Author

Mohammad Atif Quamar

h-index 1 View on Semantic Scholar

We have 2 of 7 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Aug 2026

Prefix-Denoising Consistency: Test-Time Verification for Diffusion Language Models

PDC consistently improves upon the initial sample, outperforms independent generations under a computational constrained comparison, and is robust to different unmasking strategies and parameter settings, highlighting prefix-conditioned regeneration as an effective DLM-specific primitive for test-time verification.

Yuki Ichihara, Naoto Iwase, Mohammad Atif Quamar et al. · 0 citations

Preprint Aug 2026

Privileged Solutions or Context-Induced Teacher Behavior? Dissecting On-Policy Self-Distillation

The role of target-specific privilege is investigated with On-Policy Self-Distillation from Other Problems (On-Policy Self-Distillation from Other Problems), which replaces the paired reference with a problem and solution from a different example, while preserving the student rollout, teacher, and distillation objective.

Yuki Ichihara, Naoto Iwase, Mohammad Atif Quamar et al. · 2 citations