arXiv subjects

Jiawei Shen

Publications and source records attributed to Jiawei Shen.

At least 19 recordsLinked to original sources

arXiv2026

SDesc3D: Towards Layout-Aware 3D Indoor Scene Generation from Short Descriptions

3D indoor scene generation conditioned on short textual descriptions provides a promising avenue for interactive 3D environment construction without the need for labor-intensive layout specification. Despite recent progress in text-conditioned 3D scene generation, existing works suffer from poor physical plausibility and insufficient detail richness in such semantic condensation cases, largely due to their reliance on explicit semantic cues about compositional objects and their spatial relationships. This limitation highlights the need for enhanced 3D reasoning capabilities, particularly in terms of prior integration and spatial anchoring. Motivated by this, we propose SDesc3D, a short-text conditioned 3D indoor scene generation framework, that leverages multi-view structural priors and regional functionality implications to enable 3D layout reasoning under sparse textual guidance. Specifically, we introduce a Multi-view scene prior augmentation that enriches underspecified textual inputs with aggregated multi-view structural knowledge, shifting from inaccessible semantic relation cues to multi-view relational prior aggregation. Building on this, we design a Functionality-aware layout grounding, employing regional functionality grounding for implicit spatial anchors and conducting hierarchical layout reasoning to enhance scene organization and semantic plausibility. Furthermore, an Iterative reflection-rectification scheme is employed for progressive structural plausibility refinement via self-rectification. Extensive experiments show that our method outperforms existing approaches on short-text conditioned 3D indoor scene generation. Code will be publicly available.

cs.CV↗

arXiv2026

DynaGen: Unifying Temporal Knowledge Graph Reasoning with Dynamic Subgraphs and Generative Regularization

Temporal Knowledge Graph Reasoning (TKGR) aims to complete missing factual elements along the timeline. Depending on the temporal position of the query, the task is categorized into interpolation and extrapolation. Existing interpolation methods typically embed temporal information into individual facts to complete missing historical knowledge, while extrapolation techniques often leverage sequence models over graph snapshots to identify recurring patterns for future event prediction. These methods face two critical challenges: limited contextual modeling in interpolation and cognitive generalization bias in extrapolation. To address these, we propose a unified method for TKGR, dubbed DynaGen. For interpolation, DynaGen dynamically constructs entity-centric subgraphs and processes them with a synergistic dual-branch GNN encoder to capture evolving structural context. For extrapolation, it applies a conditional diffusion process, which forces the model to learn underlying evolutionary principles rather than just superficial patterns, enhancing its ability to predict unseen future events. Extensive experiments on six benchmark datasets show DynaGen achieves state-of-the-art performance. On average, compared to the second-best models, DynaGen improves the Mean Reciprocal Rank (MRR) score by 2.61 points for interpolation and 1.45 points for extrapolation.

cs.LG↗

arXiv2026

ACR: Adaptive Context Refactoring via Context Refactoring Operators for Multi-Turn Dialogue

Large Language Models (LLMs) have shown remarkable performance in multi-turn dialogue. However, in multi-turn dialogue, models still struggle to stay aligned with what has been established earlier, follow dependencies across many turns, and avoid drifting into incorrect facts as the interaction grows longer. Existing approaches primarily focus on extending the context window, introducing external memory, or applying context compression, yet these methods still face limitations such as \textbf{contextual inertia} and \textbf{state drift}. To address these challenges, we propose the \textbf{A}daptive \textbf{C}ontext \textbf{R}efactoring \textbf{(ACR)} Framework, which dynamically monitors and reshapes the interaction history to mitigate contextual inertia and state drift actively. ACR is built on a library of context refactoring operators and a teacher-guided self-evolving training paradigm that learns when to intervene and how to refactor, thereby decoupling context management from the reasoning process. Extensive experiments on multi-turn dialogue demonstrate that our method significantly outperforms existing baselines while reducing token consumption.

cs.CL↗

arXiv2025

Dispersive decay for the Inter-critical nonlinear Schrödinger equation in $\mathbb{R}^3$

This paper investigates the Cauchy problem for the nonlinear Schrödinger equation (NLS) in the mass-supercritical and energy-subcritical regime within three spatial dimensions. For initial data in the critical homogeneous Sobolev space $\dot{H}^{s_c}(\mathbb{R}^3)$ (where $s_c = \frac{5}{6}$), we get a uniform decay estimate for the long-time dynamics of solutions, which extends the previous results.

math.AP↗

arXiv2025

EduEval: A Hierarchical Cognitive Benchmark for Evaluating Large Language Models in Chinese Education

Large language models (LLMs) demonstrate significant potential for educational applications. However, their unscrutinized deployment poses risks to educational standards, underscoring the need for rigorous evaluation. We introduce EduEval, a comprehensive hierarchical benchmark for evaluating LLMs in Chinese K-12 education. This benchmark makes three key contributions: (1) Cognitive Framework: We propose the EduAbility Taxonomy, which unifies Bloom's Taxonomy and Webb's Depth of Knowledge to organize tasks across six cognitive dimensions including Memorization, Understanding, Application, Reasoning, Creativity, and Ethics. (2) Authenticity: Our benchmark integrates real exam questions, classroom conversation, student essays, and expert-designed prompts to reflect genuine educational challenges; (3) Scale: EduEval comprises 24 distinct task types with over 11,000 questions spanning primary to high school levels. We evaluate 14 leading LLMs under both zero-shot and few-shot settings, revealing that while models perform well on factual tasks, they struggle with classroom dialogue classification and exhibit inconsistent results in creative content generation. Interestingly, several open source models outperform proprietary systems on complex educational reasoning. Few-shot prompting shows varying effectiveness across cognitive dimensions, suggesting that different educational objectives require tailored approaches. These findings provide targeted benchmarking metrics for developing LLMs specifically optimized for diverse Chinese educational tasks.

cs.CL↗

arXiv2025

A validated fluid-structure interaction simulation model for vortex-induced vibration of a flexible pipe in steady flow

We propose a validated fluid-structure interaction simulation framework based on strip methods for the vortex-induced vibration of a flexible pipe. The numerical results are compared with the experimental data from three previous steady flow conditions: uniform, linearly sheared, and bidirectionally sheared flow. The Reynolds number ranges from $10^4$ to $10^5$. The flow field is simulated via the RANS model, which is based on the open-source software OpenFOAM. The solid field is modeled based on Euler-Bernoulli beam theory, and fluid-structure coupling is implemented via a weak coupling algorithm developed in MATLAB. The vortex-induced vibration response is assessed in terms of amplitude and frequency, along with the differences in strain. Additionally, wavelet analysis and traveling wave phenomena are investigated. The numerical simulation codes and experimental data in this manuscript are openly available, providing a foundation for more complex vortex-induced vibration simulations in the future.

physics.flu-dyn↗

arXiv2025

Automatic Failure Attribution and Critical Step Prediction Method for Multi-Agent Systems Based on Causal Inference

Multi-agent systems (MAS) are critical for automating complex tasks, yet their practical deployment is severely hampered by the challenge of failure attribution. Current diagnostic tools, which rely on statistical correlations, are fundamentally inadequate; on challenging benchmarks like Who\&When, state-of-the-art methods achieve less than 15\% accuracy in locating the root-cause step of a failure. To address this critical gap, we introduce the first failure attribution framework for MAS grounded in multi-granularity causal inference. Our approach makes two key technical contributions: (1) a performance causal inversion principle, which correctly models performance dependencies by reversing the data flow in execution logs, combined with Shapley values to accurately assign agent-level blame; (2) a novel causal discovery algorithm, CDC-MAS, that robustly identifies critical failure steps by tackling the non-stationary nature of MAS interaction data. The framework's attribution results directly fuel an automated optimization loop, generating targeted suggestions whose efficacy is validated via counterfactual simulations. Evaluations on the Who\&When and TRAIL benchmarks demonstrate a significant leap in performance. Our method achieves up to 36.2\% step-level accuracy. Crucially, the generated optimizations boost overall task success rates by an average of 22.4\%. This work provides a principled and effective solution for debugging complex agent interactions, paving the way for more reliable and interpretable multi-agent systems.

cs.AI↗

arXiv2025

Consistent and Invariant Generalization Learning for Short-video Misinformation Detection

Short-video misinformation detection has attracted wide attention in the multi-modal domain, aiming to accurately identify the misinformation in the video format accompanied by the corresponding audio. Despite significant advancements, current models in this field, trained on particular domains (source domains), often exhibit unsatisfactory performance on unseen domains (target domains) due to domain gaps. To effectively realize such domain generalization on the short-video misinformation detection task, we propose deep insights into the characteristics of different domains: (1) The detection on various domains may mainly rely on different modalities (i.e., mainly focusing on videos or audios). To enhance domain generalization, it is crucial to achieve optimal model performance on all modalities simultaneously. (2) For some domains focusing on cross-modal joint fraud, a comprehensive analysis relying on cross-modal fusion is necessary. However, domain biases located in each modality (especially in each frame of videos) will be accumulated in this fusion process, which may seriously damage the final identification of misinformation. To address these issues, we propose a new DOmain generalization model via ConsisTency and invariance learning for shORt-video misinformation detection (named DOCTOR), which contains two characteristic modules: (1) We involve the cross-modal feature interpolation to map multiple modalities into a shared space and the interpolation distillation to synchronize multi-modal learning; (2) We design the diffusion model to add noise to retain core features of multi modal and enhance domain invariant features through cross-modal guided denoising. Extensive experiments demonstrate the effectiveness of our proposed DOCTOR model. Our code is public available at https://github.com/ghh1125/DOCTOR.

cs.CV↗

arXiv2025

Understanding LLM-Centric Challenges for Deep Learning Frameworks: An Empirical Analysis

Large language models (LLMs) have driven significant progress across a wide range of real-world applications. Realizing such models requires substantial system-level support. Deep learning (DL) frameworks provide this foundation by enabling efficient model construction, distributed execution, and optimized deployment. The large parameter scale and extended execution cycles impose exacting demands on deep learning frameworks, particularly in terms of scalability, stability, and efficiency. Therefore, poor usability, limited functionality, and subtle bugs in DL frameworks may hinder development efficiency and cause severe failures or resource waste. However, a fundamental question has not been thoroughly investigated in previous studies, i.e., what challenges do DL frameworks face in supporting LLMs? To answer this question, we analyze issue reports from three major DL frameworks (i.e., MindSpore, PyTorch, and TensorFlow) and eight associated LLM toolkits such as Megatron. Based on a manual review of these reports, we construct a taxonomy that captures LLM-centric framework bugs, user requirements, and user questions. We then refine and enrich this taxonomy through interviews with 11 LLM users and eight DL framework developers. Based on the constructed taxonomy and findings summarized from interviews, our study further reveals key technical challenges and mismatches between LLM user needs and developer priorities.

cs.SE↗

arXiv2025

Seed-Prover: Deep and Broad Reasoning for Automated Theorem Proving

LLMs have demonstrated strong mathematical reasoning abilities by leveraging reinforcement learning with long chain-of-thought, yet they continue to struggle with theorem proving due to the lack of clear supervision signals when solely using natural language. Dedicated domain-specific languages like Lean provide clear supervision via formal verification of proofs, enabling effective training through reinforcement learning. In this work, we propose \textbf{Seed-Prover}, a lemma-style whole-proof reasoning model. Seed-Prover can iteratively refine its proof based on Lean feedback, proved lemmas, and self-summarization. To solve IMO-level contest problems, we design three test-time inference strategies that enable both deep and broad reasoning. Seed-Prover proves $78.1\%$ of formalized past IMO problems, saturates MiniF2F, and achieves over 50\% on PutnamBench, outperforming the previous state-of-the-art by a large margin. To address the lack of geometry support in Lean, we introduce a geometry reasoning engine \textbf{Seed-Geometry}, which outperforms previous formal geometry engines. We use these two systems to participate in IMO 2025 and fully prove 5 out of 6 problems. This work represents a significant advancement in automated mathematical reasoning, demonstrating the effectiveness of formal verification with long chain-of-thought reasoning.

cs.AI↗

arXiv2025

CriticLean: Critic-Guided Reinforcement Learning for Mathematical Formalization

Translating natural language mathematical statements into formal, executable code is a fundamental challenge in automated theorem proving. While prior work has focused on generation and compilation success, little attention has been paid to the critic phase-the evaluation of whether generated formalizations truly capture the semantic intent of the original problem. In this paper, we introduce CriticLean, a novel critic-guided reinforcement learning framework that elevates the role of the critic from a passive validator to an active learning component. Specifically, first, we propose the CriticLeanGPT, trained via supervised fine-tuning and reinforcement learning, to rigorously assess the semantic fidelity of Lean 4 formalizations. Then, we introduce CriticLeanBench, a benchmark designed to measure models' ability to distinguish semantically correct from incorrect formalizations, and demonstrate that our trained CriticLeanGPT models can significantly outperform strong open- and closed-source baselines. Building on the CriticLean framework, we construct FineLeanCorpus, a dataset comprising over 285K problems that exhibits rich domain diversity, broad difficulty coverage, and high correctness based on human evaluation. Overall, our findings highlight that optimizing the critic phase is essential for producing reliable formalizations, and we hope our CriticLean will provide valuable insights for future advances in formal mathematical reasoning.

cs.CL↗

arXiv2025

$L^p$ boundness of Oscillatory singular integral with Calder\'{o}n Type Commutators

In the paper, we study a kind of Oscillatory singular integral operator with Calder\'{o}n Type Commutators $T_{P,K,A} $ defined by \[T_{P,K,A} f(x)=\text { p.v.} \int_{\mathbb{R}^{n}} f(y) \frac{K(x-y)}{|x-y|}(A(x)-A(y)-\nabla A(y))(x-y) e^{i P(x-y)} d y, \] where $P(t)$ is a real polynomial on $\mathbb{R},$ and $K$ is a function on $\mathbb{R}^{n},$ satisfies the vanishing moment and $CZ(\delta)$ conditions. Under these conditions, we show that $T_{P,K,A}$ is bounded on $L^p(\mathbb{R}^{n})$ with uniform boundedness, which improve and extend the previous result.

math.CA↗

arXiv2025

Flow-induced vibration of twin-pipe model with varying mass and damping: A study using virtual physical framework

Flow-induced vibration (FIV) commonly occurs in rigidly coupled twin-pipe structures. However, the limited understanding of their FIV responses and hydrodynamic features presents a major challenge to the development of reliable engineering designs. To bridge this gap, the present study systematically investigates the FIV characteristics of a rigidly coupled twin-pipe model with elastic support using a virtual physical framework (VPF), which enables flexible control of structural parameters during physical testing. A distinctive feature of twin-pipe structures is the presence of in-line hydrodynamic interactions and torsional moments arising from the rigid coupling. The in-line interaction is primarily compressive and becomes more pronounced as the mass ratio increases. The torsional moment coefficient exhibits a rise-fall trend with increasing reduced velocity $U_R$ and stabilizes around 0.46 at low mass ratios. In addition, an "amplitude drop" phenomenon is observed at $U_R=6$, attributed to energy dissipation from the downstream pipe. The mass ratio significantly affects FIV amplitude, frequency, and hydrodynamic coefficients. As the mass ratio decreases, the synchronization region broadens and the hydrodynamic coefficients become more stable. At mass ratio of 1.0, a "resonance forever" behavior is observed. Damping primarily suppresses FIV amplitude, with minimal impact on dominant frequency and hydrodynamic coefficients. These findings provide valuable insights into twin-pipe FIV mechanisms and support a scientific basis for future structural design optimization.

physics.flu-dyn↗

arXiv2025

Are LLMs Correctly Integrated into Software Systems?

Large language models (LLMs) provide effective solutions in various application scenarios, with the support of retrieval-augmented generation (RAG). However, developers face challenges in integrating LLM and RAG into software systems, due to lacking interface specifications, various requirements from software context, and complicated system management. In this paper, we have conducted a comprehensive study of 100 open-source applications that incorporate LLMs with RAG support, and identified 18 defect patterns. Our study reveals that 77% of these applications contain more than three types of integration defects that degrade software functionality, efficiency, and security. Guided by our study, we propose systematic guidelines for resolving these defects in software life cycle. We also construct an open-source defect library Hydrangea.

cs.SE↗

arXiv2025

Understanding, Detecting, and Repairing Real-World In-Context-Learning-Based Text-to-SQL Errors

Large language models (LLMs) have been adopted for text-to-SQL tasks, utilizing their in-context learning (ICL) capability to translate natural language questions into SQL queries. However, such a technique faces correctness problems. In this paper, we conduct the first comprehensive study of text-to-SQL errors of ICL-based techniques. Our study covers four representative ICL-based techniques, five basic repairing methods, two benchmarks, and two LLM settings. We find that text-to-SQL errors are widespread and summarize 27 error types of 7 categories. We also find that existing repairing attempts have limited correctness improvement while having high computational overhead and many mis-repairs. Based on these findings, we propose MapleDoctor, a novel text-to-SQL error detection and repairing framework. The evaluation demonstrates that MapleDoctor outperforms existing solutions by repairing 13.8% more queries with a negligible number of mis-repairs and reducing 67.4% repair latency. The artifact is publicly available at GitHub.

cs.CL↗

arXiv2022

A CMOS-based Characterisation Platform for Emerging RRAM Technologies

Mass characterisation of emerging memory devices is an essential step in modelling their behaviour for integration within a standard design flow for existing integrated circuit designers. This work develops a novel characterisation platform for emerging resistive devices with a capacity of up to 1 million devices on-chip. Split into four independent sub-arrays, it contains on-chip column-parallel DACs for fast voltage programming of the DUT. On-chip readout circuits with ADCs are also available for fast read operations covering 5-decades of input current (20nA to 2mA). This allows a device's resistance range to be between 1k$Ω$ and 10M$Ω$ with a minimum voltage range of $\pm$1.5V on the device.

cs.ET↗

arXiv2022

A Wide Dynamic Range Read-out System For Resistive Switching Technology

The memristor, because of its controllability over a wide dynamic range of resistance, has emerged as a promising device for data storage and analog computation. A major challenge is the accurate measurement of memristance over a wide dynamic range. In this paper, a novel read-out circuit with feedback adjustment is proposed to measure and digitise input current in the range between 20nA and 2mA. The magnitude of the input currents is estimated by a 5-stage logarithmic current-to-voltage amplifier which scales a linear analog-to-digital converter. This way the least significant bit tracks the absolute input magnitude. This circuit is applicable to reading single memristor conductance, and is also preferable in analog computing where read-out accuracy is particularly critical. The circuits have been realized in Bipolar-CMOS-DMOS (BCD) Gen2 technology.

cs.ET↗

arXiv2022

A High-Voltage Characterisation Platform For Emerging Resistive Switching Technologies

Emerging memristor-based array architectures have been effectively employed in non-volatile memories and neuromorphic computing systems due to their density, scalability and capability of storing information. Nonetheless, to demonstrate a practical on-chip memristor-based system, it is essential to have the ability to apply large programming voltage ranges during the characterisation procedures for various memristor technologies. This work presents a 16x16 high voltage memristor characterisation array employing high voltage CMOS circuitry. The proposed system has a maximum programming range of $\pm22V$ to allow on-chip electroforming and I-V sweep. In addition, a Kelvin voltage sensing system is implemented to improve the readout accuracy for low memristance measurements. This work addresses the limitation of conventional CMOS-memristor platforms which can only operate at low voltages, thus limiting the characterisation range and integration options of memristor technologies.

cs.ET↗