arXiv subjects

Yu Wu

Publications and source records attributed to Yu Wu.

At least 55 records · Page 3Linked to original sources

arXiv2026

Dual Murnaghan-Nakayama rule for Hecke algebras in Type $A$

Let $χ^λ_μ$ be the value of the irreducible character $χ^λ$ of the Hecke algebra of the symmetric group on the conjugacy class of type $μ$. The usual Murnaghan-Nakayama rule provides an iterative algorithm based on reduction of the lower partition $μ$. In this paper, we establish a dual Murnaghan-Nakayama rule for Hecke algebras of type $A$ using vertex operators by applying reduction to the upper partition $λ$. We formulate an explicit recursion of the dual Murnaghan-Nakayama rule by employing the combinatorial model of ``brick tabloids", which refines a previous result by two of us (J. Algebra 598 (2022), 24--47).

math.RT↗

arXiv2026

SafeRedirect: Defeating Internal Safety Collapse via Task-Completion Redirection in Frontier LLMs

Internal Safety Collapse (ISC) is a failure mode in which frontier LLMs, when executing legitimate professional tasks whose correct completion structurally requires harmful content, spontaneously generate that content with safety failure rates exceeding 95%. Existing input-level defenses achieve a 100% failure rate against ISC, and standard system prompt defenses provide only partial mitigation. We propose SafeRedirect, a system-level override that defeats ISC by redirecting the model's task-completion drive rather than suppressing it. SafeRedirect grants explicit permission to fail the task, prescribes a deterministic hard-stop output, and instructs the model to preserve harmful placeholders unresolved. Evaluated on seven frontier LLMs across three AI/ML-related ISC task types in the single-turn setting, SafeRedirect reduces average unsafe generation rates from 71.2% to 8.0%, compared to 55.0% for the strongest viable baseline. Multi-model ablation reveals that failure permission and condition specificity are universally critical, while the importance of other components varies across models. Cross-attack evaluation confirms state-of-the-art defense against ISC with generalization performance at least on par with the baseline on other attack families. Code is available at https://github.com/fzjcdt/SafeRedirect.

cs.CR↗

arXiv2026

Breaking the Generator Barrier: Disentangled Representation for Generalizable AI-Text Detection

As large language models (LLMs) generate text that increasingly resembles human writing, the subtle cues that distinguish AI-generated content from human-written content become increasingly challenging to capture. Reliance on generator-specific artifacts is inherently unstable, since new models emerge rapidly and reduce the robustness of such shortcuts. This generalizes unseen generators as a central and challenging problem for AI-text detection. To tackle this challenge, we propose a progressively structured framework that disentangles AI-detection semantics from generator-aware artifacts. This is achieved through a compact latent encoding that encourages semantic minimality, followed by perturbation-based regularization to reduce residual entanglement, and finally a discriminative adaptation stage that aligns representations with task objectives. Experiments on MAGE benchmark, covering 20 representative LLMs across 7 categories, demonstrate consistent improvements over state-of-the-art methods, achieving up to 24.2% accuracy gain and 26.2% F1 improvement. Notably, performance continues to improve as the diversity of training generators increases, confirming strong scalability and generalization in open-set scenarios. Our source code will be publicly available at https://github.com/PuXiao06/DRGD.

cs.CL↗

arXiv2026

Coupling of phase transition, anharmonicity, and thermal transport in CaSnF$_6$

Understanding the coupling between structural phase transitions and thermal transport is essential for designing functional materials with tunable properties. Here, we investigate this interplay in CaSnF$_6$ by combining first-principles calculations with a machine-learned neuroevolution potential that enables large-scale molecular dynamics simulations across a wide temperature range. The simulations accurately capture the first-order structural phase transition and associated lattice dynamics. We show that the negative thermal expansion originates from low-energy rigid unit modes involving cooperative rotations of corner-sharing [CaF$_6$]$^{4-}$ octahedra, which induce bond-angle bending and volume contraction. At the same time, strong anharmonicity, dominated by four-phonon scattering, plays a central role in suppressing lattice thermal conductivity ($κ_L$). Crucially, non-equilibrium simulations reveal a pronounced non-monotonic anomaly in $κ_L$ near the phase transition, deviating from the conventional $\sim 1/T^α$ behavior and providing direct transport evidence of lattice reconstruction. These results establish a unified mechanism linking lattice geometry, anharmonic vibrational dynamics, and thermal transport, and highlight the potential of machine-learned potentials for bridging atomic-scale phase transitions with macroscopic transport properties.

cond-mat.mtrl-sci↗

arXiv2026

From UAV Imagery to Agronomic Reasoning: A Multimodal LLM Benchmark for Plant Phenotyping

To improve crop genetics, high-throughput, effective and comprehensive phenotyping is a critical prerequisite. While such tasks were traditionally performed manually, recent advances in multimodal foundation models, especially in vision-language models (VLMs), have enabled more automated and robust phenotypic analysis. However, plant science remains a particularly challenging domain for foundation models because it requires domain-specific knowledge, fine-grained visual interpretation, and complex biological and agronomic reasoning. To address this gap, we develop PlantXpert, an evidence-grounded multimodal reasoning benchmark for soybean and cotton phenotyping. Our benchmark provides a structured and reproducible framework for agronomic adaptation of VLMs, and enables controlled comparison between base models and their domain-adapted counterparts. We constructed a dataset comprising 385 digital images and more than 3,000 benchmark samples spanning key plant science domains including disease, pest control, weed management, and yield. The benchmark can assess diverse capabilities including visual expertise, quantitative reasoning, and multi-step agronomic reasoning. A total of 11 state-of-the-art VLMs were evaluated. The results indicate that task-specific fine-tuning leads to substantial improvement in accuracy, with models such as Qwen3-VL-4B and Qwen3-VL-30B achieving up to 78%. At the same time, gains from model scaling diminish beyond a certain capacity, generalization across soybean and cotton remains uneven, and quantitative as well as biologically grounded reasoning continue to pose substantial challenges. These findings suggest that PlantXpert can serve as a foundation for assessing evidence-grounded agronomic reasoning and for advancing multimodal model development in plant science.

cs.CV↗

arXiv2026

A plug-and-play superconducting quantum controller at millikelvin temperatures enables exceeding 99.9% average gate fidelity

The development of large-scale superconducting quantum computing requires efficient in-situ control methods that allow high-fidelity operations at millikelvin temperatures. Superconducting circuits based on Josephson junctions offer a promising solution due to their high speed, low power dissipation, and cryogenic nature. Here, we report a superconducting quantum controller that enables direct chip-to-chip interconnection with qubits at 10 mK and high-fidelity, all-digital manipulation. Randomized benchmarking reveals a uniformly high average Clifford fidelity of 99.9% with leakage to high energy levels on the order of $10^{-4}$, and an estimated average gate operation energy of 0.121 fJ, demonstrating the potential to resolve the control bottleneck in superconducting quantum computing.

quant-ph↗

arXiv2026

Visual Prototype Conditioned Focal Region Generation for UAV-Based Object Detection

Unmanned aerial vehicle (UAV) based object detection is a critical but challenging task, when applied in dynamically changing scenarios with limited annotated training data. Layout-to-image generation approaches have proved effective in promoting detection accuracy by synthesizing labeled images based on diffusion models. However, they suffer from frequently producing artifacts, especially near layout boundaries of tiny objects, thus substantially limiting their performance. To address these issues, we propose UAVGen, a novel layout-to-image generation framework tailored for UAV-based object detection. Specifically, UAVGen designs a Visual Prototype Conditioned Diffusion Model (VPC-DM) that constructs representative instances for each class and integrates them into latent embeddings for high-fidelity object generation. Moreover, a Focal Region Enhanced Data Pipeline (FRE-DP) is introduced to emphasize object-concentrated foreground regions in synthesis, combined with a label refinement to correct missing, extra and misaligned generations. Extensive experimental results demonstrate that our method significantly outperforms state-of-the-art approaches, and consistently promotes accuracy when integrated with distinct detectors. The source code is available at https://github.com/Sirius-Li/UAVGen.

cs.CV↗

arXiv2026

RiskProp: Collision-Anchored Self-Supervised Risk Propagation for Early Accident Anticipation

Accident anticipation aims to predict impending collisions from dashcam videos and trigger early alerts. Existing methods rely on binary supervision with manually annotated "anomaly onset" frames, which are subjective and inconsistent, leading to inaccurate risk estimation. In contrast, we propose RiskProp, a novel collision-anchored self-supervised risk propagation paradigm for early accident anticipation, which removes the need for anomaly onset annotations and leverages only the reliably annotated collision frame. RiskProp models temporal risk evolution through two observation-driven losses: first, since future frames contain more definitive evidence of an impending accident, we introduce a future-frame regularization loss that uses the model's next-frame prediction as a soft target to supervise the current frame, enabling backward propagation of risk signals; second, inspired by the empirical trend of rising risk before accidents, we design an adaptive monotonic constraint to encourage a non-decreasing progression over time. Experiments on CAP and Nexar demonstrate that RiskProp achieves state-of-the-art performance and produces smoother, more discriminative risk curves, improving both early anticipation and interpretability.

cs.CV↗

arXiv2026

Towards One-step Causal Video Generation via Adversarial Self-Distillation

Recent hybrid video generation models combine autoregressive temporal dynamics with diffusion-based spatial denoising, but their sequential, iterative nature leads to error accumulation and long inference times. In this work, we propose a distillation-based framework for efficient causal video generation that enables high-quality synthesis with extremely limited denoising steps. Our approach builds upon the Distribution Matching Distillation (DMD) framework and proposes a novel Adversarial Self-Distillation (ASD) strategy, which aligns the outputs of the student model's n-step denoising process with its (n+1)-step version at the distribution level. This design provides smoother supervision by bridging small intra-student gaps and more informative guidance by combining teacher knowledge with locally consistent student behavior, substantially improving training stability and generation quality in extremely few-step scenarios (e.g., 1-2 steps). In addition, we present a First-Frame Enhancement (FFE) strategy, which allocates more denoising steps to the initial frames to mitigate error propagation while applying larger skipping steps to later frames. Extensive experiments on VBench demonstrate that our method surpasses state-of-the-art approaches in both one-step and two-step video generation. Notably, our framework produces a single distilled model that flexibly supports multiple inference-step settings, eliminating the need for repeated re-distillation and enabling efficient, high-quality video synthesis.

cs.CV↗

arXiv2026

Evolution of Phonon Transport Across Structural Phase Transitions in MgAgSb

MgAgSb, a promising thermoelectric material, undergoes reversible phase transitions that drastically alter its thermal transport behavior. Using first-principles calculations, we systematically investigate the lattice thermal conductivity ($κ_L$) of its three phases: $α$, $β$, and $γ$, revealing a progressive increase following $α< β< γ$. This trend originates from distinct scattering mechanisms. Four-phonon scattering substantially suppresses the particle-like conductivity ($κ_p$) in the $β$ and $γ$ phases, while electron-phonon scattering provides a minor additional reduction. In contrast, the wave-like conductivity ($κ_c$) from coherent phonon tunneling is highest in the complex $α$ phase, contributing up to 44\% of $κ_L$. Notably, the temperature dependence of $κ_L$ differs fundamentally between phases: in $β$, the weak $κ_p$ variation arises from a decreasing Grüneisen parameter with temperature; in $α$, the strong rise in $κ_c$ with temperature counteracts the decay of $κ_p$. Our findings establish a comprehensive picture of thermal transport in MgAgSb, highlighting the phase-dependent interplay between particle-like and wave-like phonon contributions.

cond-mat.mtrl-sci↗

arXiv2026

xLLM Technical Report

We introduce xLLM, an intelligent and efficient Large Language Model (LLM) inference framework designed for high-performance, large-scale enterprise-grade serving, with deep optimizations for diverse AI accelerators. To address these challenges, xLLM builds a novel decoupled service-engine architecture. At the service layer, xLLM-Service features an intelligent scheduling module that efficiently processes multimodal requests and co-locates online and offline tasks through unified elastic scheduling to maximize cluster utilization. This module also relies on a workload-adaptive dynamic Prefill-Decode (PD) disaggregation policy and a novel Encode-Prefill-Decode (EPD) disaggregation policy designed for multimodal inputs. Furthermore, it incorporates a distributed architecture to provide global KV Cache management and robust fault-tolerant capabilities for high availability. At the engine layer, xLLM-Engine co-optimizes system and algorithm designs to fully saturate computing resources. This is achieved through comprehensive multi-layer execution pipeline optimizations, an adaptive graph mode and an xTensor memory management. xLLM-Engine also further integrates algorithmic enhancements such as optimized speculative decoding and dynamic EPLB, collectively serving to substantially boost throughput and inference efficiency. Extensive evaluations demonstrate that xLLM delivers significantly superior performance and resource efficiency. Under identical TPOT constraints, xLLM achieves throughput up to 1.7x that of MindIE and 2.2x that of vLLM-Ascend with Qwen-series models, while maintaining an average throughput of 1.7x that of MindIE with Deepseek-series models. xLLM framework is publicly available at https://github.com/jd-opensource/xllm and https://github.com/jd-opensource/xllm-service.

cs.DC↗

arXiv2026

CRCC: Contrast-Based Robust Cross-Subject and Cross-Site Representation Learning for EEG

EEG-based neural decoding models often fail to generalize across acquisition sites due to structured, site-dependent biases implicitly exploited during training. We reformulate cross-site clinical EEG learning as a bias-factorized generalization problem, in which domain shifts arise from multiple interacting sources. We identify three fundamental bias factors and propose a general training framework that mitigates their influence through data standardization and representation-level constraints. We construct a standardized multi-site EEG benchmark for Major Depressive Disorder and introduce CRCC, a two-stage training paradigm combining encoder-decoder pretraining with joint fine-tuning via cross-subject/site contrastive learning and site-adversarial optimization. CRCC consistently outperforms state-of-the-art baselines and achieves a 10.7 percentage-point improvement in balanced accuracy under strict zero-shot site transfer, demonstrating robust generalization to unseen environments.

q-bio.NC↗

arXiv2026

Incentivizing Agentic Reasoning in LLM Judges via Tool-Integrated Reinforcement Learning

Large Language Models (LLMs) are widely used as judges to evaluate response quality, providing a scalable alternative to human evaluation. However, most LLM judges operate solely on intrinsic text-based reasoning, limiting their ability to verify complex constraints or perform accurate computation. Motivated by the success of tool-integrated reasoning (TIR) in numerous tasks, we propose TIR-Judge, an end-to-end RL framework for training LLM judges that integrates a code executor for precise evaluation. TIR-Judge is built on three principles: (i) diverse training across verifiable and non-verifiable domains, (ii) flexible judgment formats (pointwise, pairwise, listwise), and (iii) iterative RL that bootstraps directly from the initial model without distillation. On seven public benchmarks, TIR-Judge surpasses strong reasoning-based judges by up to 6.4% (pointwise) and 7.7% (pairwise), and achieves listwise performance comparable to Claude-Opus-4 despite having only 8B parameters. Remarkably, TIR-Judge-Zero - trained entirely without distilled judge trajectories, matches the performance of distilled variants, demonstrating that tool-augmented judges can self-evolve through iterative reinforcement learning.

cs.CL↗

arXiv2026

Rotational Soft Modes and Octahedral Distortion as Design Principles for Ultralow Thermal Conductivity in Halide Materials

We establish that ultralow lattice thermal conductivity in halide perovskites and related octahedral framework materials arises from two distinct and complementary mechanisms: (i) halogen-halogen-enabled rotational soft modes that reshape the low-frequency spectrum and intensify phonon scattering, and (ii) static octahedral distortions that further enhance anharmonicity and reduce phonon lifetimes. Using first-principles calculations on CsPbBr3, we demonstrate that Br-Br interactions induce rotational soft modes that decongest the phonon spectrum and enhance three- and four-phonon scattering, strongly suppressing particle-like thermal conductivity (kappa_p). Independently, static octahedral distortions further reduce kappa_p by amplifying anharmonicity while leaving wave-like conductivity (kappa_c) intact. Based on these mechanistic insights, we introduce a geometric distortion factor rho and perform a high-throughput screening that first selects materials with halogen-coordinated octahedral building blocks-ensuring the presence of rotational soft modes-and then identifies those with pronounced distortion. This strategy uncovers TaGaI8 with an ultralow kappa_L = 0.11 W/mK at room temperature. This work establishes halogen-halogen-enabled rotational soft modes and octahedral distortions as transferable design principles for octahedra-containing halides, spanning both extended frameworks and molecular-cluster motifs, for discovering ultralow-kappa_L materials.

cond-mat.mtrl-sci↗

arXiv2026

Detecting Latin in Historical Books with Large Language Models: A Multimodal Benchmark

This paper presents a novel task of extracting low-resourced and noisy Latin fragments from mixed-language historical documents with varied layouts. We benchmark and evaluate the performance of large foundation models against a multimodal dataset of 724 annotated pages. The results demonstrate that reliable Latin detection with contemporary zero-shot models is achievable, yet these models lack a functional comprehension of Latin. This study establishes a comprehensive baseline for processing Latin within mixed-language corpora, supporting quantitative analysis in intellectual history and historical linguistics. Both the dataset and code are available at https://github.com/COMHIS/EACL26-detect-latin.

cs.CL↗

arXiv2026

From Deferral to Learning: Online In-Context Knowledge Distillation for LLM Cascades

Standard LLM cascades improve efficiency by deferring difficult queries from weak to strong models. However, these systems are typically static: when faced with repeated or semantically similar queries, they redundantly consult the expensive model, failing to adapt during inference. To address this, we propose Inter-Cascade, an online, interactive framework that transforms the strong model from a temporary helper into a long-term teacher. In our approach, when the strong model resolves a deferred query, it generates a generalized, reusable problem-solving strategy. These strategies are stored in a dynamic repository and retrieved via similarity matching to augment the weak model's context for future queries. This enables the weak model to learn on the job without expensive parameter fine-tuning. We theoretically show that this mechanism improves the weak model's confidence calibration. Empirically, Inter-Cascade outperforms standard cascades on multiple benchmarks, improving weak model and overall system accuracy by up to 33.06 percent and 6.35 percent, while reducing strong model calls by up to 48.05 percent and saving fee by up to 49.63 percent. Inter-Cascade demonstrates effective in-context knowledge transfer between LLMs and provides a general, scalable framework applicable to both open-source and API-based LLMs.

cs.AI↗

arXiv2026

BNMusic: Blending Environmental Noises into Personalized Music

While being disturbed by environmental noises, the acoustic masking technique is a conventional way to reduce the annoyance in audio engineering that seeks to cover up the noises with other dominant yet less intrusive sounds. However, misalignment between the dominant sound and the noise-such as mismatched downbeats-often requires an excessive volume increase to achieve effective masking. Motivated by recent advances in cross-modal generation, in this work, we introduce an alternative method to acoustic masking, aiming to reduce the noticeability of environmental noises by blending them into personalized music generated based on user-provided text prompts. Following the paradigm of music generation using mel-spectrogram representations, we propose a Blending Noises into Personalized Music (BNMusic) framework with two key stages. The first stage synthesizes a complete piece of music in a mel-spectrogram representation that encapsulates the musical essence of the noise. In the second stage, we adaptively amplify the generated music segment to further reduce noise perception and enhance the blending effectiveness, while preserving auditory quality. Our experiments with comprehensive evaluations on MusicBench, EPIC-SOUNDS, and ESC-50 demonstrate the effectiveness of our framework, highlighting the ability to blend environmental noise with rhythmically aligned, adaptively amplified, and enjoyable music segments, minimizing the noticeability of the noise, thereby improving overall acoustic experiences. Project page: https://d-fas.github.io/BNMusic_page/.

cs.SD↗

arXiv2026

Will It Zero-Shot?: Predicting Zero-Shot Classification Performance For Arbitrary Queries

Vision-Language Models like CLIP create aligned embedding spaces for text and images, making it possible for anyone to build a visual classifier by simply naming the classes they want to distinguish. However, a model that works well in one domain may fail in another, and non-expert users have no straightforward way to assess whether their chosen VLM will work on their problem. We build on prior work using text-only comparisons to evaluate how well a model works for a given natural language task, and explore approaches that also generate synthetic images relevant to that task to evaluate and refine the prediction of zero-shot accuracy. We show that generated imagery to the baseline text-only scores substantially improves the quality of these predictions. Additionally, it gives a user feedback on the kinds of images that were used to make the assessment. Experiments on standard CLIP benchmark datasets demonstrate that the image-based approach helps users predict, without any labeled examples, whether a VLM will be effective for their application.

cs.CV↗