Add EvoTokenDLM and DiSE (ACL 2026)

[celve]

Adds two ACL 2026 main-conference papers on diffusion language models:

• Beyond Hard Masks: Progressive Token Evolution for Diffusion Language Models (arXiv:2601.07351) — replaces hard binary masks with evolving soft-token distributions; introduces continuous trajectory supervision training. Code: https://github.com/aim-uofa/EvoTokenDLM → Foundation Model
• Efficient Self-Evaluation for Diffusion Language Models via Sequence Regeneration (arXiv:2603.02760) — training-free self-evaluation via token-regeneration probability; enables flexible-length generation. → Fast Sampling > Advanced Sampling Method

Entries follow the existing bold-title + single-line-date format. Chronological ordering preserved (appended to end of each section).

Disclosure: I am a co-author on both papers.

[Copilot]

Pull request overview

This PR updates the project’s README paper list to include two new ACL 2026 diffusion language model papers in the appropriate topical sections, maintaining the existing formatting and chronological ordering.

Changes:

• Add “Beyond Hard Masks: Progressive Token Evolution for Diffusion Language Models” under Foundation Model (with paper + code links).
• Add “Efficient Self-Evaluation for Diffusion Language Models via Sequence Regeneration” under Fast Sampling → Advanced Sampling Method (with paper link).

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.