Fine-tuning LLMs on observational logs can amplify spurious patterns such as weekday effects, and a confounder-subtraction method, DeconfoundLM, restores causal signal in self-built simulations.
Frontiers: Can large language models capture human prefer- ences? Marketing Science, 43(4):709–722, 2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Aligning Language Models with Observational Data: Opportunities and Risks from a Causal Perspective
Fine-tuning LLMs on observational logs can amplify spurious patterns such as weekday effects, and a confounder-subtraction method, DeconfoundLM, restores causal signal in self-built simulations.