Searcharxiv⌕ Search

arXiv subjects

Kangjie Zhou

Publications and source records attributed to Kangjie Zhou.

At least 19 recordsLinked to original sources

Learning and Verifying Progress: Closed-Loop Agentic Planning with Progress-Enhanced Execution for Long-Horizon Vision-Language Navigation

Long-horizon vision-language navigation (VLN) requires agents to track instruction progress and recover from local errors before they propagate across subtasks. However, executors trained on expert trajectories encounter policy-induced state shifts during deployment, and their stop signals do not reliably indicate subtask completion. We introduce Learning and Verifying Progress for VLN (LeaP-VLN), an agentic VLN system that couples progress-aware execution with independent progress verification through plan--execute--verify--recover cycles. LeaP-VLN leverages progress both explicitly, through a progress verifier that verifies subtask completion before the agent advances and triggers dual-phase recovery upon route deviation, and implicitly, by co-training the executor on progress reasoning and action prediction to enhance the executor's progress awareness and corrective control. To adapt the executor to this closed-loop interface, we propose a two-stage progress-enhanced curriculum: behavior cloning on expert demonstrations, followed by joint training on demonstrations, expert action corrections, and progress-reasoning supervision. Policy-generated trajectories expose the executor to both intra-subtask drift and inter-subtask state shifts, aligning training with the deviations encountered during closed-loop execution. Extensive experiments on R2R-CE and RxR-CE demonstrate state-of-the-art success rates among monocular methods trained without additional navigation datasets, while real-world experiments further validate the effectiveness of LeaP-VLN in long-horizon navigation.

cs.RO↗

CovNav: Conformal Prediction-Guided Adaptive Inference for Object Goal Navigation

In object goal navigation (ObjectNav), an embodied agent must find a target object (e.g., "find a bed") in a previously unseen environment, relying on spatial commonsense such as object co-occurrence. Large Language Models (LLMs) encode this commonsense, but deploying them on robots exposes a trade-off between reasoning quality and efficiency: cloud-based models incur latency and privacy risks, whereas small on-device LLMs are fast but prone to hallucinated spatial judgments. Hybrid reasoning LLMs offer both a fast non-thinking mode and a slow thinking mode, which raises a key question for embodied agents: when is slow thinking worth its cost? We propose CovNav, a training-free framework that answers this question with Conformal Prediction (CP), turning dual-process (System 1 / System 2) inference into an uncertainty-aware decision. Much like recent decision models such as Jev, CovNav casts each spatial query as a structured binary decision rather than free-form generation: a local LLM in fast mode judges whether the target is likely near an observed object, CP converts its token probabilities into a prediction set, and only ambiguous queries are escalated to slow chain-of-thought reasoning. The object-pair formulation keeps calibration tractable and supports open-set objects through off-the-shelf detectors. Experiments on Gibson and HM3D show that CovNav surpasses the strongest zero-shot baseline by 0.8 and 1.7 percentage points in success rate on HM3D v1 and v2, while invoking slow reasoning for only 12.4% of queries and retaining most of the gain of always-on thinking at a fraction of its latency. Real-robot experiments on a TurtleBot 4 further validate the approach. CovNav shows how calibrated uncertainty can decide when an embodied agent should think slowly, offering a principled form of adaptive test-time reasoning for LLM-based robots.

cs.RO↗

CLOSER: Evidence-Aware Active Visual Verification for Embodied Object-Goal Navigation

Object-goal navigation (ObjectNav) requires an embodied agent to find an instance of a specified category in an unseen environment and stop near it. The agent typically relies on an object detector but must judge online from a limited egocentric video stream, sometimes a single frame, whether a detected candidate is the target. Detectors often confidently mistake similar objects, such as a sofa for a bed, and stopping at a false positive fails the episode. Prevailing systems also approach candidates greedily, leaving them occluded or partially observed. We present CLOSER, an evidence-aware active verification framework that treats stopping as closed-loop evidence acquisition, coupling vision-language model (VLM) verification with active perception. CLOSER fine-tunes an open-source VLM to label each view of a candidate tracked across frames as target, non-target, or uncertain, selects complementary views with an embedding-based diversity gate, and fuses their per-view probabilities with a lightweight network. If the fused judgment remains uncertain, a diffusion policy drives the agent toward the candidate using only the tracked mask and depth until the decision becomes decisive. CLOSER runs locally and plugs into existing ObjectNav systems without retraining them. On Gibson and Habitat-Matterport 3D, it raises average success rate by 3.6 points and success weighted by path length by 2.0 points, showing that verifying evidence before committing improves embodied navigation.

cs.CV↗

Replica Symmetry Breaking and Algorithmic Thresholds in Empirical Risk Minimization under Multi-Index Model

Modern machine learning models are trained by optimizing high-dimensional non-convex empirical risk functions. Such cost functions can have a multitude of local optima and yet, gradient-based optimization appears to converge to near-global optima. Within a simple supervised learning setting, we develop a precise picture of which parts of the empirical risk landscape are accessible by polynomial-time algorithms. We are given i.i.d. pairs $\{(\boldsymbol{x}_i,y_i):\; 1 \le i\le n\}$ with $\boldsymbol{x}_i\in \mathbb{R}^d$ standard Gaussian feature vectors, and $y_i\in\mathbb{R}$ response variables that depend on $\boldsymbol{x}_i$ through their projections on an unknown $k$-dimensional subspace. We use empirical risk minimization to learn a model that depends on an $m$-dimensional projection of the data (e.g., an $m$-neurons neural network). We propose an incremental approximate message passing (IAMP) algorithm and precisely characterize the training error it achieves, as well as the relation between test and training error, in the high dimensional asymptotics $n,d\to\infty$, with $n/d\toα\in (0, +\infty)$. Based on earlier work in related models, we expect that the performance achieved by our algorithm is optimal among polynomial-time algorithms.

cs.LG↗

Efficient and Adaptive Estimation of Portfolio Weights with Spectral Risk Measures

Spectral risk measures, including conditional value-at-risk (CVaR), generate a family of convex criteria for portfolio estimation. We study how the criterion should be chosen when different criteria share the same population minimizer, and whether the efficient criterion can itself be learned from data. We develop a general asymptotic theory for empirical spectral-risk minimization under estimated linear constraints. Under normal scale-mixture elliptical returns, all spectral risk measures identify the same population efficient portfolio, but their empirical minimizers have different sampling distributions. Their asymptotic covariance decomposes into a common component and a positive-semidefinite component scaled by a functional of the spectral measure, reducing efficiency to an optimization over probability measures. We characterize the efficiency-optimal spectral measure and show that single-level CVaR is generally inefficient. We then construct a fully data-adaptive estimator that learns the radial distribution and optimal spectral measure from the same observations used for portfolio estimation, yet has the same first-order distribution as the infeasible oracle. Simulations and an empirical application illustrate the method.

math.ST↗

Basic Inequalities for First-Order Optimization with Applications to Statistical Risk Analysis

In this work, we introduce $\textit{basic inequalities}$ for first-order iterative optimization algorithms, forming a simple yet versatile framework which connects implicit and explicit regularization. Building on related comparison inequalities for optimization iterates that already exist in the literature, we extend and unify these arguments to produce a general framework, which can be used as a tool for statistical analysis. In more detail, let $f$ denote the objective function to be optimized. Given a first-order iterative algorithm initialized at $θ_0$, with current iterate $θ_T$, the basic inequality upper bounds $f(θ_T) - f(z)$ for any reference point $z$ in terms of the accumulated step sizes, and the distances between $θ_0$, $θ_T$, and $z$. These distances are measured in a geometry inherent to the optimization algorithm, which then translates into a notion of regularization being applied across the path of iterates. In addition to refining existing results on gradient descent, we provide new results for mirror descent and other first-order methods. We then show how to use these basic inequalities to derive elementary yet useful bounds on the prediction risk of early-stopped gradient descent and exponentiated gradient descent iterates in generalized linear models. We also supplement these findings with numerical experiments.

math.ST↗

GroupKV: Hierarchical KV Cache Management for Long-Context Diffusion LLM Inference

Diffusion large language models (dLLMs) are emerging as a promising generative paradigm that complements autoregressive decoding. In long-context settings, KV cache bloat and offloading transfer overhead have become primary bottlenecks in inference systems. Meanwhile, the periodic full-sequence recomputation and localized token updates in dLLMs make the KV lifecycle substantially more dynamic, complicating cache management and prefetch scheduling while making heavyweight token-level indexing or clustering schemes harder to amortize effectively during decoding. To address these challenges, we present \textsc{GroupKV}, a lightweight hierarchical KV cache management system for long-context dLLM inference. We observe that under block-wise decoding, tokens within the same generation block tend to access highly overlapping and spatially concentrated context regions, making group-level sparse selection effective. Building on this observation, \textsc{GroupKV} partitions the context into contiguous groups and performs coarse-to-fine sparse selection. \textsc{GroupKV} further exploits cross-layer consistency to enable predictive prefetching, and incorporates a staleness correction mechanism to maintain cache coherence under dynamic KV updates. Additionally, \textsc{GroupKV} adopts streaming prefill to reduce peak memory consumption during prefilling. Experiments show that \textsc{GroupKV} extends the maximum serviceable context length by up to $48.00\times$ under constrained GPU memory, improves end-to-end inference performance by up to $3.73\times$ in offload-based long-context settings, and maintains competitive task accuracy.

cs.OS↗

When Does Model Collapse Occur in Structured Interactive Learning?

The proliferation of generative artificial intelligence has given rise to an interactive learning environment, where model parameters are continuously updated using not only data generated by natural processes, but also synthetic outputs produced by other models. This paradigm introduces two major challenges: (1) training data are no longer drawn exclusively from the target population, undermining a core assumption of classical statistical learning, and (2) model training processes become inherently correlated, as models interact with one another through repeated exposure to each other's synthetic outputs in a potentially complex manner. Establishing reliable statistical inference in such structured interactive learning environments therefore remains an important open problem. In particular, there is growing concern about model collapse, a phenomenon in which the performance of generative models progressively degrades as they are trained on synthetic data produced by earlier model generations. Prior work on model collapse primarily focuses on a single model trained on its own output, failing to capture model performance in multi-model interactive settings. In this work, we fill this gap by investigating the performance of generative models in an interactive learning environment with general interaction patterns. In particular, we formalize model interactions using directed graphs and show that the occurrence of model collapse depends critically on the topology of the interaction graph. We further derive an explicit necessary and sufficient condition characterizing when model collapse occurs, and establish finite-sample results for linear regression and asymptotic guarantees for general M-estimators. We support our theoretical findings through extensive numerical experiments.

cs.LG↗

CoINS: Counterfactual Interactive Navigation via Skill-Aware VLM

Recent Vision-Language Models (VLMs) have demonstrated significant potential in robotic planning. However, they typically function as semantic reasoners, lacking an intrinsic understanding of the specific robot's physical capabilities. This limitation is particularly critical in interactive navigation, where robots must actively modify cluttered environments to create traversable paths. Existing VLM-based navigators are predominantly confined to passive obstacle avoidance, failing to reason about when and how to interact with objects to clear blocked paths. To bridge this gap, we propose Counterfactual Interactive Navigation via Skill-aware VLM (CoINS), a hierarchical framework that integrates skill-aware reasoning and robust low-level execution. Specifically, we fine-tune a VLM, named InterNav-VLM, which incorporates skill affordance and concrete constraint parameters into the input context and grounds them into a metric-scale environmental representation. By internalizing the logic of counterfactual reasoning through fine-tuning on the proposed InterNav dataset, the model learns to implicitly evaluate the causal effects of object removal on navigation connectivity, thereby determining interaction necessity and target selection. To execute the generated high-level plans, we develop a comprehensive skill library through reinforcement learning, specifically introducing traversability-oriented strategies to manipulate diverse objects for path clearance. A systematic benchmark in Isaac Sim is proposed to evaluate both the reasoning and execution aspects of interactive navigation. Extensive simulations and real-world experiments demonstrate that CoINS significantly outperforms representative baselines, achieving a 17\% higher overall success rate and over 80\% improvement in complex long-horizon scenarios compared to the best-performing baseline

cs.RO↗

AINav: Large Language Model-Based Adaptive Interactive Navigation

Robotic navigation in complex environments remains a critical research challenge. Traditional navigation methods focus on optimal trajectory generation within fixed free workspace, therefore struggling in environments lacking viable paths to the goal, such as disaster zones or cluttered warehouses. To address this problem, we propose AINav, an adaptive interactive navigation approach that proactively interacts with environments to create feasible paths to achieve originally unreachable goals. Specifically, we present a primitive skill tree for task planning with large language models (LLMs), facilitating effective reasoning to determine interaction objects and sequences. To ensure robust subtask execution, we adopt reinforcement learning to pre-train a comprehensive skill library containing versatile locomotion and interaction behaviors for motion planning. Furthermore, we introduce an adaptive replanning approach featuring two LLM-based modules: an advisor serving as a flexible replanning trigger and an arborist for autonomous plan adjustment. Integrated with the tree structure, the replanning mechanism allows for convenient node addition and pruning, enabling rapid plan adaptation in a priori unknown environments. Comprehensive simulations and experiments have demonstrated AINav's effectiveness and adaptivity in diverse scenarios. The supplementary video is available at: https://youtu.be/CjXm5KFx9AI.

cs.RO↗

ReSPIRe: Informative and Reusable Belief Tree Search for Robot Probabilistic Search and Tracking in Unknown Environments

Target search and tracking (SAT) is a fundamental problem for various robotic applications such as search and rescue and environmental exploration. This paper proposes an informative trajectory planning approach, namely ReSPIRe, for SAT in unknown cluttered environments under considerably inaccurate prior target information and limited sensing field of view. We first develop a novel sigma point-based approximation approach to fast and accurately estimate mutual information reward under non-Gaussian belief distributions, utilizing informative sampling in state and observation spaces to mitigate the computational intractability of integral calculation. To tackle significant uncertainty associated with inadequate prior target information, we propose the hierarchical particle structure in ReSPIRe, which not only extracts critical particles for global route guidance, but also adjusts the particle number adaptively for planning efficiency. Building upon the hierarchical structure, we develop the reusable belief tree search approach to build a policy tree for online trajectory planning under uncertainty, which reuses rollout evaluation to improve planning efficiency. Extensive simulations and real-world experiments demonstrate that ReSPIRe outperforms representative benchmark methods with smaller MI approximation error, higher search efficiency, and more stable tracking performance, while maintaining outstanding computational efficiency.

cs.RO↗

Which exceptional low-dimensional projections of a Gaussian point cloud can be found in polynomial time?

Given $d$-dimensional standard Gaussian vectors $\boldsymbol{x}_1,\dots, \boldsymbol{x}_n$, we consider the set of all empirical distributions of its $m$-dimensional projections, for $m$ a fixed constant. Diaconis and Freedman (1984) proved that, if $n/d\to \infty$, all such distributions converge to the standard Gaussian distribution. In contrast, we study the proportional asymptotics, whereby $n,d\to \infty$ with $n/d\to α\in (0, \infty)$. In this case, the projection of the data points along a typical random subspace is again Gaussian, but the set $\mathscr{F}_{m,α}$ of all probability distributions that are asymptotically feasible as $m$-dimensional projections contains non-Gaussian distributions corresponding to exceptional subspaces. Non-rigorous methods from statistical physics yield an indirect characterization of $\mathscr{F}_{m,α}$ in terms of a generalized Parisi formula. Motivated by the goal of putting this formula on a rigorous basis, and to understand whether these projections can be found efficiently, we study the subset $\mathscr{F}^{\rm alg}_{m,α}\subseteq \mathscr{F}_{m,α}$ of distributions that can be realized by a class of iterative algorithms. We prove that this set is characterized by a certain stochastic optimal control problem, and obtain a dual characterization of this problem in terms of a variational principle that extends Parisi's formula. As a byproduct, we obtain computationally achievable values for a class of random optimization problems including `generalized spherical perceptron' models.

math.PR↗

Implicit Bias of Gradient Descent for Non-Homogeneous Deep Networks

We establish the asymptotic implicit bias of gradient descent (GD) for generic non-homogeneous deep networks under exponential loss. Specifically, we characterize three key properties of GD iterates starting from a sufficiently small empirical risk, where the threshold is determined by a measure of the network's non-homogeneity. First, we show that a normalized margin induced by the GD iterates increases nearly monotonically. Second, we prove that while the norm of the GD iterates diverges to infinity, the iterates themselves converge in direction. Finally, we establish that this directional limit satisfies the Karush-Kuhn-Tucker (KKT) conditions of a margin maximization problem. Prior works on implicit bias have focused exclusively on homogeneous networks; in contrast, our results apply to a broad class of non-homogeneous networks satisfying a mild near-homogeneity condition. In particular, our results apply to networks with residual connections and non-homogeneous activation functions, thereby resolving an open problem posed by Ji and Telgarsky (2020).

cs.LG↗

Discrepancy Algorithms for the Binary Perceptron

The binary perceptron problem asks us to find a sign vector in the intersection of independently chosen random halfspaces with intercept $-κ$. We analyze the performance of the canonical discrepancy minimization algorithms of Lovett-Meka and Rothvoss/Eldan-Singh for the asymmetric binary perceptron problem. We obtain new algorithmic results in the $κ= 0$ case and in the large-$|κ|$ case. In the $κ\to-\infty$ case, we additionally characterize the storage capacity and complement our algorithmic results with an almost-matching overlap-gap lower bound.

cs.DS↗

SwarmDiff: Swarm Robotic Trajectory Planning in Cluttered Environments via Diffusion Transformer

Swarm robotic trajectory planning faces challenges in computational efficiency, scalability, and safety, particularly in complex, obstacle-dense environments. To address these issues, we propose SwarmDiff, a hierarchical and scalable generative framework for swarm robots. We model the swarm's macroscopic state using Probability Density Functions (PDFs) and leverage conditional diffusion models to generate risk-aware macroscopic trajectory distributions, which then guide the generation of individual robot trajectories at the microscopic level. To ensure a balance between the swarm's optimal transportation and risk awareness, we integrate Wasserstein metrics and Conditional Value at Risk (CVaR). Additionally, we introduce a Diffusion Transformer (DiT) to improve sampling efficiency and generation quality by capturing long-range dependencies. Extensive simulations and real-world experiments demonstrate that SwarmDiff outperforms existing methods in computational efficiency, trajectory validity, and scalability, making it a reliable solution for swarm robotic trajectory planning.

cs.RO↗

Overparametrized linear dimensionality reductions: From projection pursuit to two-layer neural networks

Given a cloud of $n$ data points in $\mathbb{R}^d$, consider all projections onto $m$-dimensional subspaces of $\mathbb{R}^d$ and, for each such projection, the empirical distribution of the projected points. What does this collection of probability distributions look like when $n,d$ grow large? We consider this question under the null model in which the points are i.i.d. standard Gaussian vectors, focusing on the asymptotic regime in which $n,d\to\infty$, with $n/d\toα\in (0,\infty)$, while $m$ is fixed. Denoting by $\mathscr{F}_{m, α}$ the set of probability distributions in $\mathbb{R}^m$ that arise as low-dimensional projections in this limit, we establish new inner and outer bounds on $\mathscr{F}_{m, α}$. In particular, we characterize the Wasserstein radius of $\mathscr{F}_{m,α}$ up to constant multiplicative factors, and determine it exactly for $m=1$. We also prove sharp bounds in terms of Kullback-Leibler divergence and Rényi information dimension. The previous question has application to unsupervised learning methods, such as projection pursuit and independent component analysis. We introduce a version of the same problem that is relevant for supervised learning, and prove a sharp Wasserstein radius bound. As an application, we establish an upper bound on the interpolation threshold of two-layers neural networks with $m$ hidden neurons.

stat.ML↗

Dynamic Factor Analysis of High-dimensional Recurrent Events

Recurrent event time data arise in many studies, including biomedicine, public health, marketing, and social media analysis. High-dimensional recurrent event data involving many event types and observations have become prevalent with advances in information technology. This paper proposes a semiparametric dynamic factor model for the dimension reduction of high-dimensional recurrent event data. The proposed model imposes a low-dimensional structure on the mean intensity functions of the event types while allowing for dependencies. A nearly rate-optimal smoothing-based estimator is proposed. An information criterion that consistently selects the number of factors is also developed. Simulation studies demonstrate the effectiveness of these inference tools. The proposed method is applied to grocery shopping data, for which an interpretable factor structure is obtained.

stat.ME↗

Lower Bounds for the Convergence of Tensor Power Iteration on Random Overcomplete Models

Tensor decomposition serves as a powerful primitive in statistics and machine learning, and has numerous applications in problems such as learning latent variable models or mixture of Gaussians. In this paper, we focus on using power iteration to decompose an overcomplete random tensor. Past work studying the properties of tensor power iteration either requires a non-trivial data-independent initialization, or is restricted to the undercomplete regime. Moreover, several papers implicitly suggest that logarithmically many iterations (in terms of the input dimension) are sufficient for the power method to recover one of the tensor components. Here we present a novel analysis of the dynamics of tensor power iteration from random initialization in the overcomplete regime, where the tensor rank is much greater than its dimension. Surprisingly, we show that polynomially many steps are necessary for convergence of tensor power iteration to any of the true component, which refutes the previous conjecture. On the other hand, our numerical experiments suggest that tensor power iteration successfully recovers tensor components for a broad range of parameters in polynomial time. To further complement our empirical evidence, we prove that a popular objective function for tensor decomposition is strictly increasing along the power iteration path. Our proof is based on the Gaussian conditioning technique, which has been applied to analyze the approximate message passing (AMP) algorithm. The major ingredient of our argument is a conditioning lemma that allows us to generalize AMP-type analysis to non-proportional limit and polynomially many iterations of the power method.

cs.LG↗