arXiv subjects

Yang Liu

Publications and source records attributed to Yang Liu.

At least 649 records · Page 36Linked to original sources

arXiv2026

AudioTrust: Benchmarking the Multifaceted Trustworthiness of Audio Large Language Models

The rapid development and widespread adoption of Audio Large Language Models (ALLMs) demand rigorous evaluation of their trustworthiness. However, existing evaluation frameworks are primarily designed for text and fail to capture vulnerabilities introduced by the acoustic properties of audio. We find that significant trustworthiness risks in ALLMs arise from non-semantic acoustic cues, such as timbre, accent, and background noise, which can be exploited to manipulate model behavior. To address this gap, we propose AudioTrust, the first large-scale and systematic framework for evaluating ALLM trustworthiness under audio-specific risks. AudioTrust covers six key dimensions: fairness, hallucination, safety, privacy, robustness, and authenticition. It includes 26 sub-tasks and a curated dataset of more than 4,420 audio samples collected from real-world scenarios, including daily conversations, emergency calls, and voice assistant interactions, and is specifically designed to probe trustworthiness across multiple dimensions. Our comprehensive evaluation spans 18 experimental settings and uses human-validated automated pipelines to enable objective and scalable assessment of model outputs. Experimental results on 14 state-of-the-art open-source and closed-source ALLMs reveal important limitations and failure boundaries under diverse high-risk audio scenarios, providing critical insights for the secure and trustworthy deployment of future audio models. Our platform and benchmark are publicly available at https://github.com/JusperLee/AudioTrust.

cs.SD↗

arXiv2026

Weighted Generalized Risk Measure and Risk Quadrangle: Characterization, Optimization and Application

Various financial market scenarios may cause heterogeneous risk assessments among analysts, which motivates the usage of the Generalized Risk Measure in Fadina et al. (2024, Finance and Stochastics). Effectively synthesizing these diverse assessments avoids over-relying on a single, potentially flawed or conservative forecast and promotes more robust decision-making. Motivated by this, we establish analytical characterizations of the Weighted Generalized Risk Measure (WGRM) under both discrete and continuous settings. Building upon the WGRM, we incorporate the Fundamental Risk Quadrangle (FRQ) in Rockafellar and Uryasev (2013, Surveys in Operations Research and Management Science) into the Weighted Risk Quadrangle (WRQ) and show that the intrinsic relationships among risk, deviation, regret, error, and statistics in FRQ are preserved under weighted aggregation across scenarios. Moreover, we demonstrate that certain complex risk optimization problems under the WGRM can be reformulated as tractable linear programs through the WRQ structure, thus ensuring computational feasibility. Finally, the WGRM and WRQ framework is applied to empirical analyses using constituents of the NASDAQ 100 and S&P 500 indices across recession and expansion regimes, which validates that WGRM-based portfolios exhibit superior risk-adjusted performance and enhanced downside resilience and effectively mitigate losses arising from erroneous single-scenario judgments.

q-fin.RM↗

arXiv2026

BrainMem: Brain-Inspired Evolving Memory for Embodied Agent Task Planning

Embodied task planning requires agents to execute long-horizon, goal-directed actions in complex 3D environments, where success depends on both immediate perception and accumulated experience across tasks. However, most existing LLM-based planners are stateless and reactive, operating without persistent memory and therefore repeating errors and struggling with spatial or temporal dependencies. We propose BrainMem(Brain-Inspired Evolving Memory), a training-free hierarchical memory system that equips embodied agents with working, episodic, and semantic memory inspired by human cognition. BrainMem continuously transforms interaction histories into structured knowledge graphs and distilled symbolic guidelines, enabling planners to retrieve, reason over, and adapt behaviors from past experience without any model fine-tuning or additional training. This plug-and-play design integrates seamlessly with arbitrary multi-modal LLMs and greatly reduces reliance on task-specific prompt engineering. Extensive experiments on four representative benchmarks, including EB-ALFRED, EB-Navigation, EB-Manipulation, and EB-Habitat, demonstrate that BrainMem significantly enhances task success rates across diverse models and difficulty subsets, with the largest gains observed on long-horizon and spatially complex tasks. These results highlight evolving memory as a promising and scalable mechanism for generalizable embodied intelligence.

cs.RO↗

arXiv2026

Token Cleaning: Fine-Grained Data Selection for LLM Supervised Fine-Tuning

Recent studies show that in supervised fine-tuning (SFT) of large language models (LLMs), data quality matters more than quantity. While most data cleaning methods concentrate on filtering entire samples, the quality of individual tokens within a sample can vary significantly. After pre-training, even in high-quality samples, patterns or phrases that are not task-related can be redundant, uninformative, or even harmful. Continuing to fine-tune on these patterns may offer limited benefit and even degrade downstream task performance. In this paper, we investigate token quality from a noisy-label perspective and propose a generic token cleaning pipeline for SFT tasks. Our method filters out uninformative tokens while preserving those carrying key task-specific information. Specifically, we first evaluate token quality by examining the influence of model updates on each token, then apply a threshold-based separation. The token influence can be measured in a single pass with a fixed reference model or iteratively with self-evolving reference models. The benefits and limitations of both methods are analyzed theoretically by error upper bounds. Extensive experiments show that our framework consistently improves downstream performance. Code is available at https://github.com/UCSC-REAL/TokenCleaning.

cs.CL↗

arXiv2026

Dimensionality tuning of heavy-fermion states in ultrathin CeSi2 films

Dimensionality tuning is an important method to modify the electronic states of quantum materials. However, the mechanism of such tuning in heavy fermion systems and its connection with transport properties remain largely unexplored. Here by combining molecular beam epitaxy (MBE), in-situ angle-resolved photoemission spectroscopy (ARPES) and transport measurements, we study the electronic states of the heavy-fermion compound CeSi2 as a function of film thickness. In three dimensional thick films, our measurements reveal a dispersive Kondo peak at the Fermi level (EF) and satellite peaks originating from crystal electric field (CEF) excitations, characteristic of heavy fermion systems. For two-dimensional ultrathin films, the CEF satellites are largely suppressed while the ground-state Kondo peak at EF remains strong, although it develops at lower temperatures. Simultaneously, the maximum temperature Tmax of the magnetic resistivity, \r{ho}m(T), changes from ~100 K in thick films to ~35 K in ultrathin films. This can be attributed to the dimensionality driven reduction of CEF excitations during the Kondo process, in good agreement with spectroscopic results. Our work provides direct insight to understand the quantum confinement effects on strongly correlated 4f-electron systems and opens up new opportunities to explore emergent phenomena in two-dimensional heavy-fermion materials.

cond-mat.str-el↗

arXiv2026

Droplet impact on a superhydrophobic surface under shear airflow: Lattice Boltzmann simulations and scaling analyses

Droplet impact in airflow environments is ubiquitous in nature and industry, making the understanding of this multiphase behavior crucial for technologies such as anti-icing and spray cooling. In this study, the dynamics of droplet impact on a superhydrophobic surface under shear airflow are numerically investigated using the pseudopotential multiphase lattice Boltzmann method. This three-dimensional model employs a non-orthogonal multiple-relaxation-time scheme to enhance numerical stability and a contact angle hysteresis window to effectively capture dynamic wetting. Specifically, the kinetic energy supplied by the airflow enhances streamwise spreading and significantly expands the final contact footprint due to continuous horizontal sliding. To describe the nonlinear dependence of these contact-line characteristics on the impact Weber number (We) and the airflow Reynolds number (Re), a set of composite scaling laws is developed based on a modified Weber number (We*) that incorporates the airflow contribution. Moreover, the aerodynamic effect leads to a higher velocity restitution coefficient and a deflected take-off angle. Based on an energy partition analysis at detachment, a refined power law is derived to scale the vertical restitution coefficient under shear airflow, while the streamwise restitution coefficient is formulated via the sliding velocity approximation. Integrating these two directional components enables accurate quantitative predictions of the total restitution coefficient and the take-off angle governed by the interplay of We and Re. Overall, this study clarifies the underlying mechanisms of droplet-airflow-surface interactions, providing practical insights for predicting droplet behaviors and guiding surface design under aerodynamic conditions.

physics.flu-dyn↗

arXiv2026

Modeling Trend Dynamics with Variational Neural ODEs for Information Popularity Prediction

Predicting the future popularity of information in online social networks is a crucial yet challenging task, due to the complex spatiotemporal dynamics underlying information diffusion. Existing methods typically use structural or sequential patterns within the observation window as direct inputs for subsequent popularity prediction. However, most approaches lack the ability to explicitly model the overall trend of popularity up to the prediction time, which leads to limited predictive capability. To address these limitations, we propose VNOIP, a novel method based on variational neural Ordinary Differential Equations (ODEs) for information popularity prediction. Specifically, VNOIP introduces bidirectional jump ODEs with attention mechanisms to capture long-range dependencies and bidirectional context within cascade sequences. Furthermore, by jointly considering both cascade patterns and overall trend temporal patterns, VNOIP explicitly models the continuous-time dynamics of popularity trend trajectories with variational neural ODEs. Additionally, a knowledge distillation loss is employed to align the evolution of prior and posterior latent variables. Extensive experiments on real-world datasets demonstrate that VNOIP is highly competitive in both prediction accuracy and efficiency compared to state-of-the-art baselines.

cs.SI↗

arXiv2026

Video-Based Reward Modeling for Computer-Use Agents

Computer-using agents (CUAs) are becoming increasingly capable; however, it remains difficult to scale evaluation of whether a trajectory truly fulfills a user instruction. In this work, we study reward modeling from execution video: a sequence of keyframes from an agent trajectory that is independent of the agent's internal reasoning or actions. Although video-execution modeling is method-agnostic, it presents key challenges, including highly redundant layouts and subtle, localized cues that determine success. We introduce Execution Video Reward 53k (ExeVR-53k), a dataset of 53k high-quality video--task--reward triplets. We further propose adversarial instruction translation to synthesize negative samples with step-level annotations. To enable learning from long, high-resolution execution videos, we design spatiotemporal token pruning, which removes homogeneous regions and persistent tokens while preserving decisive UI changes. Building on these components, we fine-tune an Execution Video Reward Model (ExeVRM) that takes only a user instruction and a video-execution sequence to predict task success. Our ExeVRM 8B achieves 84.7% accuracy and 87.7% recall on video-execution assessment, outperforming strong proprietary models such as GPT-5.2 and Gemini-3 Pro across Ubuntu, macOS, Windows, and Android, while providing more precise temporal attribution. These results show that video-execution reward modeling can serve as a scalable, model-agnostic evaluator for CUAs.

cs.CV↗

arXiv2026

Granulon: Awakening Pixel-Level Visual Encoders with Adaptive Multi-Granularity Semantics for MLLM

Recent advances in multimodal large language models largely rely on CLIP-based visual encoders, which emphasize global semantic alignment but struggle with fine-grained visual understanding. In contrast, DINOv3 provides strong pixel-level perception yet lacks coarse-grained semantic abstraction, leading to limited multi-granularity reasoning. To address this gap, we propose Granulon, a novel DINOv3-based MLLM with adaptive granularity augmentation. Granulon introduces a text-conditioned granularity Controller that dynamically adjusts the visual abstraction level according to the semantic scope of the textual input, and an Adaptive Token Aggregation module that performs granularity-guided pooling and relation-aware clustering to produce compact, semantically rich visual tokens. This design enables unified "pixel-to-fine-to-coarse" reasoning within a single forward pass. Extensive and interpretable experiments demonstrate that Granulon improves accuracy by ~30% and reduces hallucination by ~20%, outperforming all visual encoders under identical settings.

cs.CV↗

arXiv2026

Kernel Debiased Plug-in Estimation based on the Universal Least Favorable Submodel

We propose ULFS-KDPE, a kernel debiased plug-in estimator based on the universal least favorable submodel, for estimating pathwise differentiable parameters in nonparametric models. The method constructs a data-adaptive debiasing flow in a reproducing kernel Hilbert space (RKHS), producing a plug-in estimator that achieves semiparametric efficiency without requiring explicit derivation or evaluation of efficient influence functions. We place ULFS-KDPE on a rigorous functional-analytic foundation by formulating the universal least favorable update as a nonlinear ordinary differential equation on probability densities. We establish existence, uniqueness, stability, and finite-time convergence of the empirical score along the induced flow. Under standard regularity conditions, the resulting estimator is regular, asymptotically linear, and attains the semiparametric efficiency bound simultaneously for a broad class of pathwise differentiable parameters. The method admits a computationally tractable implementation based on finite-dimensional kernel representations and principled stopping criteria. In finite samples, the combination of solving a rich collection of score equations with RKHS-based smoothing and avoidance of direct influence-function evaluation leads to improved numerical stability. Simulation studies illustrate the method and support the theoretical results.

math.ST↗

arXiv2026

How Emotion Shapes the Behavior of LLMs and Agents: A Mechanistic Study

Emotion plays an important role in human cognition and performance. Motivated by this, we investigate whether analogous emotional signals can shape the behavior of large language models (LLMs) and agents. Existing emotion-aware studies mainly treat emotion as a surface-level style factor or a perception target, overlooking its mechanistic role in task processing. To address this limitation, we propose E-STEER, an interpretable emotion steering framework that enables direct representation-level intervention in LLMs and agents. It embeds emotion as a structured, controllable variable in hidden states, and with it, we examine the impact of emotion on objective reasoning, subjective generation, safety, and multi-step agent behaviors. The results reveal non-monotonic emotion-behavior relations consistent with established psychological theories, and show that specific emotions not only enhance LLM capability but also improve safety, and systematically shape multi-step agent behaviors.

cs.AI↗

arXiv2026

ACCURATE: Arbitrary-shaped Continuum Reconstruction Under Robust Adaptive Two-view Estimation

Accurate reconstruction of arbitrary-shaped long slender continuum bodies, such as guidewires, catheters and other soft continuum manipulators, is essential for accurate mechanical simulation. However, existing image-based reconstruction approaches often suffer from limited accuracy because they often underutilize camera geometry, or lack generality as they rely on rigid geometric assumptions that may fail for continuum robots with complex and highly deformable shapes. To address these limitations, we propose ACCURATE, a 3D reconstruction framework integrating an image segmentation neural network with a geometry-constrained topology traversal and dynamic programming algorithm that enforces global biplanar geometric consistency, minimizes the cumulative point-to-epipolar-line distance, and remains robust to occlusions and epipolar ambiguities cases caused by noise and discretization. Our method achieves high reconstruction accuracy on both simulated and real phantom datasets acquired using a clinical X-ray C-arm system, with mean absolute errors below 1.0 mm.

cs.RO↗

arXiv2026

Software Development Life Cycle Perspective: A Survey of Benchmarks for Code Large Language Models and Agents

Code large language models (CodeLLMs) and agents are increasingly being integrated into complex software engineering tasks spanning the entire Software Development Life Cycle (SDLC). Benchmarking is critical for rigorously evaluating these capabilities. However, despite their growing significance, there remains a lack of comprehensive reviews that examine these benchmarks from an SDLC perspective. To bridge this gap, we propose a tiered analysis framework to systematically review 178 benchmarks from 461 papers, comprehensively characterizing them from the perspective of the SDLC. Our findings reveal a notable imbalance in the coverage of current benchmarks, with approximately 61\% focused on the software implementation phase in SDLC, while requirements engineering and software design phases receive minimal attention at only 5\% and 3\%, respectively. % Additionally, anti-contamination strategies are largely absent from current benchmarks, leading to an increased risk of data leakage. Furthermore, current benchmarks lack effective anti-contamination strategies, posing significant risks of data leakage and potentially inflated performance assessments. Finally, we identify key open challenges in current research and outline future directions to narrow the gap between the theoretical capabilities of CodeLLMs and agents and their practical effectiveness in real-world scenarios.

cs.SE↗

arXiv2026

Spatial4D-Bench: A Versatile 4D Spatial Intelligence Benchmark

4D spatial intelligence involves perceiving and processing how objects move or change over time. Humans naturally possess 4D spatial intelligence, supporting a broad spectrum of spatial reasoning abilities. To what extent can Multimodal Large Language Models (MLLMs) achieve human-level 4D spatial intelligence? In this work, we present Spatial4D-Bench, a versatile 4D spatial intelligence benchmark designed to comprehensively assess the 4D spatial reasoning abilities of MLLMs. Unlike existing spatial intelligence benchmarks that are often small-scale or limited in diversity, Spatial4D-Bench provides a large-scale, multi-task evaluation benchmark consisting of ~40,000 question-answer pairs covering 18 well-defined tasks. We systematically organize these tasks into six cognitive categories: object understanding, scene understanding, spatial relationship understanding, spatiotemporal relationship understanding, spatial reasoning and spatiotemporal reasoning. Spatial4D-Bench thereby offers a structured and comprehensive benchmark for evaluating the spatial cognition abilities of MLLMs, covering a broad spectrum of tasks that parallel the versatility of human spatial intelligence. We benchmark various state-of-the-art open-source and proprietary MLLMs on Spatial4D-Bench and reveal their substantial limitations in a wide variety of 4D spatial reasoning aspects, such as route plan, action recognition, and physical plausibility reasoning. We hope that the findings provided in this work offer valuable insights to the community and that our benchmark can facilitate the development of more capable MLLMs toward human-level 4D spatial intelligence. More resources can be found on our project page.

cs.CV↗

arXiv2026

Melting of quantum Hall Wigner and bubble crystals

A two-dimensional crystal melts via the proliferation and unbinding of topological defects, yet quantitatively predicting the melting temperature $T_m$ in real systems is challenging. Here we resolve this discrepancy in quantum Hall electron bubble phases by combining Corbino-geometry transport experiment in an ultraclean GaAs/AlGaAs quantum well for Landau levels 2 to 5 with Hartree--Fock elasticity and the full Kosterlitz--Thouless--Halperin--Nelson--Young melting criterion including the finite-temperature renormalization-group calculation. The theoretically obtained $T_m$ quantitatively captures the measured solid-liquid phase transition boundaries across all probed ranges, validating the bubble-crystal interpretation and establishing defect--mediated melting as a predictive framework for strongly interacting electronic solids. This agreement further supports using bulk transport to probe the energetics of topological defects and screening in quantum Hall physics, and the approach is readily extendable to other electronic crystals, including the generalized Wigner crystal in moiré Chern bands.

cond-mat.mes-hall↗

arXiv2026

A Classification of Flexible Kokotsakis Polyhedra with Reducible Quadrilaterals

We study a class of mechanisms known as Kokotsakis polyhedra with a quadrangular base. These are $3\times3$ quadrilateral meshes whose faces are rigid bodies and joined by hinges at the common edges. In contrast to existing work, the quadrilateral faces do not necessarily have to be planar. In general, such a mesh is rigid. The problem of finding and classifying the flexible ones is old, but until now largely unsolved. It appears that the tangent values of the dihedral angles between different faces are algebraically related through polynomials. Specifically, this article deals with the case when these polynomials are reducible. We explore the conditions for reducibility to characterize all possible shape restrictions that lead to flexible polyhedra.

math.AG↗

arXiv2026

A Scalable Benchmark for Repository-Oriented Long-Horizon Conversational Context Management

In recent years, large language models (LLMs) have advanced rapidly, substantially enhancing their code understanding and generation capabilities and giving rise to powerful code assistants. However, in practical repository development, excessively long-horizon conversational context may overwhelm models, causing the loss of critical information and degraded performance, thereby limiting the utility of code assistants. Existing context management methods proposed to mitigate this context dilemma primarily target general-purpose conversations, while repository-oriented solutions remain largely unexplored, which is largely due to the lack of reliable evaluation benchmarks. To bridge this gap, we present LoCoEval, the first long-horizon conversational context management benchmark tailored to repository-oriented development scenarios. Adhering to three key principles, LoCoEval is constructed via an LLM-driven pipeline that generates realistic and diverse repository-oriented conversations, capturing key interaction patterns such as iterative requirements, noisy input, and retrospective questions. We evaluate 7 baselines, including 4 representative context management methods, using 3 advanced backbone LLMs on LoCoEval. The results reveal substantial challenges faced by standalone LLMs and existing approaches, especially memory systems, in repository-oriented conversational scenarios. To address these limitations, we further propose an improved method integrating conversational and repository information into a unified memory, which outperforms all baselines (*Oracle* excluded) and demonstrates robustness. Additionally, we investigated the impact of various factors on method performance, providing actionable insights for future research.

cs.SE↗

arXiv2026

DDP-WM: Disentangled Dynamics Prediction for Efficient World Models

World models are essential for autonomous robotic planning. However, the substantial computational overhead of existing dense Transformerbased models significantly hinders real-time deployment. To address this efficiency-performance bottleneck, we introduce DDP-WM, a novel world model centered on the principle of Disentangled Dynamics Prediction (DDP). We hypothesize that latent state evolution in observed scenes is heterogeneous and can be decomposed into sparse primary dynamics driven by physical interactions and secondary context-driven background updates. DDP-WM realizes this decomposition through an architecture that integrates efficient historical processing with dynamic localization to isolate primary dynamics. By employing a crossattention mechanism for background updates, the framework optimizes resource allocation and provides a smooth optimization landscape for planners. Extensive experiments demonstrate that DDP-WM achieves significant efficiency and performance across diverse tasks, including navigation, precise tabletop manipulation, and complex deformable or multi-body interactions. Specifically, on the challenging Push-T task, DDP-WM achieves an approximately 9 times inference speedup and improves the MPC success rate from 90% to98% compared to state-of-the-art dense models. The results establish a promising path for developing efficient, high-fidelity world models. Codes is available at https://hcplab-sysu.github.io/DDP-WM/.

cs.CV↗