Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting

Record

Authors
Turpin, M.; Michael, J.; Perez, E.; Bowman, S.
Year
2023
Title
Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting
Published in
neural information processing systems
Type
Conference paper
DOI
10.52202/075280-3275
arXiv
arXiv:2305.04388
Links
https://arxiv.org/abs/2305.04388
Our note
none yet
Verification
verified: its identifier resolves to a record with the same title and author (checked against OpenAlex, 2026-11-14).
Identifiers found: OpenAlex W7133228402 · DOI 10.52202/075280-3275
From the lookup
doi, title, container (the pages' own wording is under "As cited" below)
Id
turpin-2023-language-models-don-t-always-say

The pages disagree on container: container NeurIPS 36 or arXiv. The record above takes the most common value.

Cited on

As cited

The pages cite this work in 2 forms.

  • Turpin, M., Michael, J., Perez, E., and Bowman, S. (2023). Language models don't always say what they think: Unfaithful explanations in chain-of-thought prompting. NeurIPS 36 (cited by the card; not read for this note).
  • Turpin, M., Michael, J., Perez, E., and Bowman, S. R. (2023). Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting. arXiv:2305.04388. https://arxiv.org/abs/2305.04388

BibLaTeX

@inproceedings{turpin-2023-language-models-don-t-always-say,
  author = {Turpin, M. and Michael, J. and Perez, E. and Bowman, S.},
  title = {{Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting}},
  booktitle = {neural information processing systems},
  date = {2023},
  doi = {10.52202/075280-3275},
  eprint = {2305.04388},
  eprinttype = {arxiv}
}

Download · JSON with CSL-JSON · the whole bibliography