4 papers
Taming Variability: Randomized and Bootstrapped Conformal Risk Control for LLMs
Lingyou Pang, Lei Huang, Jianyu Lin +3
We transform the randomness of LLMs into precise assurances using an actuator at the API interface that applies a user-defined risk constraint in finite samples via Conformal Risk…
Unsupervised Conformal Inference: Bootstrapping and Alignment to Control LLM Uncertainty
Lingyou Pang, Lei Huang, Jianyu Lin +4
Deploying black-box LLMs requires managing uncertainty in the absence of token-level probability or true labels. We propose introducing an unsupervised conformal inference framewor…
LLM Web Dynamics: Tracing Model Collapse in a Network of LLMs
Tianyu Wang, Akira Horiguchi, Lingyou Pang +1
The increasing use of synthetic data from the public Internet has enhanced data usage efficiency in large language model (LLM) training. However, the potential threat of model coll…
Investigating social alignment via mirroring in a system of interacting language models
Harvey McGuinness, Tianyu Wang, Carey E. Priebe +1
Alignment is a social phenomenon wherein individuals share a common goal or perspective. Mirroring, or mimicking the behaviors and opinions of another individual, is one mechanism…