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.
Counterplanning using Goal Recognition and Landmarks
Authors: Alberto Pozanco, Yolanda E-Martín, Susana Fernández, Daniel Borrajo
IJCAI 2018 | Venue PDF | LLM Run Details | Input Tokens: 16,843 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 4,729 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 | Experimental results in several domains show the benefits of our novel approach. |
| Researcher Affiliation | Academia | Alberto Pozanco, Yolanda E-Mart ın, Susana Fern andez, Daniel Borrajo Departamento de Inform atica, Universidad Carlos III de Madrid Avda. de la Universidad, 30. 28911 Legan es, Madrid, Spain |
| Pseudocode | Yes | Algorithm 1 DOMAIN-INDEPENDENT COUNTERPLANNING |
| Open Source Code | No | The paper does not provide any explicit statements about releasing source code for the methodology described, nor does it include links to a code repository. |
| Open Datasets | Yes | We empirically evaluate our approach on the new previously described TERRORIST domain as well as in other domains usually used in goal recognition works such as LOGISTICS, EASY IPC GRID, BLOCKS, and INTRUSION DETECTION. |
| Dataset Splits | No | The paper mentions generating '10 random problems' for each domain and varying 'observed actions' percentage, but it does not specify explicit training, validation, or test dataset splits needed for reproduction. |
| Hardware Specification | Yes | All the experiments were ran on a Ubuntu machine with Intel Core 2 Quad Q8400 running at 2.66 GHz. |
| Software Dependencies | No | The paper mentions using specific planners like 'HSP*f [Haslum, 2008]' and 'LAMA [Richter et al., 2011]', but it does not provide specific version numbers for these software components. |
| Experiment Setup | Yes | The planning times for all the planners were set to 1800 seconds. In all the domains πφ is computed using GREEDY LAMA. For optimal plan computations of FCL, we use HSP*f. The set of candidate goals Gφ always consists of a 20% of all the possible goals in each problem. The set of observed actions was taken to be a subset of the plan solution πφ, ranging from 10% of the actions, up to 70% of the actions. |