arXiv subjects

Zihao Li

Publications and source records attributed to Zihao Li.

At least 91 records · Page 5Linked to original sources

arXiv2026

GPU-Fuzz: Finding Memory Errors in Deep Learning Frameworks

GPU memory errors are a critical threat to deep learning (DL) frameworks, leading to crashes or even security issues. We introduce GPU-Fuzz, a fuzzer locating these issues efficiently by modeling operator parameters as formal constraints. GPU-Fuzz utilizes a constraint solver to generate test cases that systematically probe error-prone boundary conditions in GPU kernels. Applied to PyTorch, TensorFlow, and PaddlePaddle, we uncovered 13 unknown bugs, demonstrating the effectiveness of GPU-Fuzz in finding memory errors.

cs.CR↗

arXiv2026

SAGE-LLM: Towards Safe and Generalizable LLM Controller with Fuzzy-CBF Verification and Graph-Structured Knowledge Retrieval for UAV Decision

In UAV dynamic decision, complex and variable hazardous factors pose severe challenges to the generalization capability of algorithms. Despite offering semantic understanding and scene generalization, Large Language Models (LLM) lack domain-specific UAV control knowledge and formal safety assurances, restricting their direct applicability. To bridge this gap, this paper proposes a train-free two-layer decision architecture based on LLMs, integrating high-level safety planning with low-level precise control. The framework introduces three key contributions: 1) A fuzzy Control Barrier Function verification mechanism for semantically-augmented actions, providing provable safety certification for LLM outputs. 2) A star-hierarchical graph-based retrieval-augmented generation system, enabling efficient, elastic, and interpretable scene adaptation. 3) Systematic experimental validation in pursuit-evasion scenarios with unknown obstacles and emergent threats, demonstrating that our SAGE-LLM maintains performance while significantly enhancing safety and generalization without online training. The proposed framework demonstrates strong extensibility, suggesting its potential for generalization to broader embodied intelligence systems and safety-critical control domains.

cs.RO↗

arXiv2026

The Dark Side of ChatGPT: Legal and Ethical Challenges from Stochastic Parrots and Hallucination

With the launch of ChatGPT, Large Language Models (LLMs) are shaking up our whole society, rapidly altering the way we think, create and live. For instance, the GPT integration in Bing has altered our approach to online searching. While nascent LLMs have many advantages, new legal and ethical risks are also emerging, stemming in particular from stochastic parrots and hallucination. The EU is the first and foremost jurisdiction that has focused on the regulation of AI models. However, the risks posed by the new LLMs are likely to be underestimated by the emerging EU regulatory paradigm. Therefore, this correspondence warns that the European AI regulatory paradigm must evolve further to mitigate such risks.

cs.CY↗

arXiv2026

Hypergraphs as Weighted Directed Self-Looped Graphs: Spectral Properties, Clustering, Cheeger Inequality

Hypergraphs naturally arise when studying group relations and have been widely used in the field of machine learning. To the best of our knowledge, the recently proposed edge-dependent vertex weights (EDVW) modeling is one of the most generalized modeling methods of hypergraphs, i.e., most existing hypergraph conceptual modeling methods can be generalized as EDVW hypergraphs without information loss. However, the relevant algorithmic developments on EDVW hypergraphs remain nascent: compared to the spectral theories for graphs, its formulations are incomplete, the spectral clustering algorithms are not well-developed, and the hypergraph Cheeger Inequality is not well-defined. To this end, deriving a unified random walk-based formulation, we propose our definitions of hypergraph Rayleigh Quotient, NCut, boundary/cut, volume, and conductance, which are consistent with the corresponding definitions on graphs. Then, we prove that the normalized hypergraph Laplacian is associated with the NCut value, which inspires our proposed HyperClus-G algorithm for spectral clustering on EDVW hypergraphs. Finally, we prove that HyperClus-G can always find an approximately linearly optimal partitioning in terms of both NCut and conductance. Additionally, we provide extensive experiments to validate our theoretical findings from an empirical perspective. Code of HyperClus-G is available at https://github.com/iDEA-iSAIL-Lab-UIUC/HyperClus-G.

cs.SI↗

arXiv2026

MTFM: A Scalable and Alignment-free Foundation Model for Industrial Recommendation in Meituan

Industrial recommendation systems typically involve multiple scenarios, yet existing cross-domain (CDR) and multi-scenario (MSR) methods often require prohibitive resources and strict input alignment, limiting their extensibility. We propose MTFM (Meituan Foundation Model for Recommendation), a transformer-based framework that addresses these challenges. Instead of pre-aligning inputs, MTFM transforms cross-domain data into heterogeneous tokens, capturing multi-scenario knowledge in an alignment-free manner. To enhance efficiency, we first introduce a multi-scenario user-level sample aggregation that significantly enhances training throughput by reducing the total number of instances. We further integrate Grouped-Query Attention and a customized Hybrid Target Attention to minimize memory usage and computational complexity. Furthermore, we implement various system-level optimizations, such as kernel fusion and the elimination of CPU-GPU blocking, to further enhance both training and inference throughput. Offline and online experiments validate the effectiveness of MTFM, demonstrating that significant performance gains are achieved by scaling both model capacity and multi-scenario training data.

cs.IR↗

arXiv2026

UEREBot: Learning Safe Quadrupedal Locomotion under Unstructured Environments and High-Speed Dynamic Obstacles

Quadruped robots are increasingly deployed in unstructured environments. Safe locomotion in these settings requires long-horizon goal progress, passability over uneven terrain and static constraints, and collision avoidance against high-speed dynamic obstacles. A single system cannot fully satisfy all three objectives simultaneously: planning-based decisions can be too slow, while purely reactive decisions can sacrifice goal progress and passability. To resolve this conflict, we propose UEREBot (Unstructured-Environment Reflexive Evasion Robot), a hierarchical framework that separates slow planning from instantaneous reflexive evasion and coordinates them during execution. UEREBot formulates the task as a constrained optimal control problem blueprint. It adopts a spatial--temporal planner that provides reference guidance toward the goal and threat signals. It then uses a threat-aware handoff to fuse navigation and reflex actions into a nominal command, and a control barrier function shield as a final execution safeguard. We evaluate UEREBot in Isaac Lab simulation and deploy it on a Unitree Go2 quadruped equipped with onboard perception. Across diverse environments with complex static structure and high-speed dynamic obstacles, UEREBot achieves higher avoidance success and more stable locomotion while maintaining goal progress than representative baselines, demonstrating improved safety--progress trade-offs.

cs.RO↗

arXiv2026

Dynamic Mechanism Design without Monetary Transfers: A Queueing Theory Approach

We study the design of optimal allocation mechanisms in an environment where agents and goods arrive stochastically. Agents have private types that determine the principal payoff. Either agents or goods can be held in a queue at a flow cost until allocation. The principal cannot use monetary transfers, but can verify agents types at a cost. We characterize the optimal mechanism at the steady state of the system. It is a dynamic threshold mechanism in which the principal sets type thresholds for agent admission and goods allocation. These thresholds depend on the current state of the mechanism. The model applies to public programs such as public housing and grant allocation, and to allocation problems within organizations such as capital budgeting.

econ.TH↗

arXiv2026

Unveiling Traffic Wave of Linear Adaptive Cruise Control: A Second-order Macroscopic Traffic Flow Model

Traffic waves, the spatiotemporal propagation of congestion, are a key feature of traffic flow. As Adaptive Cruise Control (ACC) systems gain widespread adoption and show promise for improving both efficiency and safety, understanding how these waves evolve under ACC becomes increasingly important. Yet most existing analyses rely on steady-state metrics (e.g., equilibrium spacing) and neglect the ACC control-law parameters, such as feedback gains, that fundamentally shape higher-order traffic dynamics. To overcome this limitation, we embed the ACC control law directly into the momentum equation while retaining mass conservation law. The result is a higher-order macroscopic model whose dynamics are governed by a second-order partial differential equation equivalent to the linear ACC feedback law. Analyzing the flux Jacobian confirms that the system is strictly hyperbolic, thereby preserving anisotropy and ensuring physical consistency. The derivation also shows that traffic wave evolution depends on both the initial state and the ACC control parameters. We analyze wave-propagation characteristics, linear degeneracy, admissible discontinuities, and their connection to ACC string stability, with the corresponding derivations. Numerical experiments confirm that the second-order model yields markedly lower vehicle-pair speed deviations along wave paths than a first-order model subject to the same non-steady disturbances, underscoring both the necessity of a second-order treatment and the soundness of the proposed framework.

math.AP↗

arXiv2026

SpatialV2A: Visual-Guided High-fidelity Spatial Audio Generation

While video-to-audio generation has achieved remarkable progress in semantic and temporal alignment, most existing studies focus solely on these aspects, paying limited attention to the spatial perception and immersive quality of the synthesized audio. This limitation stems largely from current models' reliance on mono audio datasets, which lack the binaural spatial information needed to learn visual-to-spatial audio mappings. To address this gap, we introduce two key contributions: we construct BinauralVGGSound, the first large-scale video-binaural audio dataset designed to support spatially aware video-to-audio generation; and we propose a end-to-end spatial audio generation framework guided by visual cues, which explicitly models spatial features. Our framework incorporates a visual-guided audio spatialization module that ensures the generated audio exhibits realistic spatial attributes and layered spatial depth while maintaining semantic and temporal alignment. Experiments show that our approach substantially outperforms state-of-the-art models in spatial fidelity and delivers a more immersive auditory experience, without sacrificing temporal or semantic consistency. The demo page can be accessed at https://github.com/renlinjie868-web/SpatialV2A.

cs.CV↗

arXiv2026

Cyclic Modulation Control of Multi-Conflict Connected Automated Traffic

Multi-conflict traffic is ubiquitous. Connected Automated Vehicles (CAVs) offer unprecedented opportunities to enhance safety, reduce emissions, and increase throughput through precise coordination and automation. However, existing CAV strategies remain confined to specialized scenarios, such as highway on-ramp merging or single-lane roundabouts, and traditional traffic signals sacrifice efficiency for safety via rigid phasing and all-red intervals. In this paper, we present Cyclic Modulation Control of Multi-Conflict Connected Automated Traffic (CMAT), a unified, geometry-agnostic framework that embeds each conflict point into a repeating sequence of "micro-phases". Vehicles dynamically form platoons with demand-responsive sizes and negotiate time slots for occupying conflict points, enabling collision-free traversal and high intersection utilization. CMAT aims to minimize delay, guarantee safety, and accommodate arbitrary merging, diverging, and crossing patterns without manual retuning. We formalize CMAT as a mixed-integer linear programming model constructed on a directed graph abstracted from the physical intersection layout. The performance of CMAT is evaluated across a suite of multi-conflict tests, including simple two-way crossings, four-leg intersections, complex connected intersections. The results demonstrate substantial reductions in delay and significant throughput improvements compared with state-of-the-art CAV coordination methods and traditional signal timing strategies.

math.OC↗

arXiv2026

Test-Time Scaling of Reasoning Models for Machine Translation

Test-time scaling (TTS) has enhanced the performance of Reasoning Models (RMs) on various tasks such as math and coding, yet its efficacy in machine translation (MT) remains underexplored. This paper investigates whether increased inference-time computation improves translation quality. We evaluate 12 RMs across a diverse suite of MT benchmarks spanning multiple domains, examining three scenarios: direct translation, forced-reasoning extrapolation, and post-editing. Our findings show that for general-purpose RMs, TTS provides limited and inconsistent benefits for direct translation, with performance quickly plateauing. However, the effectiveness of TTS is unlocked by domain-specific fine-tuning, which aligns a model's reasoning process with task requirements, leading to consistent improvements up to an optimal, self-determined reasoning depth. We also find that forcing a model to reason beyond its natural stopping point consistently degrades translation quality. In contrast, TTS proves highly effective in a post-editing context, reliably turning self-correction into a beneficial process. These results indicate that the value of inference-time computation in MT lies not in enhancing single-pass translation with general models, but in targeted applications like multi-step, self-correction workflows and in conjunction with task-specialized models.

cs.CL↗

arXiv2025

Nearly query-optimal classical shadow estimation of unitary channels

Classical shadow estimation (CSE) is a powerful tool for learning the properties of quantum states and quantum processes. Here we consider the CSE task for quantum unitary channels. By querying an unknown unitary channel $\mathcal{U}$ multiple times in quantum experiments, the goal is to learn a classical description from which one can accurately predict many different linear properties of the channel, i.e., the expectation values of arbitrary observables measured on the output of $\mathcal{U}$ upon arbitrary input states. Based on collective measurements on multiple systems, we propose a query efficient protocol for this task, whose query complexity has a quadratic advantage over the previous best approach for this problem, and almost saturates the information-theoretic lower bound. To further enhance practicality, we also present a variant protocol using only single-copy measurements, which still offers much better query performance than previous protocols that do not use quantum memory, and can serve as a key subroutine for learning an arbitrary unknown Hamiltonian from dynamics. In addition to linear properties of unitary channels, our protocol can also be applied to simultaneously predict many non-linear properties, such as out-of-time-ordered correlators.

quant-ph↗

arXiv2025

EMMA-500: Enhancing Massively Multilingual Adaptation of Large Language Models

In this work, we introduce EMMA-500, a large-scale multilingual language model continue-trained on texts across 546 languages designed for enhanced multilingual performance, focusing on improving language coverage for low-resource languages. To facilitate continual pre-training, we compile the MaLA corpus, a comprehensive multilingual dataset enriched with curated datasets across diverse domains. Leveraging this corpus, we conduct extensive continual pre-training of the Llama 2 7B model, resulting in EMMA-500, which demonstrates robust performance across a wide collection of benchmarks, including a comprehensive set of multilingual tasks. Our results highlight the effectiveness of continual pre-training in expanding large language models' language capacity, particularly for underrepresented languages, demonstrating significant gains in cross-lingual transfer, task generalization, and language adaptability. We release the MaLA corpus, EMMA-500 model weights, scripts, and model generations.

cs.CL↗

arXiv2025

Massively Multilingual Adaptation of Large Language Models Using Bilingual Translation Data

This paper investigates a critical design decision in the practice of massively multilingual continual pre-training -- the inclusion of parallel data. Specifically, we study the impact of bilingual translation data for massively multilingual language adaptation of the Llama3 family of models to 500 languages. To this end, we construct the MaLA bilingual translation corpus, containing data from more than 2,500 language pairs. Subsequently, we develop the EMMA-500 Llama 3 suite of four massively multilingual models -- continually pre-trained from the Llama 3 family of base models extensively on diverse data mixes up to 671B tokens -- and explore the effect of continual pre-training with or without bilingual translation data. Comprehensive evaluation across 7 tasks and 12 benchmarks demonstrates that bilingual data tends to enhance language transfer and performance, particularly for low-resource languages. We open-source the MaLA corpus, EMMA-500 Llama 3 suite artefacts, code, and model generations.

cs.CL↗

arXiv2025

High-Precision Fidelity Estimation with Common Randomized Measurements

Efficient fidelity estimation of multiqubit quantum states is crucial to many applications in quantum information processing. However, to estimate the infidelity $ε$ with multiplicative precision, conventional estimation protocols require (order) $1/ε^2$ different circuits in addition to $1/ε^2$ samples, which is quite resource-intensive for high-precision fidelity estimation. Here we introduce an efficient estimation protocol by virtue of common randomized measurements (CRM) integrated with shadow estimation based on the Clifford group, which only requires $1/ε$ circuits. Moreover, in many scenarios of practical interest, in the presence of depolarizing or Pauli noise for example, our protocol only requires a constant number of circuits, irrespective of the infidelity $ε$ and the qubit number. For large and intermediate quantum systems, quite often one circuit is already sufficient. In the course of study, we clarify the performance of CRM shadow estimation based on the Clifford group and 4-designs and highlight its advantages over standard and thrifty shadow estimation.

quant-ph↗

arXiv2025

MAMMOTH-Grism: Revisiting the Mass-Metallicity Relation in Protocluster Environments at Cosmic Noon

We present one of the first measurements of the mass-metallicity relation (MZR) in multiple massive protoclusters at cosmic noon, using Hubble Space Telescope (HST) G141 slitless spectroscopy from the MAMMOTH-Grism survey. We identify 63 protocluster member galaxies across three overdense structures at $z = 2\text{-}3$ with robust detections of [OIII], H$β$, and [OII] emission. The sample spans gas-phase metallicities of $12 + \log(\text{O/H}) = 8.2\text{-}8.6$, dust-corrected H$β$-based star formation rates (SFRs) of $10$-$250\,M_\odot\,\text{yr}^{-1}$, and stellar masses of $M_\ast \sim 10^{9.4}$-$10^{10.5}\,M_\odot$, derived via spectral energy distribution fitting using deep HST and ground-based photometry. We stack spectra in five $M_\ast$ bins to obtain average metallicities and SFRs. Relative to field galaxies at similar redshifts, protocluster members show elevated SFRs at $M_\ast < 10^{10.25}\,M_\odot$ and a systematically shallower MZR: $12 + \log(\text{O/H}) = (6.96 \pm 0.13) + (0.143 \pm 0.017) \times \log(M_{\ast}/M_{\odot})$. We detect a mass-dependent environmental offset: massive protocluster galaxies are metal-poor compared to field counterparts of similar mass, whereas lower-mass systems exhibit comparable or mildly enhanced metallicities. This trend is consistent with a scenario where cold-mode accretion dilutes the interstellar medium (ISM) across the full mass range, while efficient recycling of feedback-driven outflows preferentially enriches the ISM in low-mass galaxies. Finally, we assess the dependence of metallicity offsets on local overdensity and find no significant trend, likely reflecting the survey's bias toward protocluster cores.

astro-ph.GA↗

arXiv2025

Knowledge-Grounded Agentic Large Language Models for Multi-Hazard Understanding from Reconnaissance Reports

Post-disaster reconnaissance reports contain critical evidence for understanding multi-hazard interactions, yet their unstructured narratives make systematic knowledge transfer difficult. Large language models (LLMs) offer new potential for analyzing these reports, but often generate unreliable or hallucinated outputs when domain grounding is absent. This study introduces the Mixture-of-Retrieval Agentic RAG (MoRA-RAG), a knowledge-grounded LLM framework that transforms reconnaissance reports into a structured foundation for multi-hazard reasoning. The framework integrates a Mixture-of-Retrieval mechanism that dynamically routes queries across hazard-specific databases while using agentic chunking to preserve contextual coherence during retrieval. It also includes a verification loop that assesses evidence sufficiency, refines queries, and initiates targeted searches when information remains incomplete. We construct HazardRecQA by deriving question-answer pairs from GEER reconnaissance reports, which document 90 global events across seven major hazard types. MoRA-RAG achieves up to 94.5 percent accuracy, outperforming zero-shot LLMs by 30 percent and state-of-the-art RAG systems by 10 percent, while reducing hallucinations across diverse LLM architectures. MoRA-RAG also enables open-weight LLMs to achieve performance comparable to proprietary models. It establishes a new paradigm for transforming post-disaster documentation into actionable, trustworthy intelligence for hazard resilience.

cs.CL↗

arXiv2025

Transformer Copilot: Learning from The Mistake Log in LLM Fine-tuning

Large language models are typically adapted to downstream tasks through supervised fine-tuning on domain-specific data. While standard fine-tuning focuses on minimizing generation loss to optimize model parameters, we take a deeper step by retaining and leveraging the model's own learning signals, analogous to how human learners reflect on past mistakes to improve future performance. We first introduce the concept of Mistake Log to systematically track the model's learning behavior and recurring errors throughout fine-tuning. Treating the original transformer-based model as the Pilot, we correspondingly design a Copilot model to refine the Pilot's inference performance via logits rectification. We name the overall Pilot-Copilot framework the Transformer Copilot, which introduces (i) a novel Copilot model design, (ii) a joint training paradigm where the Copilot continuously learns from the evolving Mistake Log alongside the Pilot, and (iii) a fused inference paradigm where the Copilot rectifies the Pilot's logits for enhanced generation. We provide both theoretical and empirical analyses on our new learning framework. Experiments on 12 benchmarks spanning commonsense, arithmetic, and recommendation tasks demonstrate that Transformer Copilot consistently improves performance by up to 34.5%, while introducing marginal computational overhead to Pilot models and exhibiting strong scalability and transferability. Our code is released at https://github.com/jiaruzouu/TransformerCopilot.

cs.CL↗