A retrieval method that uses semantically similar wrong answers as anchors to find corrective correct demonstrations improves few-shot in-context learning accuracy on seven reasoning datasets.
Advances in neural information processing systems 35, 22199–22213 (2022) Title Suppressed Due to Excessive Length 9
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Failures Are the Stepping Stones to Success: Enhancing Few-Shot In-Context Learning by Leveraging Negative Samples
A retrieval method that uses semantically similar wrong answers as anchors to find corrective correct demonstrations improves few-shot in-context learning accuracy on seven reasoning datasets.