5 papers
Isolating a Language Model's Judgment with a Label-Free Subspace
Chaemin Jang, Donggyu Lee, Jeesoo Kim +2
Language models are often asked to make judgments: whether a proposed answer is correct, whether one statement follows logically from another, or whether a response is safe. Yet ch…
Learning a Fact Is Not Learning How to Retrieve It
Chaemin Jang, Jihee Kim, Dongman Lee
A model trained on "The capital of X is Y" may produce "Y" after "The capital of X is" but fail after "The capital of X:." We call these different ways of eliciting the same fact r…
From Simulated Citizens to Simulated Deliberation: Challenges in Representation and Interaction
Chaemin Jang, Junsik Min, Jaewoo Choi +10
Can LLM agents simulate how a real public would deliberate? For such simulations to be meaningful, agents must reflect the population's opinions, and their views must be shaped by…
Instruction-Tuned Language Models Cannot Sample from Distributions They Can Describe
Chaemin Jang, Dongman Lee, Jihee Kim
Silicon sampling uses language models as proxies for human survey respondents, treating each model call as an independent draw from the persona's response distribution. We show thi…
In-Context Examples Suppress Scientific Knowledge Recall in LLMs
Chaemin Jang, Woojin Park, Hyeok Yun +2
Scientific reasoning rarely stops at what is directly observable; it often requires uncovering hidden structure from data. From estimating reaction constants in chemistry to inferr…