Search arXiv⌕ Search

arXiv subjects

Pei Liu

Publications and source records attributed to Pei Liu.

At least 19 recordsLinked to original sources
arXiv2026

The maximum spectral radius of outerplanar and planar $k$-uniform hypergraphs

For $k\ge3$, a $k$-angulation is a simple $2$-connected outerplane graph whose interior faces are bounded by $k$-cycles, and a closed $k$-angulation is a simple $2$-connected plane graph all of whose faces are bounded by $k$-cycles. Their face hypergraphs have the same vertices as the graphs; the hyperedges are the vertex sets of the interior faces, respectively all faces. For $k=3$, the extremal problems over these hypergraphs agree with those for the outerplanar and planar $3$-uniform hypergraphs of Ellingham, Lu and Wang, who determined the outerplanar maximizer for large $n$ and conjectured the planar one. In this paper, for each fixed $k$, we determine the hypergraphs of maximum spectral radius in both classes for all sufficiently large $n$ satisfying the corresponding face-count divisibility condition. In the outerplanar case, the unique extremal hypergraph is the face hypergraph of the fan, in which one vertex lies on every interior face. In the planar case with $k=3$, it is the face hypergraph of $K_2+P_{n-2}$, which confirms the conjecture of Ellingham, Lu and Wang. For $k\ge4$, the extremal hypergraphs are the face hypergraphs of balanced theta graphs, in which two vertices lie on every face. For $k=4$, the result holds for every $n\ge5$, with extremal hypergraph ${\mathcal H}(K_{2,n-2})$. For $k\ge6$ and an even number of faces, there are exactly $\lfloor(k-2)/2\rfloor$ extremal hypergraphs up to isomorphism.

math.CO↗
arXiv2026

The More the Better? A Comparison of Single- and Multi-Agent LLM Setups on a Financial Review Task

Practitioner guidance holds that a large language model reviews a document better when the work is divided among agents. We test a narrow form of it in a controlled comparison of single-agent and multi-agent setups on a financial task, reviewing credit assessment memos against requirements. One agent checks six requirements in one call. A team of three on the same model checks the same memo, two requirements each, in the same words, with no roles or later stage. Labels are known for every requirement on 300 synthetic memos, so misses and false alarms are measured separately. Two primary predictions are tested on four models at a Bonferroni-corrected level. Under an instruction to decide each requirement once, the team catches 0.08 to 0.10 more of the defects that carry an explanation on the three models whose single agent misses some, at the corrected level on two. It raises more false alarms on all four, at the corrected level on three. Balanced accuracy, an exploratory measure, changes by 0.02 or less, and the team uses 2.2 to 2.4 times the tokens. Showing each agent the requirements it was not given reduces the added false alarms on the two Claude models, at the corrected level on one. In exploratory comparisons, sampling the single agent three times and flagging on any draw gives higher balanced accuracy than the team on those models, and the team's results depend on the instruction.

cs.AI↗
arXiv2026

Omni-RS: Unifying Multi-Task Understanding and Generation in Remote Sensing

Practical remote sensing applications often require both understanding and generation capabilities for the same geographic scene. However, most existing methods address either understanding or generation in isolation, while existing unified remote sensing models' generative ability is limited to text-to-image. To bridge this gap, we propose Omni-RS, a unified framework supporting four remote sensing understanding tasks and five generation tasks. Specifically, we introduce the "Scene-Guided Two-Stage Training" strategy, which first adapts a multimodal large language model to remote sensing understanding and then uses its hidden states to guide a diffusion transformer for image generation. To enable joint learning across diverse generation tasks, we construct UniGen, a unified multi-task generation dataset, by organizing heterogeneous remote sensing data into a unified format. Furthermore, we introduce "Detailed Visual Conditioning", which uses visual conditions encoded by a VAE to guide the generation process, thereby supporting additional generation tasks that require reference images. Extensive experiments demonstrate that Omni-RS performs exceptionally well in understanding tasks and has achieved outstanding results in a variety of generative tasks.

cs.CV↗
arXiv2026

DA-WAM: Decision-Aligned Future Latents for Driving World Models

World models for autonomous driving should anticipate the consequences that matter for driving decisions rather than reconstruct every detail of future scenes. However, predictive representations alone are not guaranteed to distinguish candidate actions that lead to different planning outcomes.We propose DA-WAM, a decision-aligned world action model that couples a Joint-Embedding Predictive Architecture (JEPA) with candidate-level trajectory scoring. During joint training, the JEPA objective grounds future representations in observed scene dynamics, while the planning objectives encourage these representations to capture distinctions relevant to safety and utility. An action-conditioned predictor generates a distinct future latent for each candidate trajectory, and a multi-head trajectory scorer uses the corresponding latent to estimate planning factors and overall utility. In addition, safety-critical post-training with hard negatives further improves discrimination between geometrically similar trajectories with different safety outcomes. DA-WAM achieves state-of-the-art performance among learned camera-based planners on NAVSIM-v1, NAVSIM-v2, and Bench2Drive. Ablation studies demonstrate the effectiveness of each component in driving decision-making.

cs.RO↗
arXiv2026

Panoramic modelling of single-cell transcriptomes at scale with CellNet

Comprehensively characterizing cellular diversity is critical for understanding tissue development, identifying disease-associated populations and investigating gene programmes linked to specialized functions. Advances in single-cell profiling reveal additional cell-specific transcripts and molecular associations, yet many foundation models represent only a restricted portion of this increasingly rich molecular information. Here we present CellNet, which enables panoramic modelling of single-cell transcriptomes by integrating DNA-derived gene identities and measured RNA expression in a shared image representation. Pretrained on 231 million cell profiles spanning 40 species, CellNet improves rare-population and condition-associated state recognition in evaluated cohorts while retaining competitive general cell representations. It predicts withheld coding-gene expression and selected non-coding targets in total-RNA data, and recovers lineage-specific expression patterns in spatial measurements. Prediction also extends across species, including to targets without expression-pretraining exposure. Its gene representations retrieve known non-coding modules and developmental or neural programmes, including from sparsely detected RNA queries. Together, these results establish CellNet as a scalable framework for interpreting cellular diversity through increasingly comprehensive transcriptomic measurements.

q-bio.GN↗
arXiv2026

Understanding Gene Expression-to-Image Learning: Spatial Structure and Information Fidelity

Single-cell RNA sequencing (scRNA-seq) enables cell-level analysis of biological heterogeneity, but learning effective representations from high-dimensional, unordered gene expression profiles remains challenging. Gene Expression-to-Image (GE2I) offers an effective learning strategy by arranging gene values on pixel grids for convolutional neural networks, but its underlying mechanisms remain unclear. We study GE2I on full-transcriptome data while keeping the network and training procedure fixed. Our results reveal a trade-off between spatial smoothness and gene-level fidelity. Structured mappings produce smooth, low-frequency patterns that support faster early optimization, whereas assigning multiple genes to the same pixel obscures expression differences and can reduce predictive performance. This trade-off explains why random mappings can compete with structured layouts when many genes are represented. One-to-one gene-to-pixel assignment improves prediction across the tested visual backbones even when image smoothness decreases. We further find that the benefits of pretraining and augmentation depend on the backbone and mapping. These findings provide practical guidance for pixel allocation and the transfer of vision priors to gene expression data.

cs.AI↗
arXiv2026

LeapBot-JAC: Joint Action Chains for World Action Models

Action-chunking policies predict future controls but execute only a prefix before replanning. How should a fixed action-token budget be organized to support execution? We propose LeapBot-JAC, which allocates sparse future-control references and dense execution targets within one shared action expert. Both chains are supervised by demonstrated controls and jointly denoised at aligned physical-time positions. Directed attention allows execution tokens to read reference tokens subject to the host's visibility constraints. JAC adds neither action tokens nor parameters and retains the host's world-modeling process; both chains participate in inference, while only the execution chain drives the robot. On LIBERO, JAC reaches 98.3% success with FastWAM and LeapBot-WA, up from 97.0% and 97.3%. On zero-shot LIBERO-Plus, their success rates improve from 53.5% to 70.4% and from 73.1% to 77.3%, respectively. RoboTwin and two real-world tasks provide additional deployment evaluations. Single-training-seed comparisons and FastWAM ablations support action-role organization as a practical design choice under a fixed action-token budget.

cs.RO↗
arXiv2026

Recon2Servo: Robotic Ultrasound Visual Servoing via Learned Image-to-Motion Inference

Ultrasound visual servoing is essential for autonomous robotic ultrasound, yet 6-DoF probe control from 2D B-mode images remains challenging due to limited and ambiguous out-of-plane motion cues. Existing methods typically rely on anatomical priors or handcrafted visual features, limiting their generalizability across imaging targets. Inspired by trackerless 3D ultrasound reconstruction, we propose Recon2Servo, a visual servoing framework that learns image-to-motion inference directly from B-mode images for 6-DoF probe control. A DINOv3 encoder with low-rank adaptation and a bidirectional relation module estimate the relative probe pose between current and target images to guide iterative closed-loop target-view alignment. The framework combines supervised relative-pose learning, reconstruction-guided closed-loop adaptation, and bounded residual pose correction to improve motion inference during servoing. Evaluations on a public dataset and an in-house dataset collected from 12 healthy volunteers using different ultrasound systems demonstrate its effectiveness in reconstructed-volume servoing. Additional real-robot demonstrations of target-view alignment and dynamic tracking on a human forearm are provided in the supplementary video: https://youtu.be/qwsOdI-GMYk.

cs.RO↗
arXiv2026

MagServo: Uncertainty-Resilient Hierarchical Magnetic Servoing via Learned Latent Representations

Magnetic navigation provides contact-free and line-of-sight-independent feedback for robotic systems, yet existing approaches typically rely on explicit pose estimation or direct use of raw magnetic measurements, making accurate control susceptible to modeling errors, measurement noise, and disturbances. This work presents MagServo, a hierarchical learning-based framework for robust 6-DoF magnetic servoing directly using the learned latent magnetic feature. MagServo learns uncertainty-resilient magnetic representations through masked reconstruction and captures state-dependent interaction dynamics between robot motion and latent magnetic transitions without analytical magnetic models or explicit Jacobian supervision. Based on the learned dynamics, a hierarchical controller combines nonlinear model predictive control for coarse approach with local Jacobian inversion for precise fine regulation. Extensive physical experiments demonstrate submillimeter and subdegree accuracy, achieving mean terminal errors of 0.386 mm and 0.479 degree for 6-DoF pose reaching. MagServo further outperforms a localization-based control baseline in complex trajectory tracking and maintains robust performance under unseen magnetic-source configurations without retraining. A supplementary video of the real-robot experiments is available at https://youtu.be/rZt1NUP1Mr0.

cs.RO↗
arXiv2026

CoEvolve: Construct-to-Edit Visual Grounding with Bidirectional State Refinement

Visual grounding localizes an object described by language with a bounding box. Most multimodal grounding models compress target identification, spatial reasoning, and boundary estimation into one terminal prediction. Free-form rationales make reasoning linguistically explicit but do not necessarily expose measurable, editable spatial states. Intermediate localization errors are therefore difficult to diagnose and correct, allowing incorrect region choices and imprecise boundaries to persist in the final box. We introduce CoEvolve, a construct-to-edit framework that separates grounding into explicit state construction and state editing. Region-Evolution Reinforcement (RER) organizes grounding analysis into a progressive semantic--spatial trajectory, with each reasoning step committing to an explicit candidate region. Bidirectional Denoising Refiner (BDR) treats the reasoning text as fixed semantic context and refines the trajectory's coordinate fields through bidirectional same-position reconstruction. Geometry- and behavior-level objectives provide target geometry and edit-preference signals for consolidating reliable candidates, preserving accurate inputs, or correcting toward annotations. Evaluations cover natural-image and remote-sensing grounding. With a 9B backbone, CoEvolve rivals models up to 241B parameters in grounding accuracy. Under controlled corruption, a single BDR pass improves mean box overlap by over 27 percentage points, demonstrating strong recovery from substantial localization errors. State-source comparisons further support the complementarity of explicit state construction and source-matched editing. The project is at https://sundongwei.github.io/CoEvolve_Project/.

cs.AI↗
arXiv2026

Nonvolatile optical switching of surface metallicity in 1T-TaSe2

The surface of a quantum material provides a distinct electronic environment in which reconstruction can stabilize states absent from the bulk. However, whether light can persistently reconfigure this surface-bulk electronic contrast remains unclear. Here we demonstrate nonvolatile and reversible optical insulator-to-metal switching at the surface of the charge density wave (CDW) material 1T-TaSe2, whose bulk is metallic. Angle-resolved photoemission spectroscopy reveals that a single femtosecond pulse creates a persistent metallic surface state. Heating or optical pulse trains restore the insulating state. Comparison of the measured band structures with first-principles calculations associates the two electronic states with distinct near-surface CDW stacking configurations that modify interlayer-mediated electron hopping. Layer-resolved calculations further show insulating outer layers above a metallic interior in the equilibrium configuration, whereas metallic states reach the outermost layer in the configuration associated with the photoinduced state. Our results establish persistent optical control of surface metallicity and open a route to reconfiguring the effective surface-bulk electronic boundary in layered quantum materials.

cond-mat.str-el↗
arXiv2026

ReWAM: Reciprocal World Action Models for Interactive Autonomous Driving

In interactive scenarios, an autonomous driving system is required to generate ego actions under the influence of other agents' behaviors. Existing World Action Models (WAMs) typically model other agents as components of the world model rather than as decision-makers that fundamentally shape the action of the ego agent, which impairs their performance in dense interaction scenarios. We introduce Reciprocal World Action Models (ReWAM), a game-theoretic world action modeling framework that captures the reciprocal influence between the ego agent and other agents by representing them as conditional responders whose actions are mutually influenced. We instantiate this framework with a Level-$k$ response hierarchy, where role-specific ego and other action DiTs exchange compact strategy tokens through cross-agent attention while remaining grounded in a shared representation of the future driving world. To learn the response policy of the ego agent from demonstrations, we formulate expert actions as samples from the best response distribution and jointly optimize the entire hierarchy using conditional flow matching. Our framework is evaluated on the NAVSIM dataset and achieves state-of-the-art performance compared to baselines. The improvement is particularly significant in interactive scenarios, validating that modeling reciprocal responses provides a more effective foundation for interaction-aware world action generation.

cs.RO↗
arXiv2026

Orbital-engineered px,y-kagome lattice in a halogen monolayer

Multi-orbital kagome lattices with explicit orbital degrees of freedom remain largely unexplored, as most experimentally realized systems rely on complex d-electron manifolds that are approximated by isotropic single-orbital models. Here, we overcome this limitation by realizing a px,y-orbital kagome lattice through deposition of a Br monolayer on Ag(111), where orbital filtering selectively suppresses the pz channel. Scanning tunneling microscopy, angle-resolved photoemission spectroscopy, and density-functional-theory calculations reveal a large-area, highly ordered kagome structure whose band dispersions quantitatively match the anisotropic px,y tight-binding model. To extract the intrinsic manifold from the substrate background, we construct an effective H-passivated model, which uncover the intrinsic electronic structure and reveals nontrivial topological characteristics of the px,y kagome manifold driven by first-order spin-orbit coupling effect. Our work establishes Br/Ag(111) as an experimentally accessible platform for multi-orbital kagome physics, extending the kagome paradigm from the conventional d-orbital regime to an orbitally engineered topological setting.

cond-mat.mtrl-sci↗
arXiv2026

Where Do Embodied Decisions Come From? Rethinking Latent and Explicit Reasoning

Chain-of-Thought (CoT) reasoning is increasingly incorporated into Vision-Language-Action (VLA) models, yet it can degrade the performance of stronger embodied agents. We investigate this capability-dependent effect by distinguishing explicit reasoning from latent decision computation, i.e., perception-grounded computation that directly supports action prediction. Under the standard \textit{think-then-act} (TTA) paradigm, intervening on the generated CoT while fixing the visual input and model parameters causes a substantial performance collapse, with Navigation F1 dropping from 72.14% to 11.84%, demonstrating the strong influence of explicit reasoning on action generation. We then propose \textit{decide-then-explain} (DTE), which predicts actions before generating explanations, and introduce Visual Conditional Contribution (VCC) and Reasoning Conditional Contribution (RCC) to characterize the resulting decision process. Across autonomous driving and robotic manipulation, DTE consistently outperforms TTA and conventional \textit{no-CoT} baselines, while exhibiting greater reliance on perception-grounded computation. Further TTA-trained, DTE-inference experiments show that this benefit is not solely attributable to retraining under the new factorization. Our results suggest that for capable embodied agents, explicit CoT may be better used to shape decision computation during training rather than mediate action generation at inference time. Code: https://github.com/ocean-luna/openvla-decide-then-explain.

cs.CE↗
arXiv2026

GeoForward: Geometry-Guided Cross-View Alignment for Feedforward 3D Gaussian Splatting

Feedforward 3D reconstruction has largely followed an {appearance-driven paradigm}: multi-view images are encoded into visual representations, cross-view correspondence is implicitly inferred from learned features, and 3D primitives are subsequently predicted from these representations. However, calibrated cameras and sparse metric measurements already provide explicit constraints on where the same physical point should appear across views. We propose a novel paradigm, \textbf{Geometry-Conditioned Feedforward Reconstruction}, in which metric geometry establishes cross-view correspondence before the 3D representation is formed, while learned visual features refine rather than discover this correspondence. Based on this paradigm, we introduce \textbf{GeoForward}. Sparse LiDAR points are unprojected into 3D and analytically projected across calibrated cameras to provide metric correspondence anchors. A bounded continuous local search then refines these anchors using visual features, and the resulting geometry-conditioned representation is fused with global multi-view context from VGGT. The shared representation drives separate metric-depth and Gaussian-attribute prediction, establishing a direct path from explicit geometry to both Gaussian centers and spatial attributes. A sparse 3D refinement module subsequently improves local attribute compatibility while preserving the geometrically determined centers. For temporal reconstruction, endpoint Gaussians are composed using vehicle masks and motion priors. On nuScenes, GeoForward achieves the lowest cross-camera depth-consistency error and Chamfer distance among the compared methods while maintaining competitive reconstruction quality. These results demonstrate that explicitly conditioning correspondence formation on metric geometry provides an effective alternative to purely appearance-driven feedforward 3D reconstruction.

cs.CV↗
arXiv2026

From Scene Graphs to Answers: Selective Neuro-Symbolic Reasoning for Autonomous Driving

Autonomous-driving question answering requires reasoning over structured scene information, yet existing vision-language approaches largely delegate heterogeneous reasoning operations to a single neural inference process. We argue that this uniform strategy overlooks a fundamental distinction: some queries admit exact symbolic solutions, while others require semantic interpretation. We introduce a query-adaptive neuro-symbolic reasoning framework that explicitly allocates computation according to the nature of the query. At its core is a hierarchical Spatiotemporal Scene Graph (STSG) that separates persistent object identities from frame-specific states and represents spatial relations and temporal transitions as explicit directed structures. Given a query, a symbolic executor first attempts to resolve it through exact graph operations; only when symbolic execution abstains is an LLM invoked for semantic reasoning. For these unresolved queries, query-conditioned graph retrieval and evidence filtering preserve relation direction, temporal locality, and object semantics, providing the LLM with compact and verified task-relevant evidence. This design shifts the role of the LLM from a universal reasoning engine to a targeted semantic reasoner, while allowing deterministic computation to be handled exactly and efficiently. We evaluate the framework on 5,916 NuScenes-QA questions across all ten scenes of nuScenes v1.0-mini under an oracle-perception setting. The complete system achieves 80.63 percent overall accuracy with GPT-5.4-mini, improving over the corresponding LLM-only configuration by 5.48 percentage points; with DeepSeek-V4-Flash, the improvement reaches 6.64 points. The largest gains occur on counting questions, with improvements of 10.20 and 12.61 points, respectively. These results show that selective reasoning improves both accuracy and inference efficiency.

cs.AI↗
arXiv2026

Semantic-Anchored Evidential Fusion for Domain-Robust Whole-Slide Survival Analysis

Whole-slide images (WSIs) are widely used for computational cancer prognosis. However, most existing methods primarily focus on in-domain performance and fail to generalize across clinical centers. This limitation stems from their reliance on pixel-derived representations that are highly susceptible to domain-specific artifacts caused by staining protocols and scanner hardware. We hypothesize that high-level pathology semantics, such as tumor grade and micro-environmental architecture, provide a domain-invariant semantic representation that mirrors the robust diagnostic logic of human pathologists. Therefore, we propose a Semantic-Anchored Evidential Fusion Survival (SAEFS) framework, where SAEFS derives semantic anchors from WSIs via Visual Question Answering (VQA), employs a dual-stream WSI evidence extraction architecture, uses Dirichlet-based Subjective Logic to model uncertainty, and fuses semantic and visual evidence through a cautious conjunction rule to avoid overconfident fusion from correlated sources. Trained exclusively on one source domain and evaluated zero-shot across four unseen domains, SAEFS consistently outperforms state-of-the-art models both in prediction accuracy and reliability, improving the average C-index by 10.2%. Quantitative analyses further show that VQA-derived semantic features exhibit significantly lower cross-center divergence than pixel-derived features, highlighting their robustness for cross-center clinical applications.

cs.CV↗
arXiv2026

Exact local spectral thresholds for perfect matchings in $3$-graphs and $3$-partite $3$-graphs

For a $3$-uniform hypergraph $H$, let $σ(H)$ be the minimum over the vertices of $H$ of the spectral radius of the link. Lin, Lu, Yuan and Zhao conjectured that $σ(H)>\tfrac{2n}3-2$ forces a perfect matching in a $3$-graph of order $n$ divisible by three, and Lu and Yuan conjectured that $σ(H)>τ(q)$ forces one in a $q$-balanced $3$-partite $3$-graph, where $τ(q)=\sqrt{q(q-1)/2}$ for odd $q$ and $τ(q)$ is given by a quartic for even $q$. In this paper, we prove both perfect matching conjectures for large order. Each proof passes through a stability theorem for fractional matchings: a fractional vertex cover of deficient weight with a zero coordinate is obtained, the link of that vertex carries the induced cover, and the spectral radius of such a link is bounded by an inequality whose extremal cases identify the corresponding space barriers.

math.CO↗