11 papers
Reliability-Aware Adaptive Self-Consistency for Efficient Sampling in LLM Reasoning
Junseok Kim, Nakyeong Yang, Kyungmin Min +1
Self-Consistency improves reasoning reliability through multi-sample aggregation, but incurs substantial inference cost. Adaptive self-consistency methods mitigate this issue by ad…
How Training Data Shapes the Use of Parametric and In-Context Knowledge in Language Models
Minsung Kim, Dong-Kyum Kim, Jea Kwon +3
Large language models leverage both parametric knowledge acquired during pretraining and in-context knowledge provided at inference time. Crucially, when these sources conflict, mo…
Rethinking Post-Unlearning Behavior of Large Vision-Language Models
Minsung Kim, Nakyeong Yang, Kyomin Jung
Large Vision-Language Models (LVLMs) can recognize individuals in images and disclose sensitive personal information about them, raising critical privacy concerns. Machine unlearni…
Erase or Hide? Suppressing Spurious Unlearning Neurons for Robust Unlearning
Nakyeong Yang, Dong-Kyum Kim, Jea Kwon +3
Large language models trained on web-scale data can memorize private or sensitive knowledge, raising significant privacy risks. Although some unlearning methods mitigate these risk…
Bilinear representation mitigates reversal curse and enables consistent model editing
Dong-Kyum Kim, Minsung Kim, Jea Kwon +2
The reversal curse--a language model's inability to infer an unseen fact "B is A" from a learned fact "A is B"--is widely considered a fundamental limitation. We show that this is…
Persona Switch: Mixing Distinct Perspectives in Decoding Time
Junseok Kim, Nakyeong Yang, Kyomin Jung
Role-play prompting is known to steer the behavior of language models by injecting a persona into the prompt, improving their zero-shot reasoning capabilities. However, such improv…