🧪 Test?View on arXiv
3, 7, 14, 17, 42: Instruction Tuning Installs and Sharpens Number-Choice Heuristics in Language Models
Not provided in the abstract
fine-tuningnumber-choice heuristicslanguage models
2610.36178
Builder Relevance
1h ago70%
Abstract
This study measures the probability distribution of integers assigned by language models when prompted for a random number, revealing the effects of instruction tuning on number-choice heuristics.
Reality Card
Core Claim
Instruction tuning significantly alters the number-choice heuristics in language models, leading to a preference for odd numbers and avoidance of round numbers.
Method / Result
15 out of 18 evaluated model pairs showed that instruction tuning made the distribution of chosen numbers less uniform.
Limitations
The study does not provide specific author names or detailed methodology, which may limit reproducibility.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.
No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.