BumJun Kim

2 accepted papers

2026

Dependency-Aware Parallel Decoding via Attention for Diffusion LLMs

Bumjun Kim, Dongjae Jeon, Moongyu Jeon, Albert No

ICML 2026poster

Parallel decoding for diffusion LLMs (dLLMs) is difficult because each denoising step provides only token-wise marginal distributions, while unmasking multiple tokens simultaneously requires accounting for inter-token dependencies. We propose Dependency-Aware Parallel Decoding (DAPD), a simple, trai…

Cited by 0 Source ↗Scholar ↗

2026

Rainbow Padding: Mitigating Early Termination in Instruction-Tuned Diffusion LLMs

BumJun Kim, Dongjae Jeon, Dueun Kim, Wonje Jeung, Albert No

ICLR 2026poster

Diffusion large language models (dLLMs) have emerged as a promising alternative to autoregressive models, offering flexible generation orders and strong performance on complex reasoning tasks. However, instruction-tuned dLLMs exhibit a critical vulnerability we term \<eos\> overflow: as allocated s…

Cited by 0 Source ↗Scholar ↗