arXiv subjects

Ming Zhang

Publications and source records attributed to Ming Zhang.

At least 145 records · Page 8Linked to original sources

arXiv2025

Finite-cutoff Holographic Thermodynamics

We develop a framework for holographic thermodynamics in finite-cutoff holography, extending the anti-de Sitter/conformal field theory (AdS/CFT) correspondence to incorporate a finite radial cutoff in the bulk and a $T^2$-deformed CFT on the boundary. We formulate the first laws of thermodynamics for a Schwarzschild-AdS (SAdS) black hole with a Dirichlet cutoff on the quasilocal boundary and its dual deformed CFT, introducing the deformation parameter as a thermodynamic variable. The holographic Euler relation for the deformed CFT and its equation of state are derived, alongside the Smarr relation for the bulk. We show that the Rupert teardrop coexistence curve defines a phase space island where deformation flow alters states, with up to three deformed CFTs or cut-off SAdS sharing a same phase transition temperature, one matching the seed CFT or original SAdS. These results offer insights into gravitational thermodynamics with boundary constraints and quantum gravity in finite spacetime regions.

hep-th↗

arXiv2025

From Web Search towards Agentic Deep Research: Incentivizing Search with Reasoning Agents

Information retrieval is a cornerstone of modern knowledge acquisition, enabling billions of queries each day across diverse domains. However, traditional keyword-based search engines are increasingly inadequate for handling complex, multi-step information needs. Our position is that Large Language Models (LLMs), endowed with reasoning and agentic capabilities, are ushering in a new paradigm termed Agentic Deep Research. These systems transcend conventional information search techniques by tightly integrating autonomous reasoning, iterative retrieval, and information synthesis into a dynamic feedback loop. We trace the evolution from static web search to interactive, agent-based systems that plan, explore, and learn. We also introduce a test-time scaling law to formalize the impact of computational depth on reasoning and search. Supported by benchmark results and the rise of open-source implementations, we demonstrate that Agentic Deep Research not only significantly outperforms existing approaches, but is also poised to become the dominant paradigm for future information seeking. All the related resources, including industry products, research papers, benchmark datasets, and open-source implementations, are collected for the community in https://github.com/DavidZWZ/Awesome-Deep-Research.

cs.IR↗

arXiv2025

Probing valence electron and hydrogen dynamics using charge-pair imaging with ultrafast electron diffraction

A key challenge in ultrafast science has been to directly track the coupled motions of electrons and nuclei in real-space and real-time. This study presents a significant step towards this goal by demonstrating the feasibility of time-resolved real-space tracking of valence electron and hydrogen dynamics during the photodissociation of ammonia (NH3) using MeV ultrafast electron diffraction. It is demonstrated that the enhanced temporal resolution, in conjunction with the analysis of the charge-pair distribution function, enables the disentanglement of the correlated motion of valence electrons and hydrogens in photoexcited ammonia molecule. The methodology employed in this study, which utilizes the charge-pair distribution function from ultrafast electron scattering to retrieve intertwined electron and nucleus dynamics, may open up new opportunities in the study of quantum dynamics for a wide range of molecules.

physics.chem-ph↗

arXiv2025

Coherent tunneling of collective excitation of Bose-Einstein condensate in a double-well potential

The Josephson effect can be observed in a Bose-Einstein condensate in a double-well potential, which is attributed to the tunneling of bosons between two wells. We propose a multi-mode theory to investigate the dynamics of local excitations in a one-dimensional condensate in a double-well potential. We show that the system can be described by two independent two-mode models. The Josephson oscillation and the self-trapping of local collective excitations are predicted analytically and confirmed by numerical simulation.

cond-mat.quant-gas↗

arXiv2025

PFDial: A Structured Dialogue Instruction Fine-tuning Method Based on UML Flowcharts

Process-driven dialogue systems, which operate under strict predefined process constraints, are essential in customer service and equipment maintenance scenarios. Although Large Language Models (LLMs) have shown remarkable progress in dialogue and reasoning, they still struggle to solve these strictly constrained dialogue tasks. To address this challenge, we construct Process Flow Dialogue (PFDial) dataset, which contains 12,705 high-quality Chinese dialogue instructions derived from 440 flowcharts containing 5,055 process nodes. Based on PlantUML specification, each UML flowchart is converted into atomic dialogue units i.e., structured five-tuples. Experimental results demonstrate that a 7B model trained with merely 800 samples, and a 0.5B model trained on total data both can surpass 90% accuracy. Additionally, the 8B model can surpass GPT-4o up to 43.88% with an average of 11.00%. We further evaluate models' performance on challenging backward transitions in process flows and conduct an in-depth analysis of various dataset formats to reveal their impact on model performance in handling decision and sequential branches. The data is released in https://github.com/KongLongGeFDU/PFDial.

cs.CL↗

arXiv2025

SMI-Editor: Edit-based SMILES Language Model with Fragment-level Supervision

SMILES, a crucial textual representation of molecular structures, has garnered significant attention as a foundation for pre-trained language models (LMs). However, most existing pre-trained SMILES LMs focus solely on the single-token level supervision during pre-training, failing to fully leverage the substructural information of molecules. This limitation makes the pre-training task overly simplistic, preventing the models from capturing richer molecular semantic information. Moreover, during pre-training, these SMILES LMs only process corrupted SMILES inputs, never encountering any valid SMILES, which leads to a train-inference mismatch. To address these challenges, we propose SMI-Editor, a novel edit-based pre-trained SMILES LM. SMI-Editor disrupts substructures within a molecule at random and feeds the resulting SMILES back into the model, which then attempts to restore the original SMILES through an editing process. This approach not only introduces fragment-level training signals, but also enables the use of valid SMILES as inputs, allowing the model to learn how to reconstruct complete molecules from these incomplete structures. As a result, the model demonstrates improved scalability and an enhanced ability to capture fragment-level molecular information. Experimental results show that SMI-Editor achieves state-of-the-art performance across multiple downstream molecular tasks, and even outperforming several 3D molecular representation models.

cs.LG↗

arXiv2025

ExLM: Rethinking the Impact of [MASK] Tokens in Masked Language Models

Masked Language Models (MLMs) have achieved remarkable success in many self-supervised representation learning tasks. MLMs are trained by randomly masking portions of the input sequences with [MASK] tokens and learning to reconstruct the original content based on the remaining context. This paper explores the impact of [MASK] tokens on MLMs. Analytical studies show that masking tokens can introduce the corrupted semantics problem, wherein the corrupted context may convey multiple, ambiguous meanings. This problem is also a key factor affecting the performance of MLMs on downstream tasks. Based on these findings, we propose a novel enhanced-context MLM, ExLM. Our approach expands [MASK] tokens in the input context and models the dependencies between these expanded states. This enhancement increases context capacity and enables the model to capture richer semantic information, effectively mitigating the corrupted semantics problem during pre-training. Experimental results demonstrate that ExLM achieves significant performance improvements in both text modeling and SMILES modeling tasks. Further analysis confirms that ExLM enriches semantic representations through context enhancement, and effectively reduces the semantic multimodality commonly observed in MLMs.

cs.CL↗

arXiv2025

Safe: Enhancing Mathematical Reasoning in Large Language Models via Retrospective Step-aware Formal Verification

Chain-of-Thought (CoT) prompting has become the de facto method to elicit reasoning capabilities from large language models (LLMs). However, to mitigate hallucinations in CoT that are notoriously difficult to detect, current methods such as process reward models (PRMs) or self-consistency operate as opaque boxes and do not provide checkable evidence for their judgments, possibly limiting their effectiveness. To address this issue, we draw inspiration from the idea that "the gold standard for supporting a mathematical claim is to provide a proof". We propose a retrospective, step-aware formal verification framework $Safe$. Rather than assigning arbitrary scores, we strive to articulate mathematical claims in formal mathematical language Lean 4 at each reasoning step and provide formal proofs to identify hallucinations. We evaluate our framework $Safe$ across multiple language models and various mathematical datasets, demonstrating a significant performance improvement while offering interpretable and verifiable evidence. We also propose $FormalStep$ as a benchmark for step correctness theorem proving with $30,809$ formal statements. To the best of our knowledge, our work represents the first endeavor to utilize formal mathematical language Lean 4 for verifying natural language content generated by LLMs, aligning with the reason why formal mathematical languages were created in the first place: to provide a robust foundation for hallucination-prone human-written proofs.

cs.CL↗

arXiv2025

FinMME: Benchmark Dataset for Financial Multi-Modal Reasoning Evaluation

Multimodal Large Language Models (MLLMs) have experienced rapid development in recent years. However, in the financial domain, there is a notable lack of effective and specialized multimodal evaluation datasets. To advance the development of MLLMs in the finance domain, we introduce FinMME, encompassing more than 11,000 high-quality financial research samples across 18 financial domains and 6 asset classes, featuring 10 major chart types and 21 subtypes. We ensure data quality through 20 annotators and carefully designed validation mechanisms. Additionally, we develop FinScore, an evaluation system incorporating hallucination penalties and multi-dimensional capability assessment to provide an unbiased evaluation. Extensive experimental results demonstrate that even state-of-the-art models like GPT-4o exhibit unsatisfactory performance on FinMME, highlighting its challenging nature. The benchmark exhibits high robustness with prediction variations under different prompts remaining below 1%, demonstrating superior reliability compared to existing datasets. Our dataset and evaluation protocol are available at https://huggingface.co/datasets/luojunyu/FinMME and https://github.com/luo-junyu/FinMME.

cs.CL↗

arXiv2025

Symmetry resolved entanglement entropy after an inhomogeneous quench

We investigate the non-equilibrium time evolution of symmetry-resolved entanglement entropy (SREE) following an inhomogeneous quench in a critical one-dimensional free fermion system. Using conformal field theory, we derive an exact expression for the SREE and analyze its behavior. We find that, at leading order in the long-time limit, the SREE grows logarithmically as $\log t$. While the equipartition of entanglement holds at leading order, we identify subleading corrections that break it. Our numerical simulations corroborate the analytical predictions with excellent agreement.

hep-th↗

arXiv2025

MARCO: Meta-Reflection with Cross-Referencing for Code Reasoning

The ability to reason is one of the most fundamental capabilities of large language models (LLMs), enabling a wide range of downstream tasks through sophisticated problem-solving. A critical aspect of this is code reasoning, which involves logical reasoning with formal languages (i.e., programming code). In this paper, we enhance this capability of LLMs by exploring the following question: how can an LLM agent become progressively smarter in code reasoning with each solution it proposes, thereby achieving substantial cumulative improvement? Most existing research takes a static perspective, focusing on isolated problem-solving using frozen LLMs. In contrast, we adopt a cognitive-evolving perspective and propose a novel framework named Meta-Reflection with Cross-Referencing (MARCO) that enables the LLM to evolve dynamically during inference through self-improvement. From the perspective of human cognitive development, we leverage both knowledge accumulation and lesson sharing. In particular, to accumulate knowledge during problem-solving, we propose meta-reflection that reflects on the reasoning paths of the current problem to obtain knowledge and experience for future consideration. Moreover, to effectively utilize the lessons from other agents, we propose cross-referencing that incorporates the solution and feedback from other agents into the current problem-solving process. We conduct experiments across various datasets in code reasoning, and the results demonstrate the effectiveness of MARCO.

cs.CL↗

arXiv2025

Autonomous Circular Drift Control for 4WD-4WS Vehicles Without Precomputed Drifting Equilibrium

Under extreme conditions, autonomous drifting enables vehicles to follow predefined paths at large slip angles, significantly enhancing the control system's capability to handle hazardous scenarios. Four-wheel-drive and four-wheel-steering (4WD-4WS) vehicles, which have been extensively studied, offer superior path-following precision and enhanced maneuverability under challenging driving conditions. In this paper, a hierarchical drifting controller is proposed for 4WD-4WS vehicles to track both path and velocity without relying on precomputed drifting equilibrium. The controller is structured into two layers: a trajectory tracking layer and an actuator regulation layer. The first layer generates the desired tire forces in the vehicle body frame, while the second layer converts these desired tire forces into steering angle commands and torque commands for the front and rear motors. The effectiveness and robustness of the proposed controller are validated through simulation.

eess.SY↗

arXiv2025

Cross-Domain Diffusion with Progressive Alignment for Efficient Adaptive Retrieval

Unsupervised efficient domain adaptive retrieval aims to transfer knowledge from a labeled source domain to an unlabeled target domain, while maintaining low storage cost and high retrieval efficiency. However, existing methods typically fail to address potential noise in the target domain, and directly align high-level features across domains, thus resulting in suboptimal retrieval performance. To address these challenges, we propose a novel Cross-Domain Diffusion with Progressive Alignment method (COUPLE). This approach revisits unsupervised efficient domain adaptive retrieval from a graph diffusion perspective, simulating cross-domain adaptation dynamics to achieve a stable target domain adaptation process. First, we construct a cross-domain relationship graph and leverage noise-robust graph flow diffusion to simulate the transfer dynamics from the source domain to the target domain, identifying lower noise clusters. We then leverage the graph diffusion results for discriminative hash code learning, effectively learning from the target domain while reducing the negative impact of noise. Furthermore, we employ a hierarchical Mixup operation for progressive domain alignment, which is performed along the cross-domain random walk paths. Utilizing target domain discriminative hash learning and progressive domain alignment, COUPLE enables effective domain adaptive hash learning. Extensive experiments demonstrate COUPLE's effectiveness on competitive benchmarks.

cs.LG↗

arXiv2025

DELTA: Dual Consistency Delving with Topological Uncertainty for Active Graph Domain Adaptation

Graph domain adaptation has recently enabled knowledge transfer across different graphs. However, without the semantic information on target graphs, the performance on target graphs is still far from satisfactory. To address the issue, we study the problem of active graph domain adaptation, which selects a small quantitative of informative nodes on the target graph for extra annotation. This problem is highly challenging due to the complicated topological relationships and the distribution discrepancy across graphs. In this paper, we propose a novel approach named Dual Consistency Delving with Topological Uncertainty (DELTA) for active graph domain adaptation. Our DELTA consists of an edge-oriented graph subnetwork and a path-oriented graph subnetwork, which can explore topological semantics from complementary perspectives. In particular, our edge-oriented graph subnetwork utilizes the message passing mechanism to learn neighborhood information, while our path-oriented graph subnetwork explores high-order relationships from sub-structures. To jointly learn from two subnetworks, we roughly select informative candidate nodes with the consideration of consistency across two subnetworks. Then, we aggregate local semantics from its K-hop subgraph based on node degrees for topological uncertainty estimation. To overcome potential distribution shifts, we compare target nodes and their corresponding source nodes for discrepancy scores as an additional component for fine selection. Extensive experiments on benchmark datasets demonstrate that DELTA outperforms various state-of-the-art approaches. The code implementation of DELTA is available at https://github.com/goose315/DELTA.

cs.LG↗

arXiv2025

Operational and Exploration Requirements and Research Capabilities for SEP Environment Monitoring and Forecasting

Mitigating risks posed by solar energetic particles (SEPs) to operations and exploration in space and Earth's atmosphere motivates the development of advanced, synergistic approaches for monitoring, modeling, and analyzing space weather conditions. The consequences of SEPs and their interactions with the near-Earth space environment are numerous, including elevated radiation levels at aviation altitudes during major events, satellite damage, and health risks to astronauts, resulting in economic impacts and potential hazards for space exploration. This contribution will present a high-level overview of the operational requirements and research capabilities for SEP event environment monitoring and forecasting that were highlighted during a workshop at Georgia State University, held on October 16-19, 2024. Specifically, it summarizes the presented activities concerning the following: (1) Identifying needs for SEP event forecasting and nowcasting, including practical forecast timeframes; (2) Reviewing availability and coverage of the current observational data and identifying tangible data resources for research, operations and the R2O2R loop; (3) Mapping existing forecast capabilities and identifying meaningful modeling advances for research and operations.

astro-ph.SR↗

arXiv2025

An Efficient deep learning model to Predict Stock Price Movement Based on Limit Order Book

In high-frequency trading (HFT), leveraging limit order books (LOB) to model stock price movements is crucial for achieving profitable outcomes. However, this task is challenging due to the high-dimensional and volatile nature of the original data. Even recent deep learning models often struggle to capture price movement patterns effectively, particularly without well-designed features. We observed that raw LOB data exhibits inherent symmetry between the ask and bid sides, and the bid-ask differences demonstrate greater stability and lower complexity compared to the original data. Building on this insight, we propose a novel approach in which leverages the Siamese architecture to enhance the performance of existing deep learning models. The core idea involves processing the ask and bid sides separately using the same module with shared parameters. We applied our Siamese-based methods to several widely used strong baselines and validated their effectiveness using data from 14 military industry stocks in the Chinese A-share market. Furthermore, we integrated multi-head attention (MHA) mechanisms with the Long Short-Term Memory (LSTM) module to investigate its role in modeling stock price movements. Our experiments used raw data and widely used Order Flow Imbalance (OFI) features as input with some strong baseline models. The results show that our method improves the performance of strong baselines in over 75$% of cases, excluding the Multi-Layer Perception (MLP) baseline, which performed poorly and is not considered practical. Furthermore, we found that Multi-Head Attention can enhance model performance, particularly over shorter forecasting horizons.

q-fin.TR↗

arXiv2025

Multifaceted Evaluation of Audio-Visual Capability for MLLMs: Effectiveness, Efficiency, Generalizability and Robustness

Multi-modal large language models (MLLMs) have recently achieved great success in processing and understanding information from diverse modalities (e.g., text, audio, and visual signals). Despite their growing popularity, there remains a lack of comprehensive evaluation measuring the audio-visual capabilities of these models, especially in diverse scenarios (e.g., distribution shifts and adversarial attacks). In this paper, we present a multifaceted evaluation of the audio-visual capability of MLLMs, focusing on four key dimensions: effectiveness, efficiency, generalizability, and robustness. Through extensive experiments, we find that MLLMs exhibit strong zero-shot and few-shot generalization abilities, enabling them to achieve great performance with limited data. However, their success relies heavily on the vision modality, which impairs performance when visual input is corrupted or missing. Additionally, while MLLMs are susceptible to adversarial samples, they demonstrate greater robustness compared to traditional models. The experimental results and our findings provide insights into the audio-visual capabilities of MLLMs, highlighting areas for improvement and offering guidance for future research.

cs.MM↗

arXiv2025

Large Language Model Agent: A Survey on Methodology, Applications and Challenges

The era of intelligent agents is upon us, driven by revolutionary advancements in large language models. Large Language Model (LLM) agents, with goal-driven behaviors and dynamic adaptation capabilities, potentially represent a critical pathway toward artificial general intelligence. This survey systematically deconstructs LLM agent systems through a methodology-centered taxonomy, linking architectural foundations, collaboration mechanisms, and evolutionary pathways. We unify fragmented research threads by revealing fundamental connections between agent design principles and their emergent behaviors in complex environments. Our work provides a unified architectural perspective, examining how agents are constructed, how they collaborate, and how they evolve over time, while also addressing evaluation methodologies, tool applications, practical challenges, and diverse application domains. By surveying the latest developments in this rapidly evolving field, we offer researchers a structured taxonomy for understanding LLM agents and identify promising directions for future research. The collection is available at https://github.com/luo-junyu/Awesome-Agent-Papers.

cs.CL↗