arXiv subjects

Yifan Lin

Publications and source records attributed to Yifan Lin.

At least 19 recordsLinked to original sources

arXiv2026

SpikingNav: Robust Embodied Navigation with Spiking Neural Policies

Embodied navigation requires an agent to make sequential decisions from egocentric observations in a physical environment. Existing Artificial Neural Network (ANN)-based navigation models have achieved strong performance, yet they often rely on dense computation and may degrade under visual corruptions. Spiking neural networks (SNNs) provide event-driven computation and intrinsic temporal dynamics, which are promising for compact and robust navigation on resource-constrained platforms. However, whether spike-based sensing and policy dynamics can improve robustness in visually rich embodied navigation remains an open problem. This paper proposes SpikingNav, a spiking framework for robust indoor embodied navigation. It contains a Spiking Sensing Encoder (SSE) and a Spiking Policy Network (SPN). The SSE extracts task-conditioned visual features with a spike-based backbone. The SPN maintains a recurrent policy state through membrane integration, thresholding, and spike-triggered reset. In this way, SpikingNav exploits the dynamic properties and spike activations of SNNs to improve navigation performance and robustness. We evaluate SpikingNav on PointNav and ObjectNav under clean observations and visual corruptions. SpikingNav achieves competitive clean performance and stronger robustness with fewer parameters and lower per-step computation than a matched ANN baseline. For instance, SpikingNav improves ObjectNav success from 31.05% to 34.12%, and raises the average success under visual corruptions from 8.45% to 13.71%, demonstrating the benefits of spike-based sensing and policy dynamics. We further validate the deployability of our spike-based sensing method on the Thruster-V2 neuromorphic chip. This physical hardware validation shows that SpikingNav can be instantiated on a real neuromorphic substrate for cyber-physical systems.

cs.RO↗

arXiv2026

Beyond Transformers: Linear Attention Policy for Open-Vocabulary Object Goal Navigation

Open-Vocabulary Object Goal Navigation (OVON) requires agents to operate under partial observability, making effective internal state updates critical for navigation performance. This update is implemented by the policy network, where recent approaches adopt Transformer-based backbones with self-attention over a context window to integrate temporal information. However, our controlled experiments show that performance does not scale with context length under Transformer-based policies, questioning the suitability of self-attention for state integration in navigation. To this end, we propose Linear Attention-based Navigation (LANav), which adopts linear attention (LA) as the policy backbone to maintain a structured state update rather than self-attention over the context window. Across multiple LA variants evaluated under identical settings, LANav consistently outperforms Transformer-based baselines. Performance improves as state update mechanisms become more structured and regulated, highlighting the importance of state update design. To improve state update effectiveness, we introduce Weighted State-Expansion Linear Attention (WSLA), which expands each attention head's state into multiple sub-states and uses learnable weighted readout to aggregate expanded sub-states. Equipped with WSLA, LANav achieves 36.4% average success rate (SR) on HM3D-OVON, outperforming Transformer-based counterparts by 6.3 percentage points in macro-averaged SR, while maintaining computational efficiency. Distance-stratified results show larger gains in long-distance episodes, while HSSD transfer and fine-tuning demonstrate robustness across scene distributions. Real-world deployment on a Unitree Go2 further achieves an 82% success rate over 50 trials, supporting the practical feasibility and sim-to-real transfer of LANav.

cs.RO↗

arXiv2026

Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened

Self-improving AI agents are designed to learn from their mistakes. We show they can also hallucinate mistakes that never happened. We study this failure mode in automated harness optimization, where an LLM-based proposer edits an agent's scaffold, including prompts, parsers, filters, validators and guardrails, to eliminate observed failures. But this process rarely asks first: was there a real failure to fix? We introduce the Counterfactual Fabrication Lab, a deterministic micro-lab where the correct action is known: do nothing. The lab plants a candidate guardrail for a failure class that provably never occurs, presents only legal episodes, and uses a byte-exact oracle to check every cited violation. The proposer behaves as expected on real violations and abstains on featureless legal input. Yet when the legal input contains a harmless pattern resembling a familiar game rule, it invents a failure: in 15/60 runs, versus 0/60 on featureless input, it enables the nonexistent-rule guardrail and cites a violation the oracle refutes. The effect is structured, not indiscriminate. In single-shot proposals it appears only when three conditions coincide: a rule-shaped pattern, an open-ended rule set and an instruction that presupposes failures. Removing any of these conditions eliminates the fabrication. Because the invented guardrail changes no true outcome and cannot improve an already-perfect suppression score, the phenomenon is neither reward hacking nor over-refusal. It is a phantom guardrail: a fix for a failure that never happened, invisible to suppression-only acceptance. Inside an add-only accept loop it re-enters even without the failure-presupposing instruction, the loop's keep-adding role supplying the demand the instruction supplied in single shot, and once in it stays. We present the Counterfactual Fabrication Lab for measuring fabricated failures in self-improving agent harnesses.

cs.CR↗

arXiv2026

Perception-Aware Time-Optimal Planning for Quadrotor Waypoint Flight

Agile quadrotor flight pushes the limits of control, actuation, and onboard perception. While time-optimal trajectory planning has been extensively studied, existing approaches typically neglect the tight coupling between vehicle dynamics, environmental geometry, and the visual requirements of onboard state estimation. As a result, trajectories that are dynamically feasible may fail in closed-loop execution due to degraded visual quality. This paper introduces a unified time-optimal trajectory optimization framework for vision-based quadrotors that explicitly incorporates perception constraints alongside full nonlinear dynamics, rotor actuation limits, aerodynamic effects, camera field-of-view constraints, and convex geometric gate representations. The proposed formulation solves minimum-time lap trajectories for arbitrary racetracks with diverse gate shapes and orientations, while remaining numerically robust and computationally efficient. We derive an information-theoretic position uncertainty metric to quantify visual state-estimation quality and integrate it into the planner through three perception objectives: position uncertainty minimization, sequential field-of-view constraints, and look-ahead alignment. This enables systematic exploration of the trade-offs between speed and perceptual reliability. To accurately track the resulting perception-aware trajectories, we develop a model predictive contouring tracking controller that separates lateral and progress errors. Experiments demonstrate real-world flight speeds up to 9.8 m/s with 0.07 m average tracking error, and closed-loop success rates improved from 55% to 100% on a challenging Split-S course. The proposed system provides a scalable benchmark for studying the fundamental limits of perception-aware, time-optimal autonomous flight.

cs.RO↗

arXiv2026

"Everyone's using it, but no one is allowed to talk about it": College Students' Experiences Navigating the Higher Education Environment in a Generative AI World

Higher education students are increasingly using generative AI in their academic work. However, existing institutional practices have not yet adapted to this shift. Through semi-structured interviews with 23 college students, our study examines the environmental and social factors that influence students' use of AI. Findings show that institutional pressure factors like deadlines, exam cycles, and grading lead students to engage with AI even when they think it undermines their learning. Social influences, particularly peer micro-communities, establish de-facto AI norms regardless of official AI policies. Campus-wide ``AI shame'' is prevalent, often pushing AI use underground. Current institutional AI policies are perceived as generic, inconsistent, and confusing, resulting in routine noncompliance. Additionally, students develop value-based self-regulation strategies, but environmental pressures create a gap between students' intentions and their behaviors. Our findings show student AI use to be a situated practice, and we discuss implications for institutions, instructors, and system tool designers to effectively support student learning with AI.

cs.CY↗

arXiv2026

Winning in the Limit: Average-Case Committee Selection with Many Candidates

We study the committee selection problem in the canonical impartial culture model with a large number of voters and an even larger candidate set. Here, each voter independently reports a uniformly random preference order over the candidates. For a fixed committee size $k$, we ask when a committee can collectively beat every candidate outside the committee by a prescribed majority level $\alpha$. We focus on two natural notions of collective dominance, $\alpha$-winning and $\alpha$-dominating sets, and we identify sharp threshold phenomena for both of them using probabilistic methods, duality arguments, and rounding techniques. We first consider $\alpha$-winning sets. A set $S$ of $k$ candidates is $\alpha$-winning if, for every outside candidate $a \notin S$, at least an $\alpha$-fraction of voters rank some member of $S$ above $a$. We show a sharp threshold at \[ \alpha_{\mathrm{win}}^\star = 1 - \frac{1}{k}. \] Specifically, an $\alpha$-winning set of size $k$ exists with high probability when $\alpha < \alpha_{\mathrm{win}}^\star$, and is unlikely to exist when $\alpha > \alpha_{\mathrm{win}}^\star$. We then study the stronger notion of $\alpha$-dominating sets. A set $S$ of $k$ candidates is $\alpha$-dominating if, for every outside candidate $a \notin S$, there exists a single committee member $b \in S$ such that at least an $\alpha$-fraction of voters prefer $b$ to $a$. Here we establish an analogous sharp threshold at \[ \alpha_{\mathrm{dom}}^\star = \frac{1}{2} - \frac{1}{2k}. \] As a corollary, our analysis yields an impossibility result for $\alpha$-dominating sets: for every $k$ and every $\alpha > \alpha_{\mathrm{dom}}^\star = 1 / 2 - 1 / (2k)$, there exist preference profiles that admit no $\alpha$-dominating set of size $k$. This corollary improves the best previously known bounds for all $k \geq 2$.

cs.GT↗

arXiv2026

Establishing the $^{40}$Ca$(p,p \alpha)$ reaction at 392 MeV under quasi-free scattering conditions

The $(p,p \alpha)$ reaction offers a direct means to probe preformed $\alpha$-cluster structures in nuclei under quasi-free scattering conditions. Previous studies around 100 MeV provided valuable insights into $\alpha$ clustering, but quantitative comparison with microscopic cluster wave functions remained limited due to strong distortion effects. At higher energies, the reaction mechanism becomes simpler and the distorted-wave impulse approximation (DWIA) provides a more reliable framework for quantitative analysis. In the present work, the $^{40}$Ca$(p,p\alpha)$ reaction was measured at an incident energy of 392 MeV using the high-resolution Grand Raiden and LAS spectrometers at RCNP. Despite the small cross section in this energy region, the achieved resolution allowed clear separation of the ground and excited states of the residual $^{36}$Ar nucleus, and corresponding momentum distributions were extracted. DWIA calculations using a Woods-Saxon $\alpha + ^{36}$Ar bound-state wave function yielded an experimental spectroscopic factor of $ S_{\mathrm{FAC}}^{\mathrm{WS}} = 0.51 \pm 0.05 $, consistent with the previous result at 101.5 MeV $(0.52 \pm 0.23 )$. This agreement demonstrates that the reaction mechanism is well described across a wide energy range. The present study establishes the feasibility of high-precision $(p,p\alpha)$ measurements at several hundred MeV and highlights their potential as a quantitative probe of $\alpha$ clustering in medium-mass nuclei, forming the basis for systematic studies in both stable and unstable systems.

nucl-ex↗

arXiv2025

CrownGen: Patient-customized Crown Generation via Point Diffusion Model

Digital crown design remains a labor-intensive bottleneck in restorative dentistry. We present CrownGen, a generative framework that automates patient-customized crown design using a denoising diffusion model on a novel tooth-level point cloud representation. The system employs two core components: a boundary prediction module to establish spatial priors and a diffusion-based generative module to synthesize high-fidelity morphology for multiple teeth in a single inference pass. We validated CrownGen through a quantitative benchmark on 496 external scans and a clinical study of 26 restoration cases. Results demonstrate that CrownGen surpasses state-of-the-art models in geometric fidelity and significantly reduces active design time. Clinical assessments by trained dentists confirmed that CrownGen-assisted crowns are statistically non-inferior in quality to those produced by expert technicians using manual workflows. By automating complex prosthetic modeling, CrownGen offers a scalable solution to lower costs, shorten turnaround times, and enhance patient access to high-quality dental care.

cs.CV↗

arXiv2025

Bayesian Risk-averse Model Predictive Control with Consistency and Stability Guarantees

Model Predictive Control (MPC) is a powerful framework for constrained control, but its performance and safety can be severely degraded when the prediction model is learned online and thus remains uncertain. In this work, we develop a Bayesian risk-averse MPC framework for stochastic, discrete-time, nonlinear systems that provides theoretical guarantees on the consistency of Bayesian learning and closed-loop stability. First, we study Bayesian learning under the conditionally independent state transitions induced by feedback control and establish explicit conditions for Bayesian consistency on an infinitely countable parameter space. Second, we introduce a general notion of risk-averse asymptotic stability (RAAS), defined via comparison function classes and independent of any specific coherent risk measure or convergence rate, and we derive a risk-averse Lyapunov stability theorem together with MPC-specific stability conditions. Third, building on these foundations, we design a practical Bayesian risk-averse MPC scheme that separates epistemic (parametric) and aleatoric (disturbance) uncertainty: additive disturbances are treated in a risk-neutral fashion, while parametric uncertainty is managed via dynamically shrinking ambiguity sets constructed from Bayesian credible intervals, approximated online using particle filtering. To enable real-time implementation, we propose both an optimal and a sub-optimal receding-horizon control policy, the latter obtained by warm-starting from the previous solution, and prove that asymptotic RAAS is recovered as the Bayesian estimator becomes consistent.

math.OC↗

arXiv2025

Bayesian Risk-Sensitive Policy Optimization For MDPs With General Loss Functions

Motivated by many application problems, we consider Markov decision processes (MDPs) with a general loss function and unknown parameters. To mitigate the epistemic uncertainty associated with unknown parameters, we take a Bayesian approach to estimate the parameters from data and impose a coherent risk functional (with respect to the Bayesian posterior distribution) on the loss. Since this formulation usually does not satisfy the interchangeability principle, it does not admit Bellman equations and cannot be solved by approaches based on dynamic programming. Therefore, We propose a policy gradient optimization method, leveraging the dual representation of coherent risk measures and extending the envelope theorem to continuous cases. We then show the stationary analysis of the algorithm with a convergence rate of $\mathcal{O}(T^{-1/2}+r^{-1/2})$, where $T$ is the number of policy gradient iterations and $r$ is the sample size of the gradient estimator. We further extend our algorithm to an episodic setting, and establish the global convergence of the extended algorithm and provide bounds on the number of iterations needed to achieve an error bound $\mathcal{O}(\epsilon)$ in each episode.

cs.LG↗

arXiv2025

Agentic Aerial Cinematography: From Dialogue Cues to Cinematic Trajectories

We present Agentic Aerial Cinematography: From Dialogue Cues to Cinematic Trajectories (ACDC), an autonomous drone cinematography system driven by natural language communication between human directors and drones. The main limitation of previous drone cinematography workflows is that they require manual selection of waypoints and view angles based on predefined human intent, which is labor-intensive and yields inconsistent performance. In this paper, we propose employing large language models (LLMs) and vision foundation models (VFMs) to convert free-form natural language prompts directly into executable indoor UAV video tours. Specifically, our method comprises a vision-language retrieval pipeline for initial waypoint selection, a preference-based Bayesian optimization framework that refines poses using aesthetic feedback, and a motion planner that generates safe quadrotor trajectories. We validate ACDC through both simulation and hardware-in-the-loop experiments, demonstrating that it robustly produces professional-quality footage across diverse indoor scenes without requiring expertise in robotics or cinematography. These results highlight the potential of embodied AI agents to close the loop from open-vocabulary dialogue to real-world autonomous aerial cinematography.

cs.RO↗

arXiv2025

Multimodal Contrastive Pretraining of CBCT and IOS for Enhanced Tooth Segmentation

Digital dentistry represents a transformative shift in modern dental practice. The foundational step in this transformation is the accurate digital representation of the patient's dentition, which is obtained from segmented Cone-Beam Computed Tomography (CBCT) and Intraoral Scans (IOS). Despite the growing interest in digital dental technologies, existing segmentation methodologies frequently lack rigorous validation and demonstrate limited performance and clinical applicability. To the best of our knowledge, this is the first work to introduce a multimodal pretraining framework for tooth segmentation. We present ToothMCL, a Tooth Multimodal Contrastive Learning for pretraining that integrates volumetric (CBCT) and surface-based (IOS) modalities. By capturing modality-invariant representations through multimodal contrastive learning, our approach effectively models fine-grained anatomical features, enabling precise multi-class segmentation and accurate identification of F\'ed\'eration Dentaire Internationale (FDI) tooth numbering. Along with the framework, we curated CBCT-IOS3.8K, the largest paired CBCT and IOS dataset to date, comprising 3,867 patients. We then evaluated ToothMCL on a comprehensive collection of independent datasets, representing the largest and most diverse evaluation to date. Our method achieves state-of-the-art performance in both internal and external testing, with an increase of 12\% for CBCT segmentation and 8\% for IOS segmentation in the Dice Similarity Coefficient (DSC). Furthermore, ToothMCL consistently surpasses existing approaches in tooth groups and demonstrates robust generalizability across varying imaging conditions and clinical scenarios.

cs.CV↗

arXiv2025

A Privacy-Preserving Framework for Advertising Personalization Incorporating Federated Learning and Differential Privacy

To mitigate privacy leakage and performance issues in personalized advertising, this paper proposes a framework that integrates federated learning and differential privacy. The system combines distributed feature extraction, dynamic privacy budget allocation, and robust model aggregation to balance model accuracy, communication overhead, and privacy protection. Multi-party secure computing and anomaly detection mechanisms further enhance system resilience against malicious attacks. Experimental results demonstrate that the framework achieves dual optimization of recommendation accuracy and system efficiency while ensuring privacy, providing both a practical solution and a theoretical foundation for applying privacy protection technologies in advertisement recommendation.

cs.CR↗

arXiv2025

Graph Neural Network Enhanced Sequential Recommendation Method for Cross-Platform Ad Campaign

In order to improve the accuracy of cross-platform advertisement recommendation, a graph neural network (GNN)- based advertisement recommendation method is analyzed. Through multi-dimensional modeling, user behavior data (e.g., click frequency, active duration) reveal temporal patterns of interest evolution, ad content (e.g., type, tag, duration) influences semantic preferences, and platform features (e.g., device type, usage context) shape the environment where interest transitions occur. These factors jointly enable the GNN to capture the latent pathways of user interest migration across platforms. The experimental results are based on the datasets of three platforms, and Platform B reaches 0.937 in AUC value, which is the best performance. Platform A and Platform C showed a slight decrease in precision and recall with uneven distribution of ad labels. By adjusting the hyperparameters such as learning rate, batch size and embedding dimension, the adaptability and robustness of the model in heterogeneous data are further improved.

cs.LG↗

arXiv2025

Suppression and Regulation of Thermal Birefringence in Optical Voltage Sensor with Isomerism Electrodes and Arbitrary Electric Field Direction Modulation

The insufficient stability and reliability of Optical Voltage Sensor is primarily caused by thermal stress induced birefringence. In this paper, a method based on arbitrary electric field direction modulation and isomerism electrodes is proposed to suppress or regulate it. With the aid of multi-physics Finite Element Method, Jones Matrix and the theory of photoelastic effect, it is found that metal or transparent isomerism electrodes can generate a special thermal stress distribution, which regulates the birefringence in the optical path and their induced measurement error. The experiment is conducted on a 10mm cubic bismuth germanite crystal, with cutting directions 110, -110 and 001. The experiment result shows that Cu isomerism electrodes with electric field angle of 59.9 degrees could generate 37% less birefringence error compared to parallel plate electrodes, in the temperature range from 25 degrees Celsius to 40 degrees Celsius. However, the Indium Tin Oxide electrodes with field angle of 29.6 degrees produces approximately 7 times error because of its bad ductility and thermal conduction. The proposed modeling and suppression method for birefringence is beneficial to design of high accuracy optical voltage sensor or electro-optical modulator.

eess.SY↗

arXiv2024

Liner Shipping Network Design with Reinforcement Learning

This paper proposes a novel reinforcement learning framework to address the Liner Shipping Network Design Problem (LSNDP), a challenging combinatorial optimization problem focused on designing cost-efficient maritime shipping routes. Traditional methods for solving the LSNDP typically involve decomposing the problem into sub-problems, such as network design and multi-commodity flow, which are then tackled using approximate heuristics or large neighborhood search (LNS) techniques. In contrast, our approach employs a model-free reinforcement learning algorithm on the network design, integrated with a heuristic-based multi-commodity flow solver, to produce competitive results on the publicly available LINERLIB benchmark. Additionally, our method also demonstrates generalization capabilities by producing competitive solutions on the benchmark instances after training on perturbed instances.

cs.AI↗

arXiv2024

Time-Optimal Planning for Long-Range Quadrotor Flights: An Automatic Optimal Synthesis Approach

Time-critical tasks such as drone racing typically cover large operation areas. However, it is difficult and computationally intensive for current time-optimal motion planners to accommodate long flight distances since a large yet unknown number of knot points is required to represent the trajectory. We present a polynomial-based automatic optimal synthesis (AOS) approach that can address this challenge. Our method not only achieves superior time optimality but also maintains a consistently low computational cost across different ranges while considering the full quadrotor dynamics. First, we analyze the properties of time-optimal quadrotor maneuvers to determine the minimal number of polynomial pieces required to capture the dominant structure of time-optimal trajectories. This enables us to represent substantially long minimum-time trajectories with a minimal set of variables. Then, a robust optimization scheme is developed to handle arbitrary start and end conditions as well as intermediate waypoints. Extensive comparisons show that our approach is faster than the state-of-the-art approach by orders of magnitude with comparable time optimality. Real-world experiments further validate the quality of the resulting trajectories, demonstrating aggressive time-optimal maneuvers with a peak velocity of 8.86 m/s.

cs.RO↗

arXiv2024

Reusing Historical Trajectories in Natural Policy Gradient via Importance Sampling: Convergence and Convergence Rate

Reinforcement learning provides a mathematical framework for learning-based control, whose success largely depends on the amount of data it can utilize. The efficient utilization of historical trajectories obtained from previous policies is essential for expediting policy optimization. Empirical evidence has shown that policy gradient methods based on importance sampling work well. However, existing literature often neglect the interdependence between trajectories from different iterations, and the good empirical performance lacks a rigorous theoretical justification. In this paper, we study a variant of the natural policy gradient method with reusing historical trajectories via importance sampling. We show that the bias of the proposed estimator of the gradient is asymptotically negligible, the resultant algorithm is convergent, and reusing past trajectories helps improve the convergence rate. We further apply the proposed estimator to popular policy optimization algorithms such as trust region policy optimization. Our theoretical results are verified on classical benchmarks.

cs.LG↗