Exemplification Modeling: Can You Give Me an Example, Please?
Authors: Edoardo Barba, Luigi Procopio, Caterina Lacerra, Tommaso Pasini, Roberto Navigli
IJCAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate the produced sentences quantitatively by employing them as additional training examples for the WSD task. The examples generated lead WSD models to perform better than when relying on manually-annotated datasets only, and to attain results higher than the current state of the art. |
| Researcher Affiliation | Academia | 1Sapienza NLP Group, Department of Computer Science, Sapienza University of Rome 2Department of Computer Science, University of Copenhagen |
| Pseudocode | No | No pseudocode or clearly labeled algorithm blocks were found in the paper. Methods are described in prose. |
| Open Source Code | Yes | We release the pretrained model, the datasets and the software at https://github.com/Sapienza NLP/exmod. |
| Open Datasets | Yes | We set k = 1, therefore serving only (D, s) pairs with |D| = 1, and draw the training pairs from the concatenation of Sem Cor and WNG... Word Net [Miller et al., 1990], Syntag Net [Maru et al., 2019], Sem Cor [Miller et al., 1993], Princeton Word Net Gloss Corpus (WNG), The Oxford Dictionary Dataset [Chang et al., 2018]. |
| Dataset Splits | Yes | we define two sets of validation instances for k = 1 and k = 2, with 300 and 580 samples, respectively,10 and calculate this similarity for each pair in each input set D, aggregating them via a macro average. |
| Hardware Specification | No | No specific hardware details (e.g., exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) were provided for running the experiments. |
| Software Dependencies | No | The paper mentions software like BART, RAdam, and BERT, but does not provide specific version numbers for these or other ancillary software components needed for reproduction. |
| Experiment Setup | Yes | We train EXMAKER with RAdam [Liu et al., 2020] for 300,000 training steps, learning rate set to 1e 5 and batches of 800 tokens, accumulating gradient for 10 steps. ... We train the model for at most 50 epochs with early stopping on the validation accuracy and patience set to 3 epochs. |