arXiv subjects

Yucheng Sun

Publications and source records attributed to Yucheng Sun.

6 recordsLinked to original sources

arXiv2025

Skyshield: Event-Driven Submillimetre Thin Obstacle Detection for Drone Flight Safety

Drones operating in complex environments face a significant threat from thin obstacles, such as steel wires and kite strings at the submillimeter level, which are notoriously difficult for conventional sensors like RGB cameras, LiDAR, and depth cameras to detect. This paper introduces SkyShield, an event-driven, end-to-end framework designed for the perception of submillimeter scale obstacles. Drawing upon the unique features that thin obstacles present in the event stream, our method employs a lightweight U-Net architecture and an innovative Dice-Contour Regularization Loss to ensure precise detection. Experimental results demonstrate that our event-based approach achieves mean F1 Score of 0.7088 with a low latency of 21.2 ms, making it ideal for deployment on edge and mobile platforms.

cs.CV↗

arXiv2025

Probing for Arithmetic Errors in Language Models

We investigate whether internal activations in language models can be used to detect arithmetic errors. Starting with a controlled setting of 3-digit addition, we show that simple probes can accurately decode both the model's predicted output and the correct answer from hidden states, regardless of whether the model's output is correct. Building on this, we train lightweight error detectors that predict model correctness with over 90% accuracy. We then extend our analysis to structured chain-of-thought traces on addition-only GSM8K problems and find that probes trained on simple arithmetic generalize well to this more complex setting, revealing consistent internal representations. Finally, we demonstrate that these probes can guide selective re-prompting of erroneous reasoning steps, improving task accuracy with minimal disruption to correct outputs. Our findings suggest that arithmetic errors can be anticipated from internal activations alone, and that simple probes offer a viable path toward lightweight model self-correction.

cs.CL↗

arXiv2025

Increasing Computation Resolves Conflicts in Vision Language Models

Cognitive control, the ability to coordinate competing information sources in pursuit of goals, is fundamental to intelligent behavior. We systematically investigate whether Vision Language Models (VLMs) exhibit cognitive control and how computational resources modulate conflict resolution. We construct a benchmark of 4,410 tasks across seven conflict paradigms (Stroop, Flanker, and five realistic variants) spanning multiple difficulty levels and visual complexities, testing 47 VLMs with rigorous experimental control. We find that VLMs exhibit robust congruency effects across all tasks, with larger models systematically resolving conflicts more effectively than smaller models. Critically, VLMs reproduce the fine-grained demand-resource relationship observed in human temporal dynamics: larger models drop below chance on incongruent high-conflict trials while smaller models fail to meaningfully engage and perform at chance, mirroring human behavior at short processing times and establishing parameter count as a proxy for conflict resolution capacity. These findings demonstrate that human-like cognitive control emerges from optimization dynamics in large-scale neural networks, suggesting that adaptive flexibility under conflict may naturally arise through scaling.

cs.NE↗

arXiv2023

Private Distribution Testing with Heterogeneous Constraints: Your Epsilon Might Not Be Mine

Private closeness testing asks to decide whether the underlying probability distributions of two sensitive datasets are identical or differ significantly in statistical distance, while guaranteeing (differential) privacy of the data. As in most (if not all) distribution testing questions studied under privacy constraints, however, previous work assumes that the two datasets are equally sensitive, i.e., must be provided the same privacy guarantees. This is often an unrealistic assumption, as different sources of data come with different privacy requirements; as a result, known closeness testing algorithms might be unnecessarily conservative, "paying" too high a privacy budget for half of the data. In this work, we initiate the study of the closeness testing problem under heterogeneous privacy constraints, where the two datasets come with distinct privacy requirements. We formalize the question and provide algorithms under the three most widely used differential privacy settings, with a particular focus on the local and shuffle models of privacy; and show that one can indeed achieve better sample efficiency when taking into account the two different "epsilon" requirements.

cs.DS↗

arXiv2022

The Tight Spanning Ratio of the Rectangle Delaunay Triangulation

Spanner construction is a well-studied problem and Delaunay triangulations are among the most popular spanners. Tight bounds are known if the Delaunay triangulation is constructed using an equilateral triangle, a square, or a regular hexagon. However, all other shapes have remained elusive. In this paper, we extend the restricted class of spanners for which tight bounds are known. We prove that Delaunay triangulations constructed using rectangles with aspect ratio $A$ have spanning ratio at most $\sqrt{2} \sqrt{1+A^2 + A \sqrt{A^2 + 1}}$, which matches the known lower bound.

cs.CG↗

arXiv2022

Optimal Closeness Testing of Discrete Distributions Made (Complex) Simple

In this note, we revisit the recent work of Diakonikolas, Gouleakis, Kane, Peebles, and Price (2021), and provide an alternative proof of their main result. Our argument does not rely on any specific property of Poisson random variables (such as stability and divisibility) nor on any "clever trick," but instead on an identity relating the expectation of the absolute value of any random variable to the integral of its characteristic function: \[ \mathbb{E}[|X|] = \frac{2}π\int_0^\infty \frac{1-\Re(\mathbb{E}[e^{i tX}])}{t^2}\, dt \] Our argument, while not devoid of technical aspects, is arguably conceptually simpler and more general; and we hope this technique can find additional applications in distribution testing.

cs.DS↗