Poster · Paris

Confidence-Based Decoding is Provably Efficient for Diffusion Language Models

Name: Confidence-Based Decoding is Provably Efficient for Diffusion Language Models
Start: 2026-12-11T14:30:00+01:00
End: 2026-12-11T16:30:00+01:00
Location: Paris Poster Hall, Paris

When: Fri 11 Dec 2026, 14:30–16:30 CET
Where: Paris Poster Hall
Session: Paris Poster Session 6
Authors: Changxiao Cai, Gen Li
Organisations: University of Michigan, The Chinese University of Hong Kong
QRS classification: Language models, Generative models

Open in schedule Official page