Search arXiv⌕ Search

arXiv subjects

Ming Yang

Publications and source records attributed to Ming Yang.

At least 73 records · Page 4Linked to original sources
arXiv2026

TMAS: Scaling Test-Time Compute via Multi-Agent Synergy

Test-time scaling has become an effective paradigm for improving the reasoning ability of large language models by allocating additional computation during inference. Recent structured approaches have further advanced this paradigm by organizing inference across multiple trajectories, refinement rounds, and verification-based feedback. However, existing structured test-time scaling methods either weakly coordinate parallel reasoning trajectories or rely on noisy historical information without explicitly deciding what should be retained and reused, limiting their ability to balance exploration and exploitation. In this work, we propose TMAS, a framework for scaling test-time compute via multi-agent synergy. TMAS organizes inference as a collaborative process among specialized agents, enabling structured information flow across agents, trajectories, and refinement iterations. To support effective cross-trajectory collaboration, TMAS introduces hierarchical memories: the experience bank reuses low-level reliable intermediate conclusions and local feedback, while the guideline bank records previously explored high-level strategies to steer subsequent rollouts away from redundant reasoning patterns. Furthermore, we design a hybrid reward reinforcement learning scheme tailored to TMAS, which jointly preserves basic reasoning capability, enhances experience utilization, and encourages exploration beyond previously attempted solution strategies. Extensive experiments on challenging reasoning benchmarks show that TMAS achieves stronger iterative scaling than existing test-time scaling baselines, with hybrid reward training further improving scaling effectiveness and stability across iterations. Code and data are available at https://github.com/IQuestLab/tmas.

cs.AI↗
arXiv2026

TravExplorer: Cross-Floor Embodied Exploration via Traversability-Aware 3-D Planning

Zero-shot Object Navigation (ZSON) has shown promise for open-vocabulary target search in unseen environments, yet most existing systems remain tied to planar representations and single-floor assumptions. These assumptions become inadequate in real buildings, where navigation involves floors, stairs, landings, and vertically overlapping spaces. This article presents TravExplorer, a cross-floor embodied exploration framework that couples zero-shot semantic guidance with traversability-aware 3-D planning. TravExplorer maintains a unified volumetric map that distinguishes occupied structures from robot-reachable support surfaces and extracts traversable frontiers from connected support surfaces, including floors, stairs, and landings. A FOV-aware active perception strategy further resolves incomplete observations during cross-floor traversal. To reduce semantic-reasoning latency, a lightweight guidance module aligns a probabilistic instance map from online open-vocabulary segmentation with a spatial value map from fast image-to-text matching. Based on these geometric and semantic memories, a hierarchical planner performs target-aware frontier touring over object hypotheses, traversable frontiers, and stair landmarks, and generates executable cross-floor motions through foothold-guided 3-D search and vertically constrained local trajectory optimization. Experiments over 4,195 simulated episodes on HM3D and MP3D demonstrate consistent advantages over representative ObjectNav baselines. Fifty real-world trials on a Unitree Go2 further validate open-vocabulary target search across single-floor and cross-floor indoor environments without prior maps or human intervention. The code will be released at https://github.com/wuyi2121/TravExplorer.

cs.RO↗
arXiv2026

Stellar Density Classification and Regression for CSST Multi-color Imaging Using Deep Learning

The Chinese Space Station Survey Telescope (CSST) aims to map the universe across an unprecedented dynamic range of stellar densities, spanning from extragalactic voids to the crowded Galactic center (e.g. a few stars and galaxies in the voids and $>10^5$ stars per detector in Galactic center). However, processing such heterogeneous data with a general source extraction pipeline introduces significant systematic uncertainties, standard algorithms exhibit poor accuracy in crowded fields and suffer from increased astrometric uncertainty in void regions. To mitigate these systematics, we propose a hierarchical, two-stage deep learning model for adaptive data reduction. The first stage ('classification') employs a ResNet-34 model to classify images into six discrete density categories, achieving $98.83\%$ in global accuracy. This classification acts as a critical decision gate, ensuring high calibration accuracy in the crowded fields. In the second stage ('regression'), a ResNet-50 regression model predicts the bright stars ($<23.5$ mag) in the field, which is essential for astrometric calibration, achieving a mean absolute error (MAE) of 0.0824 dex. By decoupling density characterization from source extraction, our model ensures that photometric and astrometric algorithms are optimally matched to the stellar density environment, thereby enhancing the fidelity and homogeneity of CSST as well as future large sky survey data products.

astro-ph.IM↗
arXiv2026

REAP: Reinforcement-Learning End-to-End Autonomous Parking with Gaussian Splatting Simulator for Real2Sim2Real Transfer

In recent years, autonomous parking has made significant advances, yet parking tasks still face challenges in extreme scenarios such as mechanical and dead-end parking slots, often resulting in failures. This is mainly due to traditional parking methods adopting a multistage approach, lacking the ability to optimize the parking problem as a whole. End-to-end methods enable joint optimization across perception and planning modules to eliminate the accumulation of errors, enhancing algorithm performance in extreme scenarios. Although several end-to-end parking methods use imitation or reinforcement learning, the former is limited by data cost and distribution coverage, while the latter suffers from inefficient exploration. To address these challenges, we propose a Reinforcement learning End-to-end Autonomous Parking method (REAP). REAP employs Soft Actor-Critic (SAC) within an asymmetric reinforcement learning framework to improve training efficiency and inference performance. To accelerate model convergence, we distill the capabilities of a rule-based planner into the end-to-end network through behavior cloning. We further introduce a soft predictive collision penalty mechanism to reduce collision rates by penalizing obstacle-approaching actions. To ensure that the trained reinforcement learning network can directly transfer to real-world scenarios, we have established a Real2Sim2Real simulator. In the Real2Sim step, we use 3D Gaussian Splatting (3DGS) to transform real-world scenes into digital scenes. In the Sim2Real step, we deploy the end-to-end model onto the vehicle to bridge the Sim2Real gap. Trained in the 3DGS simulator and deployed on physical vehicles, REAP successfully parks in various types of parking spaces, especially demonstrating the feasibility of end-to-end RL parking in extremely narrow mechanical slots.

cs.RO↗
arXiv2026

FAVOR: Efficient Filter-Agnostic Vector ANNS Based on Selectivity-Aware Exclusion Distances

Modern retrieval systems increasingly require integrating approximate nearest neighbor search (ANNS) with complex attribute filtering to handle hybrid queries in applications such as recommendation systems and retrieval-augmented generation (RAG). While HNSW-based inline-filtering methods show promise, existing approaches struggle to deliver high throughput under low-selectivity scenarios while balancing search efficiency, filtering generality, and index connectivity. To address these challenges, we propose FAVOR, an efficient filter-agnostic vector ANNS that supports arbitrary filtering conditions while maintaining stable performance across varying selectivity levels. FAVOR introduces three novel features: (1) an integrated architecture that unifies selectivity estimation and filtered ANNS execution, providing a cohesive solution for hybrid vector-attribute queries; (2) a HNSW-based inline-filtering algorithm that introduces an exclusion distance mechanism to dynamically reshape the vector distance distribution, pushing non-target vectors away from the query while promoting valid candidates toward the query, thus improving search efficiency without compromising generality or graph connectivity; and (3) a selectivity-driven search selector that estimates query selectivity and dynamically routes queries between a pre-filtering brute-force algorithm for low-selectivity cases and an optimized HNSW-based search algorithm for other scenarios, ensuring consistent performance. Extensive experiments on real-world datasets demonstrate that FAVOR achieves a 1.3-5$\times$ higher QPS at $Recall@10 = 95\%$ compared to state-of-the-art methods for arbitrary filtering conditions, while maintaining competitive performance even against tailored solutions in some filtering conditions.

cs.IR↗
arXiv2026

Generating Roadside LiDAR Datasets from Vehicle-Side Datasets via Novel View Synthesis

Intelligent Transportation Systems (ITS) require reliable environmental perception to support safe and efficient transportation. With the rapid development of Vehicle-to-everything (V2X), roadside perception has become an effective means to extend sensing coverage and improve traffic safety. However, the scarcity of large-scale annotated roadside LiDAR datasets poses a major challenge for training high-performance roadside perception models. In this paper, we introduce Vehicle-to-Roadside LiDAR Synthesis (VRS), a data synthesis framework that generates labeled roadside LiDAR datasets from vehicle-side datasets via LiDAR novel view synthesis. To mitigate the vehicle-to-roadside domain gap, VRS employs vehicle point cloud completion to compensate for missing geometry in vehicle-side observations, and introduces an occupancy-based visibility constraint to handle large viewpoint changes during cross-view rendering. The proposed framework enables flexible multi-view rendering for scalable roadside data generation. Extensive experiments on roadside 3D object detection demonstrate that the synthesized data effectively complements real roadside data, mitigates the limitations of limited real-world roadside data, and improves generalization to unseen roadside viewpoints.

cs.RO↗
arXiv2026

Field re-entrant superconductivity in Eu-doped infinite-layer nickelates

Intertwined superconducting and magnetic orders may give rise to exotic quantum phases, including field-induced and re-entrant superconductivity. However, such magnetism-enhanced superconductivity has remained elusive in superconductors with higher transition temperatures. While infinite-layer nickelates represent a new class of unconventional superconductors, the impact of rare-earth magnetism on superconducting properties remains largely unexplored. Here, we show that Eu-doped infinite-layer nickelate Sm$_{0.95-x}$Ca$_{0.05}$Eu$_x$NiO$_2$ exhibits a magnetic-field-induced re-entrant superconducting phase in the Eu-rich over-doped regime. Zero-resistance transport and high-field diamagnetic screening confirm the superconducting nature of this phase, which emerges after the initial suppression of low-field superconductivity and remains robust across a broad range of temperatures, fields and field orientations. In the same doping range, we observe nonlinear Hall transport and hysteretic magnetoresistance, indicating the unconventional nature of the re-entrant behaviour. While partially consistent with a compensation mechanism between the Eu-derived exchange field and the applied field, our data reveal pronounced deviations from this model at the highest-doping levels. Our findings establish infinite-layer nickelates as a fertile platform for exploring magnetically driven high-field superconductivity in strongly correlated oxides.

cond-mat.supr-con↗
arXiv2026

TabularMath: Understanding Math Reasoning over Tables with Large Language Models

Mathematical reasoning has long been a key benchmark for evaluating large language models. Although substantial progress has been made on math word problems, the need for reasoning over tabular data in real-world applications has been overlooked. For instance, applications such as business intelligence demand not only multi-step numerical reasoning with tables but also robustness to incomplete or inconsistent information. However, comprehensive evaluation in this area is severely limited, constrained by the reliance on manually collected tables that are difficult to scale and the lack of coverage for potential traps encountered in real-world scenarios. To address this problem, we propose AutoT2T, a neuro-symbolic framework that controllably transforms math word problems into scalable and verified tabular reasoning tasks. Building on this pipeline, we develop TabularMath, a benchmark comprising four subsets that include both text-based and image-based tables, covering table complexity, table quality, and table representation dimensions. Our study reveals three key observations: (1) Table complexity and reasoning difficulty impact reasoning performance jointly; (2) Low-quality tables pose severe risks to reliable reasoning in current LLMs; (3) Different table modalities show similar trends, with text-based tables typically being easier for models to reason over. In-depth analyses are conducted for each observation to guide future research.

cs.AI↗
arXiv2026

DiPO: Disentangled Perplexity Policy Optimization for Fine-grained Exploration-Exploitation Trade-Off

Reinforcement Learning with Verifiable Rewards (RLVR) has catalyzed significant advances in the reasoning capabilities of Large Language Models (LLMs). However, effectively managing the exploration and exploitation trade-off remains a critical challenge. In this paper, we fully analyze the exploration and exploitation dilemma of extremely hard and easy samples during the training and propose a new fine-grained trade-off mechanism. Concretely, we introduce a perplexity space disentangling strategy that divides the sample space into distinct exploration (high perplexity) and exploitation (low perplexity) subspaces, thereby mining fine-grained samples requiring exploration-exploitation trade-off. Subsequently, we propose a bidirectional reward allocation mechanism with a minimum impact on verification rewards to implement perplexity-guided exploration and exploitation, enabling more stable policy optimization. Finally, we have evaluated our method on two mainstream tasks: mathematical reasoning and function calling, and experimental results demonstrate the superiority of the proposed method, confirming its effectiveness in enhancing LLM performance by fine-grained exploration-exploitation trade-off.

cs.LG↗
arXiv2026

C$^2$T: Captioning-Structure and LLM-Aligned Common-Sense Reward Learning for Traffic--Vehicle Coordination

State-of-the-art (SOTA) urban traffic control increasingly employs Multi-Agent Reinforcement Learning (MARL) to coordinate Traffic Light Controllers (TLCs) and Connected Autonomous Vehicles (CAVs). However, the performance of these systems is fundamentally capped by their hand-crafted, myopic rewards (e.g., intersection pressure), which fail to capture high-level, human-centric goals like safety, flow stability, and comfort. To overcome this limitation, we introduce C2T, a novel framework that learns a common-sense coordination model from traffic-vehicle dynamics. C2T distills "common-sense" knowledge from a Large Language Model (LLM) into a learned intrinsic reward function. This new reward is then used to guide the coordination policy of a cooperative multi-intersection TLC MARL system on CityFlow-based multi-intersection benchmarks. Our framework significantly outperforms strong MARL baselines in traffic efficiency, safety, and an energy-related proxy. We further highlight C2T's flexibility in principle, allowing distinct "efficiency-focused" versus "safety-focused" policies by modifying the LLM prompt.

cs.MA↗
arXiv2026

LAST: Leveraging Tools as Hints to Enhance Spatial Reasoning for Multimodal Large Language Models

Spatial reasoning is a cornerstone capability for intelligent systems to perceive and interact with the physical world. However, multimodal large language models (MLLMs) frequently suffer from hallucinations and imprecision when parsing complex geometric layouts. As data-driven scaling struggles to internalize structured geometric priors and spatial constraints, integrating mature, specialized vision models presents a compelling alternative. Despite its promise, applying this paradigm to spatial reasoning is hindered by two key challenges: The difficulty of invoking heterogeneous, parameter-rich tools, as well as the challenge of understanding and effectively leveraging their diverse low-level outputs (e.g., segmentation masks, depth maps) in high-level reasoning. To address these challenges, we propose LAST, a unified framework for tool-augmented spatial reasoning. LAST features an extensible interactive sandbox, termed LAST-Box, which abstracts heterogeneous tool invocations into atomic instructions and reusable spatial skills, returning multimodal hints (e.g., annotated images and textual descriptions) that can be directly consumed by LLMs. We further design a three-stage progressive training strategy that guides models from understanding tool outputs to proficient and adaptive tool invocation. Experiments on four datasets show that LAST-7B achieves around 20\% performance gains over its backbone and outperforms strong proprietary closed-source LLMs, substantially enhancing reasoning on complex spatial tasks.

cs.CV↗
arXiv2026

A Physics-Informed Chemical Rule for Topological Materials Discovery

Topological phases of matter$\unicode{x2013}$comprising both insulators and semimetals$\unicode{x2013}$offer great potential for quantum applications, but identifying new candidates remains challenging due to expensive first-principles simulations and labor-intensive experimental workflows. Here we introduce a physics-informed chemical rule that integrates compositional, orbital and crystallographic descriptors within an interpretable linear framework. By explicitly encoding electron filling, space-group symmetry and orbital-resolved chemical environments, our method overcomes a fundamental limitation of composition-only heuristics$\unicode{x2013}$their inability to distinguish polymorphs with identical stoichiometry but different crystal structures. Using only elemental characteristics, our approach reduces a material's topological propensity to a single, physically interpretable score, enabling rapid and high-throughput assessment. The model achieves superior predictive performance while maintaining physical transparency, and identifies candidate topological materials where conventional symmetry indicators fail. Consequently, our framework enables rapid and interpretable exploration of complex materials spaces, establishing a scalable paradigm for the intelligent discovery of next-generation topological and quantum materials.

cond-mat.mtrl-sci↗
arXiv2026

TSPO: Breaking the Double Homogenization Dilemma in Multi-turn Search Policy Optimization

Multi-turn tool-integrated reasoning enables Large Language Models (LLMs) to solve complex tasks through iterative information retrieval. However, current reinforcement learning (RL) frameworks for search-augmented reasoning predominantly rely on sparse outcome-level rewards, leading to a "Double Homogenization Dilemma." This manifests as (1) Process homogenization, where the thinking, reasoning, and tooling involved in generation are ignored. (2) Intra-group homogenization, coarse-grained outcome rewards often lead to inefficiencies in intra-group advantage estimation with methods like Group Relative Policy Optimization (GRPO) during sampling. To address this, we propose Turn-level Stage-aware Policy Optimization (TSPO). TSPO introduces the First-Occurrence Latent Reward (FOLR) mechanism, allocating partial rewards to the step where the ground-truth answer first appears, thereby preserving process-level signals and increasing reward variance within groups without requiring external reward models or any annotations. Extensive experiments demonstrate that TSPO significantly outperforms state-of-the-art baselines, achieving average performance gains of 24% and 13.6% on Qwen2.5-3B and 7B models, respectively. Code is available at https://github.com/Flipped-May/TSPO.

cs.AI↗
arXiv2026

DeepSlide: From Artifacts to Presentation Delivery

Presentations are a primary medium for scholarly communication, yet most AI slide generators optimize the artifact (a visually plausible deck) while under-optimizing the delivery process (pacing, narrative, and presentation preparation). We present DeepSlide, a human-in-the-loop multi-agent system that supports preparing the full presentation process, from requirement elicitation and time-budgeted narrative planning, to evidence-grounded slide--script generation, attention augmentation, and rehearsal support. DeepSlide integrates (i) a controllable logical-chain planner with per-node time budgets, (ii) a lightweight content-tree retriever for grounding, (iii) Markov-style sequential rendering with style inheritance, and (iv) sandboxed execution with minimal repair to ensure renderability. We further introduce a dual-scoreboard benchmark that cleanly separates static artifact quality from dynamic delivery excellence. Across 20 domains and diverse audience profiles, DeepSlide matches strong baselines on artifact quality while consistently achieving larger gains on delivery metrics, improving narrative flow, pacing precision, and slide--script synergy with clearer attention guidance.

cs.AI↗
arXiv2026

AMFD: Distillation via Adaptive Multimodal Fusion for Multispectral Pedestrian Detection

Multispectral pedestrian detection has been shown to be effective in improving performance within complex illumination scenarios. However, prevalent double-stream networks in multispectral detection employ two separate feature extraction branches for multi-modal data, leading to nearly double the inference time compared to single-stream networks utilizing only one feature extraction branch. This increased inference time has hindered the widespread employment of multispectral pedestrian detection in embedded devices for autonomous systems. To address this limitation, various knowledge distillation methods have been proposed. However, traditional distillation methods focus only on the fusion features and ignore the large amount of information in the original multi-modal features, thereby restricting the student network's performance. To tackle the challenge, we introduce the Adaptive Modal Fusion Distillation (AMFD) framework, which can fully utilize the original modal features of the teacher network. Specifically, a Modal Extraction Alignment (MEA) module is utilized to derive learning weights for student networks, integrating focal and global attention mechanisms. This methodology enables the student network to acquire optimal fusion strategies independent from that of teacher network without necessitating an additional feature fusion module. Furthermore, we present the SMOD dataset, a well-aligned challenging multispectral dataset for detection. Extensive experiments on the challenging KAIST, LLVIP and SMOD datasets are conducted to validate the effectiveness of AMFD. The results demonstrate that our method outperforms existing state-of-the-art methods in both reducing log-average Miss Rate and improving mean Average Precision. The code is available at https://github.com/bigD233/AMFD.git.

cs.CV↗
arXiv2026

Flare-driven habitability: Expanding life's potential around low-mass stars

The traditional definition of the circumstellar habitable zone (HZ) focuses on liquid water, but neglects the crucial role of ultraviolet (UV) radiation in prebiotic chemistry. Low-mass stars typically emit insufficient UV radiation for photochemistry throughout the liquid water HZs during quiescent states. However, frequent flares can provide substantial UV fluxes, potentially fostering habitable conditions. We refine the concept of a UV habitable zone (UV-HZ) by incorporating a temperature-dependent model for RNA precursor synthesis. Furthermore, we explore a parameterized spectral energy distribution model and adopt an empirical flare frequency distribution for flares on different stars to quantify their UV contribution. Applying this framework to different flaring stars, we find the UV-HZ around low-mass stars can extend to inner regions, and overlap with the traditional HZ in wide ranges. Apply the analysis to 9 planets around Kepler flaring stars, three planets are located within both the refined UV-HZ and liquid water habitable zone (LW-HZ) without causing ozone depletion. Our findings highlight the significant role of flares in expanding the potential for life around low-mass stars, offering a revised perspective on exoplanet habitability criteria.

astro-ph.EP↗
arXiv2026

Thinking with Tables: Enhancing Multi-Modal Tabular Understanding via Neuro-Symbolic Reasoning

Multimodal Large Language Models (MLLMs) have demonstrated remarkable reasoning capabilities across modalities such as images and text. However, tabular data, despite being a critical real-world modality, remains relatively underexplored in multimodal learning. In this paper, we focus on the task of Tabular-Vision Multi-Modal Understanding (TVMU) and identify three core challenges: (1) high structural variability and data incompleteness in tables, (2) implicit and complex feature dependencies, and (3) significant heterogeneity in problem-solving pipelines across downstream tasks. To address these issues, we propose Thinking with Tables (TWT). TWT employs a program-aided code-based neuro-symbolic reasoning mechanism that facilitates key operations, such as information extraction and element modeling, by interacting with external environments. We evaluate TWT on eight representative datasets. Experimental results demonstrate that TWT consistently outperforms existing baselines by an average of 10\% in accuracy, achieving performance comparable to, or even surpassing, proprietary commercial SOTA LLMs on TVMU tasks. Models and codes are available at https://github.com/kunyang-YU/Thinking-with-Tables

cs.CL↗
arXiv2026

Towards Effective Experiential Learning: Dual Guidance for Utilization and Internalization

Recently, reinforcement learning~(RL) has become an important approach for improving the capabilities of large language models~(LLMs). In particular, reinforcement learning from verifiable rewards~(RLVR) has emerged as a promising paradigm for reasoning tasks. However, existing RL-based training still remains only a rough approximation to human learning. Human learners leverage both external and internal experience to guide exploration and gradually internalize useful trajectories into stable knowledge. Motivated by this gap, we ask: how can LLMs better utilize and internalize experience during RLVR training? To answer this question, we propose \textbf{D}ual \textbf{G}uidance \textbf{O}ptimization~(\textbf{DGO}), a unified framework that leverages \emph{external} and \emph{internal experience} to improve training effectiveness. Specifically, DGO first constructs an experience bank from previously explored trajectories. The policy then performs exploration under the joint guidance of the experience bank and the model's internal knowledge. The resulting trajectories are further used to refine the experience bank and optimize model parameters, forming a closed loop of experience utilization and internalization. Experiments show that DGO consistently outperforms baseline methods, suggesting that better utilization and internalization of experience lead to more effective reasoning.

cs.LG↗