Joe Stacey
YOU?
Author Swipe
View article: PyTOD: Programmable Task-Oriented Dialogue with Execution Feedback
PyTOD: Programmable Task-Oriented Dialogue with Execution Feedback Open
Programmable task-oriented dialogue (TOD) agents enable language models to follow structured dialogue policies, but their effectiveness hinges on accurate state tracking. We present PyTOD, an agent that generates executable code to track d…
View article: How to Improve the Robustness of Closed-Source Models on NLI
How to Improve the Robustness of Closed-Source Models on NLI Open
Closed-source Large Language Models (LLMs) have become increasingly popular, with impressive performance across a wide range of natural language tasks. These models can be fine-tuned to further improve performance, but this often results i…
View article: LUCID: LLM-Generated Utterances for Complex and Interesting Dialogues
LUCID: LLM-Generated Utterances for Complex and Interesting Dialogues Open
Spurred by recent advances in Large Language Models (LLMs), virtual assistants are poised to take a leap forward in terms of their dialogue capabilities. Yet a major bottleneck to achieving genuinely transformative task-oriented dialogue c…
View article: Distilling Robustness into Natural Language Inference Models with Domain-Targeted Augmentation
Distilling Robustness into Natural Language Inference Models with Domain-Targeted Augmentation Open
Knowledge distillation optimises a smaller student model to behave similarly to a larger teacher model, retaining some of the performance benefits. While this method can improve results on in-distribution examples, it does not necessarily …
View article: Atomic Inference for NLI with Generated Facts as Atoms
Atomic Inference for NLI with Generated Facts as Atoms Open
With recent advances, neural models can achieve human-level performance on various natural language tasks. However, there are no guarantees that any explanations from these models are faithful, i.e. that they reflect the inner workings of …
View article: Distilling Robustness into Natural Language Inference Models with Domain-Targeted Augmentation
Distilling Robustness into Natural Language Inference Models with Domain-Targeted Augmentation Open
Knowledge distillation optimises a smaller student model to behave similarly to a larger teacher model, retaining some of the performance benefits. While this method can improve results on in-distribution examples, it does not necessarily …
View article: When and Why Does Bias Mitigation Work?
When and Why Does Bias Mitigation Work? Open
Neural models have been shown to exploit shallow surface features to perform language understanding tasks, rather than learning the deeper language understanding and reasoning skills that practitioners desire. Previous work has developed d…
View article: Supervising Model Attention with Human Explanations for Robust Natural Language Inference
Supervising Model Attention with Human Explanations for Robust Natural Language Inference Open
Natural Language Inference (NLI) models are known to learn from biases and artefacts within their training data, impacting how well they generalise to other unseen datasets. Existing de-biasing approaches focus on preventing the models fro…
View article: Logical Reasoning with Span-Level Predictions for Interpretable and Robust NLI Models
Logical Reasoning with Span-Level Predictions for Interpretable and Robust NLI Models Open
Current Natural Language Inference (NLI) models achieve impressive results, sometimes outperforming humans when evaluating on in-distribution test sets. However, as these models are known to learn from annotation artefacts and dataset bias…
View article: Logical Reasoning with Span-Level Predictions for Interpretable and Robust NLI Models
Logical Reasoning with Span-Level Predictions for Interpretable and Robust NLI Models Open
Current Natural Language Inference (NLI) models achieve impressive results, sometimes outperforming humans when evaluating on in-distribution test sets. However, as these models are known to learn from annotation artefacts and dataset bias…
View article: Supervising Model Attention with Human Explanations for Robust Natural Language Inference
Supervising Model Attention with Human Explanations for Robust Natural Language Inference Open
Natural Language Inference (NLI) models are known to learn from biases and artefacts within their training data, impacting how well they generalise to other unseen datasets. Existing de-biasing approaches focus on preventing the models fro…
View article: Natural Language Inference with a Human Touch: Using Human Explanations to Guide Model Attention.
Natural Language Inference with a Human Touch: Using Human Explanations to Guide Model Attention. Open
Natural Language Inference (NLI) models are known to learn from biases and artefacts within their training data, impacting how well the models generalise to other unseen datasets. While previous de-biasing approaches focus on preventing mo…
View article: Avoiding the Hypothesis-Only Bias in Natural Language Inference via\n Ensemble Adversarial Training
Avoiding the Hypothesis-Only Bias in Natural Language Inference via\n Ensemble Adversarial Training Open
Natural Language Inference (NLI) datasets contain annotation artefacts\nresulting in spurious correlations between the natural language utterances and\ntheir respective entailment classes. These artefacts are exploited by neural\nnetworks …
View article: There is Strength in Numbers: Avoiding the Hypothesis-Only Bias in Natural Language Inference via Ensemble Adversarial Training.
There is Strength in Numbers: Avoiding the Hypothesis-Only Bias in Natural Language Inference via Ensemble Adversarial Training. Open
Natural Language Inference (NLI) datasets contain annotation artefacts resulting in spurious correlations between the natural language utterances and their respective entailment classes. These artefacts are exploited by neural networks eve…
View article: Avoiding the Hypothesis-Only Bias in Natural Language Inference via Ensemble Adversarial Training
Avoiding the Hypothesis-Only Bias in Natural Language Inference via Ensemble Adversarial Training Open
Natural Language Inference (NLI) datasets contain annotation artefacts resulting in spurious correlations between the natural language utterances and their respective entailment classes. These artefacts are exploited by neural networks eve…