Post-training reduces LLMs' behavioral alignment with humans across families and sizes, with the misalignment increasing in newer generations while persona induction fails to improve individual-level predictions.
Lin,et al., Mitigating the alignment tax of rlhf, inProceedings of the 2024 Conference on Empirical Methods in Natural Language Processing(2024), pp
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Post-training makes large language models less human-like
Post-training reduces LLMs' behavioral alignment with humans across families and sizes, with the misalignment increasing in newer generations while persona induction fails to improve individual-level predictions.