Search arXiv⌕ Search

arXiv subjects

Bin Shen

Publications and source records attributed to Bin Shen.

At least 19 recordsLinked to original sources

Parabolic frequency and backward uniqueness on compact Finsler metric measure spaces

We establish monotonicity formulas for parabolic frequency functions on closed Finsler metric measure spaces with fixed or evolving metrics and measures. When the reference measure is fixed, an integral bound on the metric variation yields monotonicity for the Finsler heat equation. For coupled metric measure flows, we construct positive global supersolutions and subsolutions of a linearized conjugate heat equation from local radial barriers. These weights define a frequency function whose monotonicity follows from an integral bound involving the metric variation and the change in weighted Ricci curvature. For parabolic equations with a source controlled by the solution and its gradient, we obtain a differential inequality for the frequency and prove that a solution vanishing at the terminal time must vanish throughout the preceding interval. The proof controls the comparison of the two weights through an energy quotient computed with the subsolution measure.

math.DG↗

Strong unique continuation on Riemannian manifolds for degenerate elliptic operators with convexified weights

In this manuscript, we consider a degenerate elliptic equation on a Riemannian manifold whose weight vanishes at an interior point $p$, and prove the strong unique continuation property (SUCP) at $p$. The curved setting gives rise to intrinsic geometric error terms that obstruct the analytic estimates available in the Euclidean case. To overcome this, we introduce a convexified radial weight that restores coercivity in the Carleman commutator. Combining the resulting Carleman estimate with a Hardy-type inequality and a Caccioppoli inequality, we show that any weak solution vanishing to infinite order at the degeneracy point must be identically zero on the connected domain. This result is the first SUCP theorem for this class of degenerate operators on general Riemannian manifolds.

math.DG↗

Local Aronson-Bénilan estimates for the porous medium and fast diffusion equations on collapsing manifolds

We establish local Aronson-Bénilan estimates for positive solutions to the porous medium and fast diffusion equations on complete noncompact Riemannian manifolds with a small scale-invariant integral bound on the negative part of the Ricci curvature in $L^p$, where $p>n/2$. No uniform lower bound on the volume of unit balls is assumed, and hence the results apply to manifolds that may collapse. For the fast diffusion equation, we obtain the estimate in the range $1-2/n<m<1$. The auxiliary function method, which was designed for the heat equation under integral curvature conditions, is not directly applicable to those two nonlinear equations. The pressure operator in both cases contains a solution-dependent coefficient, which produces a residual term involving $\partial_t J$ in the evolution inequality. Its treatment requires a positive lower bound for $J$, and moreover, control of $-\partial_t\log J$. We obtain these bounds by linearizing the auxiliary equation, applying Duhamel's formula, and combining local Dirichlet heat kernel estimates with a singular Volterra inequality. Additionally, in the fast diffusion case, the negativity of $v$ changes the sign structure of the evolution inequality, so a separate coercive estimate is required to close the argument. Local Aronson-Bénilan estimates follow for both diffusion regimes and integrating them along space-time curves gives the corresponding Harnack inequalities.

math.DG↗

Fixed geodesics of involutions on $3$-spheres

Let an isometric involution act on a smooth Riemannian $3$-sphere with sectional curvature at least one. We prove that every one-dimensional connected component of its fixed-point set has length at most $2π$, and that the bound is sharp. The proof realizes the fixed circle as the link of a complete totally geodesic surface in a positively curved expanding Ricci soliton and applies Bishop--Gromov comparison to its area growth. The analytic input is an equivariant version of Deruelle's smoothing of positively curved cones. We use Gianniotis--Schulze's equivariant local deformation and explain a marked compactness argument for arbitrary finite isotropy. In particular, the end chart retains the specified group action, rather than only an unspecified conjugate.

math.DG↗

Joining Large String Collections with Jaro-Winkler Similarity

The Jaro-Winkler measure, originally developed for U.S.\ Census record linkage, captures similarities between strings and accounts for common typographical variations. Despite its widespread use, finding all string pairs that exceed a given Jaro-Winkler threshold remains challenging as the string collection grows large. Existing algorithms for computing the Jaro-Winkler similarity between two strings are either slow for short strings or require quadratic time and fail to scale to long strings. To reduce the number of string comparisons, state-of-the-art techniques perform an index-nested-loop join on trie-based indexes; however, they have only been evaluated on small datasets and high similarity thresholds. For lower similarity thresholds that are relevant in practice, these techniques incur an excessive number of costly Jaro-Winkler computations. We present a novel approach for efficient Jaro-Winkler similarity joins on large string collections. Our solution is based on three key techniques: (1) JUMP is a fast algorithm for computing the Jaro-Winkler similarity between two strings that runs in linear time for bounded alphabets. (2) Dual-bound intersection bitmaps encode strings -- represented as character sets -- into fixed-size bitmaps, which are used to prune candidate string pairs by providing effective bounds on their character overlap; unlike prior bitmap techniques, our bitmaps are equipped with a cost model and adapt to the input data. (3) Our PUMA framework groups strings by prefix to tighten filter conditions, thereby reducing the number of similarity computations. Extensive experiments on large string collections demonstrate that our approach consistently outperforms the state of the art, achieving speedups of up to three orders of magnitude.

cs.DB↗

EPIC-Bench: A Perception-Centric Benchmark for Fine-Grained Embodied Visual Grounding in Vision-Language Models

While large vision-language models (VLMs) are increasingly adopted as the perceptual backbone for embodied agents, their fundamental vision-language grounding (VLG) capability in embodied scenarios remains unexplored. We present \textbf{EPIC-Bench}, \textbf{E}mbodied \textbf{P}ercept\textbf{I}on Ben\textbf{C}hmark, a fine-grained grounding benchmark designed to systematically evaluate the VLG capabilities of VLMs in real-world embodied environments. EPIC-Bench comprises 6.6k meticulously human-annotated tuples (Image, Text, Mask) spanning three core stages of the embodied interaction pipeline: Target Localization, Navigation, and Manipulation. With 23 fine-grained tasks, multi-dimensional settings and evaluation over 80 leading VLMs, EPIC-Bench provides a diagnostic analysis.

cs.CV↗

Foundation Models for Event Prediction Across Heterogeneous Event Logs

Event logs are ubiquitous across real-world domains, including business processes, manufacturing workflows, healthcare, and many other operational systems. Yet event logs can differ substantially in their event semantics, label spaces, interaction patterns, and temporal scales, making it difficult to build a single foundation model that generalizes directly to previously unseen systems. We address this challenge with the Foundation Event Predictor (FEP), a simple and scalable in-context architecture that performs next-event prediction through learned, event-level nonparametric inference. Causal self-attention encodes context and target histories into key and query representations, respectively, while the observed next context events are encoded as values. Cross-attention then performs learned, relevance-weighted aggregation of next-event outcomes over all context transitions, allowing the entire nonparametric prediction procedure to be expressed as standard attention operations and efficiently parallelized with modern kernels such as FlashAttention. To promote cross-system generalization, we introduce permutation-robust embedding learning and random fixed embeddings to decouple event IDs from persistent semantics, together with empirical quantile transforms and robust normalization to handle diverse temporal scales. We pretrain FEP on 696 real-world event datasets spanning heterogeneous event systems and evaluate it on 50 held-out datasets unseen during pretraining. Across these datasets, FEP generalizes effectively to new event systems without parameter adaptation, frequently approaching the performance of strong models trained separately on each target dataset while substantially outperforming existing foundation-model baselines, particularly on challenging prediction tasks.

cs.LG↗

Universal Multi-Modal Traceformer: Integrating Heterogeneous Context for Process Event Prediction

Event logs arise in a wide range of real-world processes, capturing not only event activities and timestamps but also multi-modal contextual information. Existing event-sequence models, including many temporal point process approaches, primarily model event activities and timestamps while overlooking heterogeneous context, such as numerical measurements, categorical attributes, textual descriptions, and metadata associated with individual events and entire traces. In this paper, we propose Universal Multi-Modal Traceformer (UMT), a unified framework for incorporating heterogeneous process context into next-event prediction. Built on a Transformer backbone, UMT introduces a universal feature encoder that maps diverse feature types into a shared representation space and handles contextual information at both the event and trace levels. UMT further develops a per-event Perceiver module that dynamically weights contextual features and adaptively integrates them into event-token representations. To accommodate the heavy-tailed and potentially multi-modal distribution of inter-arrival times, UMT represents each interval at multiple temporal scales and jointly predicts the corresponding scale-specific quantities. Experiments on 13 real-world event logs show that UMT improves both next-event activity and time prediction over existing approaches.

cs.LG↗

Kronecker-Structured Nonparametric Spatiotemporal Point Processes

Events in spatiotemporal domains arise in numerous real-world applications, where uncovering event relationships and enabling accurate prediction are central challenges. Classical Poisson and Hawkes processes rely on restrictive parametric assumptions that limit their ability to capture complex interaction patterns, while recent neural point process models increase representational capacity but integrate event information in a black-box manner, hindering interpretable relationship discovery. To address these limitations, we propose a Kronecker-Structured Nonparametric Spatiotemporal Point Process (KSTPP) that enables transparent event-wise relationship discovery while retaining high modeling flexibility. We model the background intensity with a spatial Gaussian process (GP) and the influence kernel as a spatiotemporal GP, allowing rich interaction patterns including excitation, inhibition, neutrality, and time-varying effects. To enable scalable training and prediction, we adopt separable product kernels and represent the GPs on structured grids, inducing Kronecker-structured covariance matrices. Exploiting Kronecker algebra substantially reduces computational cost and allows the model to scale to large event collections. In addition, we develop a tensor-product Gauss-Legendre quadrature scheme to efficiently evaluate intractable likelihood integrals. Extensive experiments demonstrate the effectiveness of our framework.

cs.LG↗

Magnetic and electric properties of the metallic kagome antiferromagnet CrRhAs

CrRhAs is an antiferromagnetic kagome metal predicted to host a nontrivial spin texture with vector spin chirality [Huang \textit{et al.}, \textit{npj Quantum Mater.} \textbf{8}, 32 (2023)]. We report the synthesis and basic characterization of CrRhAs single crystals, which exhibit an antiferromagnetic transition with $T_{\rm N}$ = 150~K, evidenced by electrical transport, heat capacity, and magnetization measurements. Hall resistivity varies linearly with magnetic field, i.e., there is no nonlinear Hall contribution. Intriguingly, the Hall coefficient changes sign between the configurations of $j \parallel ab, H \perp ab$ and $j \parallel c, H \perp c$, which is likely connected to a peculiar topology of the Fermi surface. Furthermore, for $j \parallel ab$, the Hall coefficient shows a pronounced and continuous enhancement below $T_{\rm N}$, signaling a significant reconstruction of the Fermi surface or an extra scattering from the magnons. Our results offer guidance for exploring anomalous electric transport phenomena in exotic magnetic systems.

cond-mat.str-el↗

Complex Magnetic Behavior of the Ce sawtooth chains in CeRhSn$_2$

Conflicting reports exist on the ground state of the intermetallic compound CeRhSn$_2$. This can be rooted in the sawtooth-like arrangement of two inequivalent Ce sites in the unit cell, which suggests potential geometric magnetic frustration. To resolve, we conducted a comprehensive study on high-quality single crystals of CeRhSn$_2$ by means of magnetization ($M$), specific heat ($C_p/T$), and resistivity ($ρ$). The system exhibits strong magnetic anisotropy, confirming the $b$-axis as the easy magnetic axis. We establish three successive transitions, an AFM order at $T_{N} = 3.65$K, a first-order FM order at $T_{C} = 1.7$K and final transition, at $T = 1.5$K. The transition temperatures are highly field-directional dependent: in a magnetic field, the lowest transition is immediately suppressed while $\mathbf{H} \parallel b$ rapidly merges $T_{C}$ and $T_{N}$ into a single second-order transition. Conversely, $\mathbf{H}\parallel c$ suppresses the FM order and reduces $T_{N}$. Additional ab initio calculations affirm the FM ground state of CeRhSn$_2$. The observation of an enhancement of the Sommerfeld coefficient ($γ= 76.5$mJ/mol$\cdot$K$^2$) may arise from geometric frustration, but it is most consistently attributed to weak Kondo hybridization as frustration cannot be conclusively established through our data.

cond-mat.str-el↗

Structured Neural Marked Point Processes for Interpretable Event Interaction Modeling

Multi-class event streams arise in numerous real-world applications, where uncovering structured, interpretable inter-event relationships, together with accurate prediction, remains a central challenge. Existing neural point process models are highly expressive but encode event interactions in a black-box manner, preventing explicit discovery of structured dependencies. In this paper, we propose a structured neural marked point process (SNMPP) that achieves high modeling flexibility while enabling explicit event-wise and class-wise relationship discovery from data. Our model constructs a product-form neural influence kernel composed of a signed interaction network over event types and a delay-aware monotonic temporal network. This design enables explicit characterization of inter-class influence topology -- including excitation, inhibition, and neutrality -- while flexibly capturing diverse temporal decay patterns and potential influence delays. For efficient learning, we develop a stratified Monte Carlo estimator for stochastic training. Extensive experiments on synthetic and real-world benchmark datasets validate the ability of our approach to uncover structured relationships and deliver strong predictive performance.

cs.LG↗

Arbitrarily Conditioned Hierarchical Flows for Spatiotemporal Events

Events in spatiotemporal systems are ubiquitous, yet modeling their complex distributions remains challenging. Existing point process models often rely on strong structural assumptions and are typically limited to autoregressive, event-by-event prediction. As a result, they struggle to support broader inference tasks such as inverse inference, trajectory reconstruction, and recovery of missing event locations. We introduce Arbitrarily Conditioned Hierarchical Flows (ARCH), a hierarchical flow matching framework for spatiotemporal event modeling. ARCH is expressive enough to capture complex event distributions while enabling tractable and accurate computation of conditional intensities, which quantify instantaneous event risk. Built on a history-encoder-generative-decoder architecture, ARCH introduces a hybrid masking strategy for flexible conditioning on arbitrary observed events. This enables a unified treatment of forecasting, inverse inference, and partial trajectory recovery within a single framework. Experiments on synthetic and real-world datasets show that ARCH consistently outperforms existing baselines across both prediction and conditional inference tasks.

cs.LG↗

SPAT: A Semantic Port-Aware Adaptive-Rate Transmission Protocol for Semantic Communication

With the evolution of 6G, semantic communication has emerged as a promising paradigm by prioritizing the delivery of task-relevant meaning over strict bit-level correctness. However, existing transport mechanisms still rely on explicit port headers and bit-level validation, making them vulnerable to header corruption and the resulting packet loss. To address this issue, this paper proposes a Semantic Port-Aware Adaptive-Rate Transmission Protocol (SPAT) for semantic communication. The proposed framework jointly embeds source and destination port information into semantic representations, thereby reducing dependence on explicit port headers while enabling robust port-aware transmission. Furthermore, a differentiated semantic processing mechanism is developed for uplink and downlink scenarios, where port identification is introduced for uplink service recognition and destination-aware conditional gating is designed for downlink selective decoding. In addition, an adaptive-rate controller is incorporated to dynamically adjust the number of transmitted semantic channels according to channel conditions and feature importance, thereby improving both robustness and transmission efficiency. Experimental results on the AFHQ and ImageNet-10 datasets, together with real-world experimental measurements, demonstrate that SPAT consistently outperforms TCP, UDP, and SITP in reconstruction quality across different SNRs while maintaining low-latency transmission.

eess.SP↗

Magnetic field and pressure tuning of the heavy fermion antiferromagnet CePdIn

Frustrated Kondo lattices are ideal platforms for studying how both the Kondo effect and quantum fluctuations compete with the magnetic exchange interactions that drive magnetic ordering. Here, we investigate the effect of tuning the heavy-fermion compound CePdIn, which crystallizes in the geometrically frustrated ZrNiAl-type structure, using applied magnetic fields and hydrostatic pressure. At ambient pressure, CePdIn exhibits two magnetic transitions, one at $T_{\rm{N}} \approx 1.65$ K and another at $T_{\rm{M}} \approx 1.15$ K, which are both suppressed by applied $c$-axis fields. Upon applying pressure in zero magnetic field, there is a non-monotonic evolution of $T_{\rm{N}}$, which decreases to 0.8 K at 2.3 GPa, before abruptly increasing to 1.5 K at 2.6 GPa. At higher pressures, $T_{\rm{N}}$ has a weak pressure dependence, and vanishes near 5 GPa. Together with the high-pressure phase being more robust to applied fields, these results suggest two distinct antiferromagnetic phases in CePdIn, which are separated near 2.6 GPa, and this change may be driven by the evolution of the underlying electronic structure due to enhanced Kondo hybridization under pressure.

cond-mat.str-el↗

Query Suggestion for Retrieval-Augmented Generation via Dynamic In-Context Learning

Retrieval-augmented generation with tool-calling agents (agentic RAG) has become increasingly powerful in understanding, processing, and responding to user queries. However, the scope of the grounding knowledge is limited and asking questions that exceed this scope may lead to issues like hallucination. While guardrail frameworks aim to block out-of-scope questions (Rodriguez et al., 2024), no research has investigated the question of suggesting answerable queries in order to complete the user interaction. In this paper, we initiate the study of query suggestion for agentic RAG. We consider the setting where user questions are not answerable, and the suggested queries should be similar to aid the user interaction. Such scenarios are frequent for tool-calling LLMs as communicating the restrictions of the tools or the underlying datasets to the user is difficult, and adding query suggestions enhances the interaction with the RAG agent. As opposed to traditional settings for query recommendations such as in search engines, ensuring that the suggested queries are answerable is a major challenge due to the RAG's multi-step workflow that demands a nuanced understanding of the RAG as a whole, which the executing LLM lacks. As such, we introduce robust dynamic few-shot learning which retrieves examples from relevant workflows. We show that our system can be self-learned, for instance on prior user queries, and is therefore easily applicable in practice. We evaluate our approach on three benchmark datasets based on two unlabeled question datasets collected from real-world user queries. Experiments on real-world datasets confirm that our method produces more relevant and answerable suggestions, outperforming few-shot and retrieval-only baselines, and thus enable safer, more effective user interaction with agentic RAG.

cs.CL↗

Bridging VLMs and Embodied Intelligence with Deliberate Practice Policy Optimization

Developing a universal and versatile embodied intelligence system presents two primary challenges: the critical embodied data bottleneck, where real-world data is scarce and expensive, and the algorithmic inefficiency of existing methods, which are resource-prohibitive. To address these limitations, we introduce Deliberate Practice Policy Optimization (DPPO), a metacognitive ``Metaloop'' training framework that dynamically alternates between supervised fine-tuning (competence expansion) and reinforcement learning (skill refinement). This enables automatic weakness identification and targeted resource allocation, specifically designed to maximize learning efficiency from sparse, finite data. Theoretically, DPPO can be formalised as a unified preference-learning framework. Empirically, training a vision-language embodied model with DPPO, referred to as Pelican-VL 1.0, yields a 20.3% performance improvement over the base model and surpasses open-source models at the 100B-parameter scale by 10.6%. We are open-sourcing both the models and code, providing the first systematic framework that alleviates the data and resource bottleneck and enables the community to build versatile embodied agents efficiently.

cs.AI↗

Pelican-VL 1.0: A Foundation Brain Model for Embodied Intelligence

This report presents Pelican-VL 1.0, a new family of open-source embodied brain models with parameter scales ranging from 7 billion to 72 billion. Our explicit mission is clearly stated as: To embed powerful intelligence into various embodiments. Pelican-VL 1.0 is currently the largest-scale open-source embodied multimodal brain model. Its core advantage lies in the in-depth integration of data power and intelligent adaptive learning mechanisms. Specifically, metaloop distilled a high-quality dataset from a raw dataset containing 4+ billion tokens. Pelican-VL 1.0 is trained on a large-scale cluster of 1000+ A800 GPUs, consuming over 50k+ A800 GPU-hours per checkpoint. This translates to a 20.3% performance uplift from its base model and outperforms 100B-level open-source counterparts by 10.6%, placing it on par with leading proprietary systems on well-known embodied benchmarks. We establish a novel framework, DPPO (Deliberate Practice Policy Optimization), inspired by human metacognition to train Pelican-VL 1.0. We operationalize this as a metaloop that teaches the AI to practice deliberately, which is a RL-Refine-Diagnose-SFT loop.

cs.LG↗