FoCore uses self-contrast on early-converging high-density tokens to boost diffusion LLM quality on reasoning benchmarks while cutting decoding steps by over 2x.
Adaptive classifier-free guidance via dynamic low-confidence masking.arXiv preprint arXiv:2505.20199, 2025a
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Focus on the Core: Empowering Diffusion Large Language Models by Self-Contrast
FoCore uses self-contrast on early-converging high-density tokens to boost diffusion LLM quality on reasoning benchmarks while cutting decoding steps by over 2x.