraining. This approach improves both predictive accuracy and computational efficiency. Experimental results across diverse real-world datasets demonstrate that TPP-LLM outperforms state-of-the-art baselines in sequence modeling and event prediction, highlighting the benefits of combining LLMs with TPPs.

cs.LG↗

arXiv2024

Application of Neural Ordinary Differential Equations for ITER Burning Plasma Dynamics

The dynamics of burning plasmas in tokamaks are crucial for advancing controlled thermonuclear fusion. This study applies the NeuralPlasmaODE, a multi-region multi-timescale transport model, to simulate the complex energy transfer processes in ITER deuterium-tritium (D-T) plasmas. Our model captures the interactions between energetic alpha particles, electrons, and ions, which are vital for understanding phenomena such as thermal runaway instability. We employ neural ordinary differential equations (Neural ODEs) for the numerical derivation of diffusivity parameters, enabling precise modeling of energy interactions between different plasma regions. By leveraging transfer learning, we utilize model parameters derived from DIII-D experimental data, enhancing the efficiency and accuracy of our simulations without training from scratch. Applying this model to ITER's inductive and non-inductive operational scenarios, our results demonstrate that radiation and transport processes effectively remove excess heat from the core plasma, preventing thermal runaway instability. This study underscores the potential of machine learning in advancing our understanding and control of burning plasma dynamics in fusion reactors.

physics.plasm-ph↗

arXiv2024

InvAgent: A Large Language Model based Multi-Agent System for Inventory Management in Supply Chains

Supply chain management (SCM) involves coordinating the flow of goods, information, and finances across various entities to deliver products efficiently. Effective inventory management is crucial in today's volatile and uncertain world. Previous research has demonstrated the superiority of heuristic methods and reinforcement learning applications in inventory management. However, the application of large language models (LLMs) as autonomous agents in multi-agent systems for inventory management remains underexplored. This study introduces a novel approach using LLMs to manage multi-agent inventory systems. Leveraging their zero-shot learning capabilities, our model, InvAgent, enhances resilience and improves efficiency across the supply chain network. Our contributions include utilizing LLMs for zero-shot learning to enable adaptive and informed decision-making without prior training, providing explainability and clarity through chain-of-thought, and demonstrating dynamic adaptability to varying demand scenarios while reducing costs and preventing stockouts. Extensive evaluations across different scenarios highlight the efficiency of our model in SCM.

cs.CL↗

arXiv2024

EconLogicQA: A Question-Answering Benchmark for Evaluating Large Language Models in Economic Sequential Reasoning

In this paper, we introduce EconLogicQA, a rigorous benchmark designed to assess the sequential reasoning capabilities of large language models (LLMs) within the intricate realms of economics, business, and supply chain management. Diverging from traditional benchmarks that predict subsequent events individually, EconLogicQA poses a more challenging task: it requires models to discern and sequence multiple interconnected events, capturing the complexity of economic logics. EconLogicQA comprises an array of multi-event scenarios derived from economic articles, which necessitate an insightful understanding of both temporal and logical event relationships. Through comprehensive evaluations, we exhibit that EconLogicQA effectively gauges a LLM's proficiency in navigating the sequential complexities inherent in economic contexts. We provide a detailed description of EconLogicQA dataset and shows the outcomes from evaluating the benchmark across various leading-edge LLMs, thereby offering a thorough perspective on their sequential reasoning potential in economic contexts. Our benchmark dataset is available at https://huggingface.co/datasets/yinzhu-quan/econ_logic_qa.

cs.CL↗

arXiv2024

AdaMoLE: Fine-Tuning Large Language Models with Adaptive Mixture of Low-Rank Adaptation Experts

We introduce AdaMoLE, a novel method for fine-tuning large language models (LLMs) through an Adaptive Mixture of Low-Rank Adaptation (LoRA) Experts. Moving beyond conventional methods that employ a static top-k strategy for activating experts, AdaMoLE dynamically adjusts the activation threshold using a dedicated threshold network, adaptively responding to the varying complexities of different tasks. By replacing a single LoRA in a layer with multiple LoRA experts and integrating a gating function with the threshold mechanism, AdaMoLE effectively selects and activates the most appropriate experts based on the input context. Our extensive evaluations across a variety of commonsense reasoning and natural language processing tasks show that AdaMoLE exceeds baseline performance. This enhancement highlights the advantages of AdaMoLE's adaptive selection of LoRA experts, improving model effectiveness without a corresponding increase in the expert count. The experimental validation not only confirms AdaMoLE as a robust approach for enhancing LLMs but also suggests valuable directions for future research in adaptive expert selection mechanisms, potentially broadening the scope for optimizing model performance across diverse language processing tasks.

cs.CL↗

arXiv2024

Application of Neural Ordinary Differential Equations for Tokamak Plasma Dynamics Analysis

In the quest for controlled thermonuclear fusion, tokamaks present complex challenges in understanding burning plasma dynamics. This study introduces a multi-region multi-timescale transport model, employing Neural Ordinary Differential Equations (Neural ODEs) to simulate the intricate energy transfer processes within tokamaks. Our methodology leverages Neural ODEs for the numerical derivation of diffusivity parameters from DIII-D tokamak experimental data, enabling the precise modeling of energy interactions between electrons and ions across various regions, including the core, edge, and scrape-off layer. These regions are conceptualized as distinct nodes, capturing the critical timescales of radiation and transport processes essential for efficient tokamak operation. Validation against DIII-D plasmas under various auxiliary heating conditions demonstrates the model's effectiveness, ultimately shedding light on ways to enhance tokamak performance with deep learning.

physics.plasm-ph↗