The ACS benchmark scores models on 11 language pairs by asking which sentence in a minimal pair is the naturally occurring code-switch; humans and larger LLMs tend to agree.
InInterspeech, pages 1938–1942
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
-
Minimal Pair-Based Evaluation of Code-Switching
The ACS benchmark scores models on 11 language pairs by asking which sentence in a minimal pair is the naturally occurring code-switch; humans and larger LLMs tend to agree.