Zero-Shot Slot Filling with Slot-Prefix Prompting and Attention Relationship Descriptor
Authors: Qiaoyang Luo, Lingqiao Liu
AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | we create a simple-but-effective zero-shot slot filling system that can achieve significantly better performance than the previous methods, as demonstrated by our experimental studies. |
| Researcher Affiliation | Academia | Qiaoyang Luo, Lingqiao Liu* The University of Adelaide {qiaoyang.luo, lingqiao.liu}@adelaide.edu.au |
| Pseudocode | No | The paper includes diagrams and prose descriptions of the method but no structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any explicit statement about releasing source code or a link to a code repository for the methodology described. |
| Open Datasets | Yes | We evaluate the proposed method on SNIPS (Coucke et al. 2018) which is a public spoken language understanding dataset... We also evaluated our model at TOP dataset (Gupta et al. 2018) that is task-oriented semantic parsing dataset with 36 slot types. |
| Dataset Splits | Yes | Following the previous zero-shot evaluation setup from (Liu et al. 2020), we select one domain as the target domain each time while using the other six domains as the source domain. We follow the settings from (Du et al. 2021) to use all seven domains of SNIPS as training data. |
| Hardware Specification | No | The paper mentions using pre-trained language models like BERT and RoBERTa, but it does not specify any hardware details (e.g., GPU/CPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper mentions using BERT and RoBERTa models but does not provide specific version numbers for these or any other software dependencies needed for replication. |
| Experiment Setup | No | The paper describes training settings like 'fine-tuning' and 'prompt tuning' and mentions using 'cross-entropy loss', but it does not provide specific hyperparameter values such as learning rate, batch size, or number of epochs. |