The geometric placement of synthetic augmentation examples in embedding space significantly affects discourse-pragmatic classification performance, with near-core examples yielding the largest macro-F gains but no improvement in AUC.
Evaluating Grammatical Well-Formedness in Large Language Models: A Comparative Study with Human Judgments
3 Pith papers cite this work, alongside 5 external citations. Polarity classification is still indexing.
3
Pith papers citing it
5
external citations · Crossref
fields
cs.CL 3years
2026 3representative citing papers
LLMs recover dominant binomial orders from corpora but align less closely with exact preference distributions, with preference strength partially encoded in middle-to-late layers and manipulable via steering.
LLMs show strong linguistic performance but use distinct processing methods from humans and are less efficient at learning and generalization for reasoning tasks.
citing papers explorer
-
The Impact of Synthetic Data Augmentation on Discourse-Pragmatic Function Classification
The geometric placement of synthetic augmentation examples in embedding space significantly affects discourse-pragmatic classification performance, with near-core examples yielding the largest macro-F gains but no improvement in AUC.
-
Behavioral and Representational Evidence of Binomial Ordering Preferences in Large Language Models
LLMs recover dominant binomial orders from corpora but align less closely with exact preference distributions, with preference strength partially encoded in middle-to-late layers and manipulable via steering.
-
LLM and Human Modes of Representation
LLMs show strong linguistic performance but use distinct processing methods from humans and are less efficient at learning and generalization for reasoning tasks.