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..
Procedural Generation of Initial States of Sokoban
Authors: Dâmaris S. Bento, André G. Pereira, Levi H. S. Lelis
IJCAI 2019 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments show that β is able to generate initial states that are harder to solve by a specialized solver than those designed by human experts. |
| Researcher Affiliation | Academia | Dˆamaris S. Bento,1 Andr e G. Pereira2 and Levi H. S. Lelis1 1Universidade Federal de Vic osa, Brazil 2Universidade Federal do Rio Grande do Sul, Brazil EMAIL, EMAIL |
| Pseudocode | No | The paper describes the search algorithm in prose but does not include any pseudocode or a clearly labeled algorithm block. |
| Open Source Code | No | The paper does not contain any statement about making its source code publicly available or provide a link to a code repository. |
| Open Datasets | No | The paper states 'We use the 90 problems of the x Sokoban benchmark in our experiments.' However, it does not provide concrete access information (link, DOI, or formal citation with authors/year) for this benchmark. |
| Dataset Splits | No | The paper does not provide specific details about training, validation, or test dataset splits, percentages, or methodology for data partitioning. |
| Hardware Specification | Yes | All experiments are run on 2.66 GHz CPUs, β is allowed 1 hour of computation time and 8 GB of memory, while the solver is allowed 10 minutes and 4 GB (more memory and running time have a small impact on PRB). |
| Software Dependencies | No | The paper mentions using a 'solver introduced by Pereira et al. [2015], which we refer to as PRB', but it does not provide specific version numbers for any software dependencies. |
| Experiment Setup | Yes | All experiments are run on 2.66 GHz CPUs, β is allowed 1 hour of computation time and 8 GB of memory, while the solver is allowed 10 minutes and 4 GB (more memory and running time have a small impact on PRB). Due to the randomness of the PDBs, we report the average results of 5 runs of each approach. We instantiate variants of β by varying the ordering [ ] in which the states are expanded and how initial states are selected. |