Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Fully Dynamic Algorithms for Chamfer Distance
Authors: Gramoz Goranci, Shaofeng Jiang, Peter Kiss, Eva Szilagyi, Qiaoyuan Yang
NeurIPS 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate our method on real-world datasets and demonstrate that it performs competitively against natural baselines. ... We implement our algorithm and validate its performance over four real datasets covering both high and low dimensions, as well as their noisy versions with injected outlier points. In all datasets, our algorithm achieves less than 10% error using only hundreds of samples, even against injected outliers, in time up to magnitudes better than a naive dynamic algorithm. |
| Researcher Affiliation | Academia | Gramoz Goranci Faculty of Computer Science University of Vienna, Austria EMAIL Shaofeng H.-C. Jiang School of Computer Science Peking University, China EMAIL Peter Kiss Faculty of Computer Science University of Vienna, Austria EMAIL Eva Szilagyi Faculty of Computer Science Uni Vie Doctoral School Computer Science Do CS University of Vienna, Austria EMAIL Qiaoyuan Yang School of Computer Science Peking University, China EMAIL |
| Pseudocode | Yes | C Pseudo-code of Our Algorithm In this section, we give pseudo-codes for the dynamic algorithm described in Section 3. Algorithm 1 UPDATE POINT(x) Algorithm 2 SAMPLER(A) procedure Algorithm 3 SAMPLER-FIND(v) procedure Algorithm 4 QUERY-CHAMFER(A, B, ϵ, α) procedure |
| Open Source Code | Yes | The supplementary material includes clear instructions and all necessary resources, including open access to the datasets and code. ... The code is included as part of the supplementary material, and all datasets used in the experiments are publicly available online. |
| Open Datasets | Yes | We employ four real datasets covering both high and low dimensions in the experiment: Text Embedding [KSKW15], Shape Net [CFG+15], Fashion MNIST [XRV17], and SIFT [JDS11]. |
| Dataset Splits | Yes | Finally, since these datasets do not contain information of dynamic update, we employ a sliding window (on B) to simulate the insertions and deletions. The detailed specification of the dataset and the experiment parameters can be found in Table 1. |
| Hardware Specification | Yes | All the experiments are run on a Mac Book Air 15.3 with an Apple M3 chip (8 cores, 2.22 GHz), 16GB RAM, and mac OS 14.4.1 (23E224). |
| Software Dependencies | Yes | All algorithms are implemented in C++ and compiled with Apple Clang version 15.0.0 at -O3 optimization level. ... We do not use any external library in the code. |
| Experiment Setup | Yes | The detailed specification of the dataset and the experiment parameters can be found in Table 1. ... Specifically, we estimate dist CH(A, B) through taking m = 240 L log n max{α2, 1} ϵ 2 = O(log2 n max{α2, 1} ϵ 2) samples S from A through the after-mentioned sampling procedure. For each a S, we then query the nearest neighbor data structure to generate a (1 + α/4)-approximation to dist CH(a, B). ... Since the window size for each dataset is relatively small, we choose to use the exact nearest neighbor algorithm, which does not introduce additional errors and allows for a more accurate evaluation. For the 3D Shape Net dataset, we implement nearest neighbor queries with KD-trees, which can efficiently perform exact search in low-dimensional spaces. |