GPT-3 shows that scaling an autoregressive language model to 175 billion parameters enables strong few-shot performance across diverse NLP tasks via in-context prompting without fine-tuning.
Probing neural network comprehension of natural language arguments
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
representative citing papers
GRASP aggregates stable local LLM interaction judgments into global argument rankings via a convergent attack-defense propagation operator on interaction graphs, yielding higher reproducibility than holistic judging and no correlation with human convincingness.
idSCD uses semantic correlation descriptors to perform dataset membership inference by comparing learned semantic structures, outperforming baselines in NLI, emotion, and medical text experiments.
citing papers explorer
-
GRASP: Deterministic argument ranking in interaction graphs
GRASP aggregates stable local LLM interaction judgments into global argument rankings via a convergent attack-defense propagation operator on interaction graphs, yielding higher reproducibility than holistic judging and no correlation with human convincingness.
-
idSCD: Identifying Training Datasets through Semantic Correlation Descriptors
idSCD uses semantic correlation descriptors to perform dataset membership inference by comparing learned semantic structures, outperforming baselines in NLI, emotion, and medical text experiments.