Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
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.
Path Planning with CPD Heuristics
Authors: Massimo Bono, Alfonso E. Gerevini, Daniel D. Harabor, Peter J. Stuckey
IJCAI 2019 | Venue PDF | LLM Run Details | Input Tokens: 16,254 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 4,731 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate CPD-Search in two online settings: optimal search and anytime search. In both experiments we use grid benchmarks drawn from Sturtevant s well known repository at http://movingai.com. |
| Researcher Affiliation | Academia | 1Dipartimento di Ingegneria dell Informazione, Universit a degli Studi di Brescia, Brescia, Italy 2Faculty of Information Technology, Monash University, Melbourne, Australia |
| Pseudocode | Yes | Algorithm 1: CPD-Search(w, w , s, t, ϵ): Variant of Weighted A* with a CPD heuristic. Algorithm 2: w CPD(s,t): Retrieving the original (using w) and new (using w ) cost of the shortest path (according to w) for an (s, t) pair. |
| Open Source Code | Yes | All codes are in C++ and available from https://bitbucket.org/koldar/astar-early-stop/. |
| Open Datasets | Yes | In both experiments we use grid benchmarks drawn from Sturtevant s well known repository at http://movingai.com. |
| Dataset Splits | No | The paper uses grid benchmarks with various instances (start-target pairs) for evaluation but does not specify a training, validation, or test dataset split in the typical machine learning sense. |
| Hardware Specification | Yes | Our test machine is a i78700 machine with 16GB memory. |
| Software Dependencies | No | The paper states 'All codes are in C++', but does not provide specific version numbers for any libraries, frameworks, or other software dependencies. |
| Experiment Setup | Yes | We use a radius r = 15 and a weight multiplication of w'(e) = w(e) (3e^(-x^2/45) + 1) where x is the number of hops from node n to edge e. We implement ALT with different numbers of landmark nodes, up to diminishing returns: 6, 12 and 18. |