Searcharxiv⌕ Search

arXiv subjects

Siming He

Publications and source records attributed to Siming He.

At least 19 recordsLinked to original sources

Racing the Clock: Teaching VLNs Budget-Conditioned Object Search

How do action budgets shape vision-language navigation (VLN) exploration, and what expert supervision best teaches budget-aware search? We study budget-conditioned object search: finding as many distinct instances of a requested category as possible within a budget, with discovery defined geometrically by camera field of view and proximity. Distinct discoveries provide a denser evaluation signal than single-target success. We cast expert design as a choice of scene abstraction: a single budget-indexed Bellman recursion, applied to different abstractions of the full environment, yields experts that differ in what they know. An object-level Bellman expert is given every target location and reduces search to budgeted routing. A room-level Bellman expert retains observed room connectivity, doorway evidence, and cross-scene semantic priors but withholds the contents of unexplored rooms, so it must search for targets that its student also cannot see. We distill both experts into Uni-NaVid under matched training conditions and also evaluate greedy-frontier supervision on held-out HM3D scenes. Changing only the stated budget changes learned trajectories. Despite higher object-level expert return, room-level supervision achieves the highest mean collection among evaluated VLNs at every tested budget. At 2000 steps, it finds 72.7% of reachable targets, versus 68.5% under greedy-frontier supervision, 65.5% under object-level supervision, and 44.8% for the pretrained VLN. These results show that expert-return rankings need not predict supervision quality for VLN exploration, and motivate choosing an expert's abstraction with the student's information in mind rather than by expert return alone.

cs.RO↗

Nonlinear Inviscid Damping of Couette Flow with Vorticity Perturbations Supported up to the Boundary

We establish nonlinear inviscid damping of the near-Couette flows in a finite channel, with vorticity perturbations vanishing to infinite order at the boundary. To study the boundary effect, we decompose the vorticity into dyadic layers whose supports expand toward the boundary, and use an inductive Newton scheme to track nonlinear interactions across layers. Finally, precise elliptic estimates are established to quantify the velocity generated by different layers.

math.AP↗

Fast separation of Euler and Navier--Stokes flows in two dimensions

We study the long-time validity of the inviscid limit problem by measuring when the two-dimensional ideal fluid ceases to accurately approximate a slightly viscous fluid with the same initial datum. We construct a stationary solution of the Euler equation from which the Navier--Stokes solution separates at a time scale proportional to the logarithm of the inverse viscosity, achieving the fastest separation from Euler steady states.

math.AP↗

BT-Agent: Deploying Onboard Robot Agents with Behavior Trees and Multi-Fidelity Evaluation

Deploying language-model agents onboard mobile robots requires task reasoning and responsive execution within limited compute and memory. We propose BT-Agent, an onboard architecture combining LLM-driven task decisions with predefined behavior-tree (BT) skills. These skills enforce execution conditions and return structured failure feedback. We demonstrate fully onboard operation with a 9B model on a quadruped across five task categories. Intervention tests show that local checks stop ongoing navigation without waiting for another model decision, while the agent remains active to retry or report constraints. To compare agent designs efficiently, we introduce a 161-task multi-fidelity benchmark, evaluated in text on all tasks and in Habitat on 39. Text supports 10.9$\times$ higher evaluation throughput, while simulation reveals how perception and motion failures change task outcomes. BT-Agent's largest success gains over the strongest Act/ReAct baseline occur at 4B and 9B: approximately 10 and 5 percentage points. Ablations identify structured tool calls and informative failure feedback as the most useful components.

cs.RO↗

Morphometric Imitation: From Morphology and Contact Aware Hand Retargeting to Sim-to-Real Visuomotor Policy

Human hand-object interactions (HOIs) provide a rich source of demonstrations for dexterous manipulation, but learning directly from them presents challenges in bridging morphology gaps, ensuring dynamical feasibility, and sim-to-real deployment. We present Morphometric Imitation, a three-stage framework that transforms reconstructed HOIs into zero-shot sim-to-real visuomotor policies. First, morphometric optimization (MMO) kinematically retargets human motion across hand morphologies while preserving demonstrated contacts. Second, residual reinforcement learning (RL) refines the kinematic reference using object pose and contact information from the human motion to produce dynamically feasible robot demonstrations. Third, these demonstrations are distilled into visuomotor policies. Across three robot hands and ten HOIs, MMO outperforms five baselines in contact F1, improving on the strongest ones by 8 to 28 points, while improving the success rate of downstream dynamic retargeting by as much as 35 points. On a Sharpa hand, the visuomotor policies achieve 89.3% zero-shot success in 300 real-world trials on 30 objects spanning 10 categories. Project page: https://morphometricimitation.github.io

cs.RO↗

A Scene Language Model for Open-Vocabulary Scene Mapping

Open-vocabulary 3D scene mapping aims to build a persistent representation of the objects in an environment. Existing systems typically rely on engineered mapping pipelines to associate observations, merge information across views, and maintain a consistent scene representation over time. Many additionally store feature-rich object representations, such as embeddings or image crops, increasing the size and complexity of the persistent memory. We introduce SceneLM, a Scene-Language Model that directly maintains a textual scene map. The full scene is represented as a structured text list of objects, which serves as the model's only persistent memory. For each input image, the model reads the current scene state and updates the map by adding, editing, and removing objects. To learn this behavior, we introduce supervision tasks for iterative scene map maintenance together with an automatic annotation pipeline that generates training data from images without human labels. We evaluate SceneLM on both a language-grounded retrieval benchmark and a localization benchmark. Across both benchmarks, the model produces a scene map that achieves competitive performance with complete mapping systems built from dedicated perception and geometric modules while producing a scene representation that is 6-12x more compact. We further show that SceneLM can be run online on an edge device through experiments on a quadruped. These results show that a persistent open-vocabulary 3D scene map can be maintained directly by a single vision-language model using only a lightweight text representation. Training and inference code is available on https://goldengait.github.io/scenelm/.

cs.CV↗

Digital Engagement, Income Disparities, and Job Seeking in the United States since 2010

Surveys often record how frequently people use the internet without measuring the infrastructures, skills, and support systems that make digital participation possible. Using the U.S. National Longitudinal Survey of Youth 1997 cohort, we study how internet-use frequency relates to labor income, employment attachment, and job seeking after 2010. The main digital-engagement analysis uses the comparable 2011, 2013, and 2015 waves, with 2017 retained as later labor-market context. Across repeated cross sections, daily internet use consistently marks higher income and stronger employment attachment. Relative to daily use, less-than-daily use is associated with roughly 11 to 20 percent lower income, while nonuse is associated with about 18 to 21 percent lower income in 2011 and 2013. Respondents reporting no internet use are also 13 to 23 percentage points less likely to report full-year work. Job-search estimates reveal a distinct mechanism: active search is governed by employment status, search intensity, and application support, so a frequency item sorts respondents more sharply on durable labor-market attachment than on short-window search. Education accounts for a substantial share of the raw digital gradient, and pooled lagged-outcome and doubly robust transition estimates separate durable stratification from positive adoption margins. The results establish internet-use frequency as an informative behavioral marker of digitally mediated labor-market stratification and clarify why routine use should not be treated as a simple measure of digital access.

cs.CY↗

Sampling and Optimization meet Enhanced Flows

It is well known that the computational realization of Gibbs probability measures, $e^{-\mathbb{U}(\mathbf{x})}/Z$, plays a central role in sampling and optimization. In this paper, we introduce two types of dynamics that exhibit rapid convergence towards these Gibbs measures. The mechanism driving this rapid convergence is the enhanced dissipation associated with these transport-diffusion dynamics. Motivated by these enhanced dynamics, we design numerical algorithms for sampling from the target Gibbs measure. Finally, we provide the corresponding particle systems that may yield other effective numerical samplers.

math.OC↗

FARM: Find Anything using Relational Spatial Memory

Robots operating in homes, warehouses, and other object-rich environments need memory systems that can find specific object instances on demand. Object-level memory alone is often insufficient: scenes contain many plausibly matching objects, and users refer to the target through relations to landmarks and surrounding objects (e.g. ``the tall lamp below the dartboard and to the left of the poster''), demanding a relational spatial memory that supports retrieval through semantic, appearance, and spatial predicates over objects. To achieve this, we present FARM (Find Anything using Relational Spatial Memory), which builds, in real time at 5-10 Hz, a compact, open-vocabulary, object-level memory with geometry, visual-language descriptors, and viewpoint evidence. At query time, FARM uses VLMs to parse the query and score visual evidence, while grounding spatial constraints explicitly through object symbols and relational predicates. This structured use of VLMs enables more accurate and robust retrieval than end-to-end reasoning over frame histories or scene-graph context. In experiments on 44k language queries spanning 67 indoor and outdoor scenes, ranging from 15 to 15,000 m^2, FARM improves Recall@5 and Recall@10 over prior methods by 164% and 224%, and a final VLM reranking stage improves Accuracy@1 by 35%, while running in real time. We further demonstrate closed-loop deployment on a quadrupedal robot using onboard sensors and compute.

cs.RO↗

Couette Flow with Robin Boundary Condition (I): the viscosity-independent friction

This article is the first paper in the series. In this series of articles, we will examine the influence of the friction factor $α$ at the solid--fluid boundary on the stability of Couette flow. Specifically, we consider the stability of Couette flow in a bounded periodic channel $\mathbb{T} \times [-1,1]$ under Robin-type boundary conditions ($u^2|_{y=\pm 1} = 0$, $[α\partial_n u^1 + u^1]|_{y=\pm1} = f $), where $α$ is the friction factor and $n$ is the unit outer normal vector. In this article, we prove that for a given friction factor $α$, as long as the fluid viscosity coefficient $ν\ll α$ is sufficiently small, the system is asymptotically stable if the initial perturbation satisfies $\|ω_{\rm in}\| \leq εν^{1/3}$. Moreover, inviscid damping and enhanced dissipation hold.

math.AP↗

Estimating the Diameter at Breast Height of Trees in a Forest from RGB

Forest inventories rely on accurate measurements of the diameter at breast height (DBH) for ecological monitoring, resource management, and carbon accounting. While LiDAR-based techniques can achieve centimeter-level precision, they are cost-prohibitive and operationally complex. We present a low-cost alternative that only needs a consumer-grade 360 video camera. Our semi-automated pipeline comprises of (i) a dense point cloud reconstruction using Structure from Motion (SfM) photogrammetry software called Agisoft Metashape, (ii) semantic trunk segmentation by projecting Grounded Segment Anything (SAM) masks onto the 3D cloud, and (iii) a robust RANSAC-based technique to estimate cross section shape and DBH. We introduce an interactive visualization tool for inspecting segmented trees and their estimated DBH. On 61 acquisitions of 43 trees under a variety of conditions, our method attains median absolute relative errors of 5-9% with respect to "ground-truth" manual measurements. This is only 2-4% higher than LiDAR-based estimates, while employing a single 360 camera that costs orders of magnitude less, requires minimal setup, and is widely available.

cs.CV↗

An Active Perception Game for Robust Exploration

Active perception approaches select future viewpoints by using some estimate of the information gain. An inaccurate estimate can be detrimental in critical situations, e.g., locating a person in distress. However the true information gained can only be calculated post hoc, i.e., after the observation is realized. We present an approach to estimate the discrepancy between the estimated information gain (which is the expectation over putative future observations while neglecting correlations among them) and the true information gain. The key idea is to analyze the mathematical relationship between active perception and the estimation error of the information gain in a game-theoretic setting. Using this, we develop an online estimation approach that achieves sub-linear regret (in the number of time-steps) for the estimation of the true information gain and reduces the sub-optimality of active perception systems. We demonstrate our approach for active perception using a comprehensive set of experiments on: (a) different types of environments, including a quadrotor in a photorealistic simulation, real-world robotic data, and real-world experiments with ground robots exploring indoor and outdoor scenes; (b) different types of robotic perception data; and (c) different map representations. On average, our approach reduces information gain estimation errors by 42%, increases the information gain by 7%, PSNR by 5%, and semantic accuracy (measured as the number of objects that are localized correctly) by 6%. In real-world experiments with a Jackal ground robot, our approach demonstrated complex trajectories to explore occluded regions.

cs.RO↗

Stability threshold of close-to-Couette shear flows with no-slip boundary conditions in 2D

In this paper, we develop a stability threshold theorem for the 2D incompressible Navier-Stokes equations on the channel, supplemented with the no-slip boundary condition. The initial datum is close to the Couette flow in the following sense: the shear component of the perturbation is small, but independent of the viscosity $ν$. On the other hand, the $x$-dependent fluctuation is assumed small in a viscosity-dependent sense, namely, $O(ν^{\frac12}|\log ν|^{-2})$. Under this setup, we prove nonlinear enhanced dissipation of the vorticity and a time-integrated inviscid damping for the velocity. These stabilizing phenomena guarantee that the Navier-Stokes solution stays close to an evolving shear flow for all time. The analytical challenge stems from a time-dependent nonlocal term that appears in the associated linearized Navier-Stokes equations.

math.AP↗

Mixing, Enhanced Dissipation and Phase Transition in the Kinetic Vicsek Model

In this paper, we study the kinetic Vicsek model, which serves as a starting point for describing the polarization phenomena observed in the experiments of fibroblasts moving on liquid crystalline substrates. The long-time behavior of the kinetic equation is analyzed, revealing that, within specific parameter regimes, the mixing and enhanced dissipation phenomena stabilize the dynamics and ensure effective information communication among agents. Consequently, the solution exhibits features similar to those of a spatially-homogeneous system. As a result, we confirm the phase transition observed in the agent-based Vicsek model at the kinetic level.

math.AP↗

Transition Threshold for Strictly Monotone Shear Flows in Sobolev Spaces

We study the stability of spectrally stable, strictly monotone, smooth shear flows in the 2D Navier-Stokes equations on $\mathbb{T} \times \mathbb{R}$ with small viscosity $ν$. We establish nonlinear stability in $H^s$ for $s \geq 2$ with a threshold of size $εν^{1/3}$ for time smaller than $c_*ν^{-1}$ with $ε, c_* \ll 1$. Additionally, we demonstrate nonlinear inviscid damping and enhanced dissipation.

math.AP↗

From NeRFs to Gaussian Splats, and Back

For robotics applications where there is a limited number of (typically ego-centric) views, parametric representations such as neural radiance fields (NeRFs) generalize better than non-parametric ones such as Gaussian splatting (GS) to views that are very different from those in the training data; GS however can render much faster than NeRFs. We develop a procedure to convert back and forth between the two. Our approach achieves the best of both NeRFs (superior PSNR, SSIM, and LPIPS on dissimilar views, and a compact representation) and GS (real-time rendering and ability for easily modifying the representation); the computational cost of these conversions is minor compared to training the two from scratch.

cs.CV↗

Pseudo-Gevrey Smoothing for the Passive Scalar Equations near Couette

In this article, we study the regularity theory for two linear equations that are important in fluid dynamics: the passive scalar equation for (time-varying) shear flows close to Couette in $\mathbb T \times [-1,1]$ with vanishing diffusivity $ν\to 0$ and the Poisson equation with right-hand side behaving in similar function spaces to such a passive scalar. The primary motivation for this work is to develop some of the main technical tools required for our treatment of the (nonlinear) 2D Navier-Stokes equations, carried out in our companion work. Both equations are studied with homogeneous Dirichlet conditions (the analogue of a Navier slip-type boundary condition) and the initial condition is taken to be compactly supported away from the walls. We develop smoothing estimates with the following three features: [1] Uniform-in-$ν$ regularity is with respect to $\partial_x$ and a time-dependent adapted vector-field $Γ$ which approximately commutes with the passive scalar equation (as opposed to `flat' derivatives), and a scaled gradient $\sqrtν \nabla$; [2] $(\partial_x, Γ)$-regularity estimates are performed in Gevrey spaces with regularity that depends on the spatial coordinate, $y$ (what we refer to as `pseudo-Gevrey'); [3] The regularity of these pseudo-Gevrey spaces degenerates to finite regularity near the center of the channel and hence standard Gevrey product rules and other amenable properties do not hold. Nonlinear analysis in such a delicate functional setting is one of the key ingredients to our companion paper, \cite{BHIW24a}, which proves the full nonlinear asymptotic stability of the Couette flow with slip boundary conditions. The present article introduces new estimates for the associated linear problems in these degenerate pseudo-Gevrey spaces, which is of independent interest.

math.AP↗