Searcharxiv⌕ Search

arXiv subjects

Xu Yang

Publications and source records attributed to Xu Yang.

At least 127 records · Page 7Linked to original sources
arXiv2026

SAGA: A Robust Self-Attention and Goal-Aware Anchor-based Planner for Safe UAV Autonomous Navigation

Agile unmanned aerial vehicle (UAV) navigation in cluttered environments demands a planning architecture that is both computationally efficient and structurally expressive enough to reason over multiple feasible motions. This paper presents SAGA, a robust self-attention and goal-aware anchor-based planner for safe UAV autonomous navigation. SAGA formulates local planning as a one-stage joint regression-and-ranking problem over a fixed lattice of motion anchors. Given a depth image and a body-frame motion state, the planner predicts refined terminal states and planning scores for all anchors in a single forward pass, after which the best candidate is decoded into a dynamically feasible trajectory. The key idea of SAGA is to transform anchor-aligned features into geometry-aware tokens and perform cross-anchor global reasoning with self-attention. To preserve directional structure in the token space, we further introduce a polar positional encoding derived from anchor yaw and pitch. In addition, a goal-aware modulation module injects velocity, acceleration, and target information into the token representation before final score prediction. Experiments in cluttered pillar-map environments under maximum speed settings of 2.0, 3.0, and 4.0~m/s show that SAGA consistently achieves a 100\% success rate, while YOPO drops from 90.91\% to 62.50\%, Ego-planner from 71.43\% to 52.63\%, and Fast-planner from 52.63\% to 38.46\%. Under the 4.0~m/s maximum speed setting, SAGA also improves average safety from 1.9843~m to 2.3888~m and minimum safety from 0.4390~m to 0.7576~m over YOPO, while reducing total flight time from 40.4631~s to 27.4901~s. The comparison with SAGA w/o PPE further shows that explicit polar positional encoding is critical for stable cross-anchor reasoning and safe passage selection in cluttered scenes.

cs.RO↗
arXiv2026

Thermodynamics and phase transitions of nonlinearly scalarized black holes in Einstein-scalar-Gauss-Bonnet theory

We investigate the thermodynamic properties of static nonlinearly scalarized black holes in Einstein-scalar-Gauss-Bonnet theory with polynomial coupling functions. Based on the scalarized solutions constructed previously, we compute thermodynamical quantities of these scalarized black holes. Moreover, we examine the first law of black hole thermodynamics and consider the phase transitions between Schwarzschild and scalarized black holes. It shows that a phase transition from Schwarzschild black hole to scalarized black hole is a first-order with non-zero latent heat.

gr-qc↗
arXiv2026

Large Language Model as An Operator: An Experience-Driven Solution for Distribution Network Voltage Control

With the advanced reasoning, contextual understanding, and information synthesis capabilities of large language models (LLMs), a novel paradigm emerges for the autonomous generation of dispatch strategies in modern power systems. In this paper, we propose an LLM-based experience-driven day-ahead Volt/Var schedule solution for distribution networks, which enables the self-evolution of LLM agent's strategies through the collaboration and interaction of multiple modules, specifically, experience storage, experience retrieval, experience generation, and experience modification. The experience storage module archives historical operational records and decisions, while the retrieval module selects relevant past cases according to current forecasting conditions. The LLM agent then leverages these retrieved experiences to generate new, context-aware decisions for current situation, which are subsequently refined by the modification module to realize self-evolution of the dispatch policy. Comprehensive experimental results validate the effectiveness of the proposed method and highlight the applicability of LLMs in power system dispatch problems facing incomplete information.

eess.SY↗
arXiv2026

Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search

LLM-based agents for machine learning engineering (MLE) predominantly rely on tree search, a form of gradient-free optimization that uses scalar validation scores to rank candidates. As LLM reasoning capabilities improve, exhaustive enumeration becomes increasingly inefficient compared to directed updates, analogous to how accurate gradients enable efficient descent over random search. We introduce Gome, an MLE agent that operationalizes gradient-based optimization. Gome maps structured diagnostic reasoning to gradient computation, success memory to momentum, and multi-trace execution to distributed optimization. Under a closed-world protocol that isolates architectural effects from external knowledge, Gome achieves a state-of-the-art 35.1\% any-medal rate on MLE-Bench with a restricted 12-hour budget on a single V100 GPU. Scaling experiments across 10 models reveal a critical crossover: with weaker models, tree search retains advantages by compensating for unreliable reasoning through exhaustive exploration; as reasoning capability strengthens, gradient-based optimization progressively outperforms, with the gap widening at frontier-tier models. Given the rapid advancement of reasoning-oriented LLMs, this positions gradient-based optimization as an increasingly favorable paradigm. We release our codebase and GPT-5 traces at https://github.com/microsoft/RD-Agent.

cs.LG↗
arXiv2026

Nano-EmoX: Unifying Multimodal Emotional Intelligence from Perception to Empathy

The development of affective multimodal language models (MLMs) has long been constrained by a gap between low-level perception and high-level interaction, leading to fragmented affective capabilities and limited generalization. To bridge this gap, we propose a cognitively inspired three-level hierarchy that organizes affective tasks according to their cognitive depth-perception, understanding, and interaction-and provides a unified conceptual foundation for advancing affective modeling. Guided by this hierarchy, we introduce Nano-EmoX, a small-scale multitask MLM, and P2E (Perception-to-Empathy), a curriculum-based training framework. Nano-EmoX integrates a suite of omni-modal encoders, including an enhanced facial encoder and a fusion encoder, to capture key multimodal affective cues and improve cross-task transferability. The outputs are projected into a unified language space via heterogeneous adapters, empowering a lightweight language model to tackle diverse affective tasks. Concurrently, P2E progressively cultivates emotional intelligence by aligning rapid perception with chain-of-thought-driven empathy. To the best of our knowledge, Nano-EmoX is the first compact MLM (2.2B) to unify six core affective tasks across all three hierarchy levels, achieving state-of-the-art or highly competitive performance across multiple benchmarks, demonstrating excellent efficiency and generalization. The code is available at https://github.com/waHAHJIAHAO/Nano-EmoX.

cs.AI↗
arXiv2026

Adversarial Decision-Making in Partially Observable Multi-Agent Systems: A Sequential Hypothesis Testing Approach

Adversarial decision-making in partially observable multi-agent systems requires sophisticated strategies for both deception and counter-deception. This paper presents a sequential hypothesis testing (SHT)-driven framework that captures the interplay between strategic misdirection and inference in adversarial environments. We formulate this interaction as a partially observable Stackelberg game, where a follower agent (blue team) seeks to fulfill its primary task while actively misleading an adversarial leader (red team). In opposition, the red team, leveraging leaked information, instills carefully designed patterns to manipulate the blue team's behavior, mitigating the misdirection effect. Unlike conventional approaches that focus on robust control under adversarial uncertainty, our framework explicitly models deception as a dynamic optimization problem, where both agents strategically adapt their policies in response to inference and counter-inference. We derive a semi-explicit optimal control solution for the blue team within a linear-quadratic setting and develop iterative and machine learning-based methods to characterize the red team's optimal response. Numerical experiments demonstrate how deception-driven strategies influence adversarial interactions and reveal the impact of leaked information in shaping equilibrium behaviors. These results provide new insights into strategic deception in multi-agent systems, with potential applications in cybersecurity, autonomous decision-making, and financial markets.

math.OC↗
arXiv2026

The Three-Dimensional Stochastic EMHD System: Local Well-Posedness and Maximal Pathwise Solutions

We study the three-dimensional stochastic electron magnetohydrodynamics (EMHD) system with fractional dissipation on the torus, driven by Stratonovich transport noise acting through divergence-free first-order operators. The noise generates an Itô correction while preserving the transport structure of the Hall nonlinearity. Since the Hall term contains one more derivative, in the stochastic setting it must be controlled together with commutators arising from the transport operators. We develop a high-order Sobolev energy method based on Littlewood--Paley analysis and refined commutator estimates, which yields uniform bounds for Galerkin approximations in $H^s$ with $s > \tfrac{5}{2}$ together with suitable time regularity. Using stochastic compactness and identification of limits, we construct martingale solutions for initial data in $L^2(Ω; H^s)$. Pathwise uniqueness follows from cancellations in the Hall term combined with a stochastic Grönwall argument. An application of a Yamada--Watanabe type result then yields local pathwise well-posedness and the existence of maximal pathwise solutions.

math.PR↗
arXiv2026

An Actor-Critic Framework for Continuous-Time Jump-Diffusion Controls with Normalizing Flows

Continuous-time stochastic control with time-inhomogeneous jump-diffusion dynamics is central in finance and economics, but computing optimal policies is difficult under explicit time dependence, discontinuous shocks, and high dimensionality. We propose an actor-critic framework that serves as a mesh-free solver for entropy-regularized control problems and stochastic games with jumps. The approach is built on a time-inhomogeneous little q-function and an appropriate occupation measure, yielding a policy-gradient representation that accommodates time-dependent drift, volatility, and jump terms. To represent expressive stochastic policies in continuous-action spaces, we parameterize the actor using conditional normalizing flows, enabling flexible non-Gaussian policies while retaining exact likelihood evaluation for entropy regularization and policy optimization. We validate the method on time-inhomogeneous linear-quadratic control, Merton portfolio optimization, and a multi-agent portfolio game, using explicit solutions or high-accuracy benchmarks. Numerical results demonstrate stable learning under jump discontinuities, accurate approximation of optimal stochastic policies, and favorable scaling with respect to dimension and number of agents.

math.OC↗
arXiv2026

Benchmarking Multi-turn Medical Diagnosis: Hold, Lure, and Self-Correction

Large language models (LLMs) achieve high accuracy in medical diagnosis when all clinical information is provided in a single turn, yet how they behave under multi-turn evidence accumulation closer to real clinical reasoning remains unexplored. We introduce MINT (Medical Incremental N-Turn Benchmark), a high-fidelity, multi-turn medical diagnosis benchmark comprising 1,035 cases with clinically labeled evidence shards, controlled turn granularity, and information-preserving decomposition. Through systematic evaluation of 11 LLMs on MINT, we uncover three persistent behavioral patterns that significantly impact diagnostic decisions: (1) intent to answer, models rush to answer before sufficient evidence has been observed, with over 55% of answers committed within the first two turns; (2) self-correction, incorrect-to-correct answer revisions occur at up to 10.6 times the rate of correct-to-incorrect flips, revealing a latent capacity for self-correction that premature commitment forecloses; and (3) strong lures, clinically salient information such as laboratory results trigger premature answering even when models are explicitly instructed to wait. We translate these findings into clinically actionable guidance: deferring the diagnostic question to later turns reduces premature answering and improves accuracy at the first point of commitment by up to 62.6%, while reserving salient clinical evidence for later turns prevents a catastrophic accuracy drop of up to 23.3% caused by premature commitment. Our work provides both a controlled evaluation framework and concrete recommendations for improving the reliability of LLMs in multi-turn medical diagnosis.

cs.CL↗
arXiv2026

Open-Loop Planning, Closed-Loop Verification: Speculative Verification for VLA

Vision-Language-Action (VLA) models, as large foundation models for embodied control, have shown strong performance in manipulation tasks. However, their performance comes at high inference cost. To improve efficiency, recent methods adopt action chunking, which predicts a sequence of future actions for open-loop execution. Although effective for reducing computation, open-loop execution is sensitive to environmental changes and prone to error accumulation due to the lack of close-loop feedback. To address this limitation, we propose Speculative Verification for VLA Control (SV-VLA), a framework that combines efficient open-loop long-horizon planning with lightweight closed-loop online verification. Specifically, SV-VLA uses a heavy VLA as a low-frequency macro-planner to generate an action chunk together with a planning context, while a lightweight verifier continuously monitors execution based on the latest observations. Conditioned on both the current observation and the planning context, the verifier compares the planned action against a closed-loop reference action and triggers replanning only when necessary. Experiments demonstrate that SV-VLA combines the efficiency of chunked prediction with the robustness of closed-loop control, enabling efficient and reliable VLA-based control in dynamic environments. Code is available: https://github.com/edsad122/SV-VLA.

cs.RO↗
arXiv2026

A Note on Weak Saturation Number of Trees

In this paper, we estimate the weak saturation numbers of trees. As a case study, we examine caterpillars and obtain several tight estimates. In particular, this implies that for any $α\in [1,2]$, there exist caterpillars with $k$ vertices whose weak saturation numbers are of order $k^α$. We call a tree good if its weak saturation number is exactly its edge number minus one. We provide a sufficient condition for a tree to be a good tree. With the additional property that all leaves are at even distances from each other, this condition fully characterizes good trees.

math.CO↗
arXiv2026

GEditBench v2: A Human-Aligned Benchmark for General Image Editing

Recent advances in image editing have enabled models to handle complex instructions with impressive realism. However, existing evaluation frameworks lag behind: current benchmarks suffer from narrow task coverage, while standard metrics fail to adequately capture visual consistency, i.e., the preservation of identity, structure and semantic coherence between edited and original images. To address these limitations, we introduce GEditBench v2, a comprehensive benchmark with 1,200 real-world user queries spanning 23 tasks, including a dedicated open-set category for unconstrained, out-of-distribution editing instructions beyond predefined tasks. Furthermore, we propose PVC-Judge, an open-source pairwise assessment model for visual consistency, trained via two novel region-decoupled preference data synthesis pipelines. Besides, we construct VCReward-Bench using expert-annotated preference pairs to assess the alignment of PVC-Judge with human judgments on visual consistency evaluation. Experiments show that our PVC-Judge achieves state-of-the-art evaluation performance among open-source models and even surpasses GPT-5.1 on average. Finally, by benchmarking 16 frontier editing models, we show that GEditBench v2 enables more human-aligned evaluation, revealing critical limitations of current models, and providing a reliable foundation for advancing precise image editing.

cs.CV↗
arXiv2026

Hierarchical Control Framework Integrating LLMs with RL for Decarbonized HVAC Operation

Heating, ventilation, and air conditioning (HVAC) systems account for a substantial share of building energy consumption. Environmental uncertainty and dynamic occupancy behavior bring challenges in decarbonized HVAC control. Reinforcement learning (RL) can optimize long-horizon comfort-energy trade-offs but suffers from exponential action-space growth and inefficient exploration in multi-zone buildings. Large language models (LLMs) can encode semantic context and operational knowledge, yet when used alone they lack reliable closed-loop numerical optimization and may result in less reliable comfort-energy trade-offs. To address these limitations, we propose a hierarchical control framework in which a fine-tuned LLM, trained on historical building operation data, generates state-dependent feasible action masks that prune the combinatorial joint action space into operationally plausible subsets. A masked value-based RL agent then performs constrained optimization within this reduced space, improving exploration efficiency and training stability. Evaluated in a high-fidelity simulator calibrated with real-world sensor and occupancy data from a 7-zone office building, the proposed method achieves a mean PPD of 7.30%, corresponding to reductions of 39.1% relative to DQN, the best vanilla RL baseline in comfort, and 53.1% relative to the best vanilla LLM baseline, while reducing daily HVAC energy use to 140.90~kWh, lower than all vanilla RL baselines. The results suggest that LLM-guided action masking is a promising pathway toward efficient multi-zone HVAC control.

eess.SY↗
arXiv2026

Compensating Visual Insufficiency with Stratified Language Guidance for Long-Tail Class Incremental Learning

Long-tail class incremental learning (LT CIL) remains highly challenging because the scarcity of samples in tail classes not only hampers their learning but also exacerbates catastrophic forgetting under continuously evolving and imbalanced data distributions. To tackle these issues, we exploit the informativeness and scalability of language knowledge. Specifically, we analyze the LT CIL data distribution to guide large language models (LLMs) in generating a stratified language tree that hierarchically organizes semantic information from coarse to fine grained granularity. Building upon this structure, we introduce stratified adaptive language guidance, which leverages learnable weights to merge multi-scale semantic representations, thereby enabling dynamic supervisory adjustment for tail classes and alleviating the impact of data imbalance. Furthermore, we introduce stratified alignment language guidance, which exploits the structural stability of the language tree to constrain optimization and reinforce semantic visual alignment, thereby alleviating catastrophic forgetting. Extensive experiments on multiple benchmarks demonstrate that our method achieves state of the art performance.

cs.AI↗
arXiv2026

A Creative Agent is Worth a 64-Token Template

Text-to-image (T2I) models have substantially improved image fidelity and prompt adherence, yet their creativity remains constrained by reliance on discrete natural language prompts. When presented with fuzzy prompts such as ``a creative vinyl record-inspired skyscraper'', these models often fail to infer the underlying creative intent, leaving creative ideation and prompt design largely to human users. Recent reasoning- or agent-driven approaches iteratively augment prompts but incur high computational and monetary costs, as their instance-specific generation makes ``creativity'' costly and non-reusable, requiring repeated queries or reasoning for subsequent generations. To address this, we introduce \textbf{CAT}, a framework for \textbf{C}reative \textbf{A}gent \textbf{T}okenization that encapsulates agents' intrinsic understanding of ``creativity'' through a \textit{Creative Tokenizer}. Given the embeddings of fuzzy prompts, the tokenizer generates a reusable token template that can be directly concatenated with them to inject creative semantics into T2I models without repeated reasoning or prompt augmentation. To enable this, the tokenizer is trained via creative semantic disentanglement, leveraging relations among partially overlapping concept pairs to capture the agent's latent creative representations. Extensive experiments on \textbf{\textit{Architecture Design}}, \textbf{\textit{Furniture Design}}, and \textbf{\textit{Nature Mixture}} tasks demonstrate that CAT provides a scalable and effective paradigm for enhancing creativity in T2I generation, achieving a $3.7\times$ speedup and a $4.8\times$ reduction in computational cost, while producing images with superior human preference and text-image alignment compared to state-of-the-art T2I models and creative generation methods.

cs.CV↗
arXiv2026

CICDWOA: A Collective Cognitive Sharing Whale Optimization Algorithm with Cauchy Inverse Cumulative Distribution for 2D/3D Path Planning and Engineering Design Problems

The Whale Optimization Algorithm (WOA) has shown strong optimization ability but still suffers from premature convergence and weak search diversity. To address these issues, this paper proposes an enhanced WOA variant called CICDWOA. The proposed algorithm introduces a Good Nodes Set (GNS) method for uniform population initialization, a Collective Cognitive Sharing (CCS) mechanism to enhance group collaboration, and an Enhanced Spiral Updating strategy based on the Cauchy Inverse Cumulative Distribution (CICD) to strengthen global exploration and local exploitation balance. In addition, a nonlinear convergence factor and a Hybrid Gaussian-Cauchy mutation based on Differential Evolution (DE) further improve convergence efficiency and population diversity. CICDWOA was evaluated on 23 benchmark functions, 2D robot path planning problems, 3D UAV path planning tasks and 10 engineering design problems. Statistical experiment results show that CICDWOA achieves faster convergence, higher accuracy, and better robustness than classical WOA and other advanced metaheuristic algorithms. CICDWOA gained average Friedman value of 1.6790, ranking first among the SOTA algorithms. And the results of engineering simulations confirm that CICDWOA provides an effective and general framework for solving complex optimization and engineering problems. The code of CICDWOA are available on \href{URL}{https://github.com/JunhaoWei-mpu/ROBIS-Lab/tree/CICDWOA}.

cs.CE↗
arXiv2026

Self-Indexing KVCache: Predicting Sparse Attention from Compressed Keys

The KV cache in self-attention has emerged as a major bottleneck in long-context and large-batch inference for LLMs. Existing approaches often treat sparsity prediction and compression as separate modules, relying on auxiliary index structures to select relevant tokens, and on complex quantization schemes to reduce memory usage. This fragmented design introduces redundant overhead and limits scalability. In this paper, we propose a novel paradigm: treating the compressed key representation not merely as storage, but as a self-indexing structure that directly enables efficient sparse attention. By designing a sign-based 1-bit vector quantization (VQ) scheme, our method unifies compression and retrieval in a single, hardware-friendly format. This approach eliminates the need for external indices or learning-based predictors, offering a lightweight yet robust solution for memory-constrained inference. All components are designed to be hardware-efficient and easy to implement. By implementing custom CUDA kernels, our method integrates seamlessly with FlashAttention, minimizing additional runtime and memory overhead. Experimental results demonstrate that our approach delivers both effectiveness and efficiency.

cs.LG↗
arXiv2026

Towards On-Policy SFT: Distribution Discriminant Theory and its Applications in LLM Training

Supervised fine-tuning (SFT) is computationally efficient but often yields inferior generalization compared to reinforcement learning (RL). This gap is primarily driven by RL's use of on-policy data. We propose a framework to bridge this chasm by enabling On-Policy SFT. We first present \textbf{\textit{Distribution Discriminant Theory (DDT)}}, which explains and quantifies the alignment between data and the model-induced distribution. Leveraging DDT, we introduce two complementary techniques: (i) \textbf{\textit{In-Distribution Finetuning (IDFT)}}, a loss-level method to enhance generalization ability of SFT, and (ii) \textbf{\textit{Hinted Decoding}}, a data-level technique that can re-align the training corpus to the model's distribution. Extensive experiments demonstrate that our framework achieves generalization performance surpassing prominent offline RL algorithms, including DPO and SimPO, while maintaining the efficiency of an SFT pipeline. The proposed framework thus offers a practical alternative in domains where RL is infeasible. We open-source the code here: https://github.com/zhangmiaosen2000/Towards-On-Policy-SFT

cs.LG↗