When a language model is asked whether its own answer is correct, three languages are involved: that of the question, that requested for the answer, and that of the verifier instruction asking for the judgment. Multilingual confidence evaluations seldom separate the three. We vary them independently in a six-language f...
J. Yeom, Jaewon Sok, Seonghyeon Park et al.· 0 citations
This paper first classify memorization- and generalization-heavy models using grokking in modular addition and compare their responses to unlearning, showing that the latter suffer greater retain damage, i.e., a larger performance drop on the retain set.
Hwiyeong Lee, H. Lim, Ingyu Bang et al.· 0 citations
Masked diffusion language models revise many masked output positions in parallel and call a response answer-first when the final answer commits before the reasoning printed ahead of it, which limits how far the intervention result can be generalized.
J. Yeom, Jaewon Sok, Seonghyeon Park et al.· 3 citations
These results reframe existing window-style samplers, previously motivated by efficiency, as the minimal fix for a reasoning pathology they were never designed to address.
J. Yeom, Jaewon Sok, Seonghyeon Park et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.