The End of the Guessing Game? Why Describing Data Beats Estimating It
1 Jul 2025
CLAIM surpasses statistical and ML methods by using LLMs for contextual imputation, though future work must address scalability and domain specificity.
It's Not Just What's Missing, It's How You Say It: CLAIM's Winning Formula
1 Jul 2025
Experiments show CLAIM outperforms baselines like k-NN and MICE across all missingness patterns, with context-specific descriptors proving most effective.
Teaching AI to Say "I Don't Know": A Four-Step Guide to Contextual Data Imputation
1 Jul 2025
CLAIM converts tabular data to natural language, then uses an LLM to generate contextual text descriptors for missing values to improve downstream tasks.
CLAIM: A Contextual Language Model for Accurate Imputation of Missing Tabular Data
1 Jul 2025
CLAIM uses LLMs to fill missing tabular data with contextual text, outperforming traditional methods and improving downstream task accuracy.