Reformulating Queries: Theory and Practice
Authors: Michael Benedikt, Egor V. Kostylev, Fabio Mogavero, Efthymia Tsamoura
IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We present a classification of the complexity of the problem, then provide algorithms for solving the problems in practice and evaluate their performance.Our work is the first look at query reformulation over a sub-vocabulary for logics that include disjunction in arbitrary arity, either either from the point of view of complexity or experimental evaluation.We report the results of our experiments with more than 3600 runs in our infrastructure, more than 600 for each of the six classes for the canonical target reformulation. |
| Researcher Affiliation | Academia | Michael Benedikt University of Oxford Egor V. Kostylev University of Oxford Fabio Mogavero University of Oxford Efthymia Tsamoura Alan Turing Institute & University of Oxford |
| Pseudocode | No | The paper describes algorithms in text and mathematical notation but does not include structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | [Benedikt et al., 2017] Michael Benedikt, Egor V. Kostylev, Fabio Mogavero, and Efthymia Tsamoura. Query reformulation: Theory and practice, 2017. Available at www.github.com/qreform/ijcai17qreform. |
| Open Datasets | No | The generator has two parts. The first part randomly generates a canonical target reformulation Q in one of the following six classes: arbitrary or positive propositional formulas either in CNF, or in DNF, or which are alternations of conjunctions and disjunctions of literals with alternation depth varying from 2 to 5. The second part takes reformulation Q as input and outputs a CQ Q, disjunctive Horn formulas Σ, and sub-vocabulary V such that Q is a reformulation of Q over Σ in V. |
| Dataset Splits | No | The paper describes a synthetic data generation process and reports on experimental runs, but does not specify training, validation, or test dataset splits. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., CPU, GPU models, or cloud instance types) used for running its experiments. |
| Software Dependencies | Yes | native interpolation of Vampire 4.1 for CASC J8 |
| Experiment Setup | No | The paper describes the generation of synthetic data and some control parameters (e.g., 'noise clauses'), but it does not provide specific experimental setup details such as hyperparameter values or system-level training configurations. |