Register Tokens for Bounded-State Reasoning in Diffusion Language Models — mentioned by E-Reverance

Thu, 17 Sep 2026 01:00:20 GMT

In my opinion its a great idea because you get reuse hidden states which allows you to do something akin to thinking/[online learning]. If you only evolve use input space outputs you're dealing with more decoding pressure. Also related to what I was saying see these: https://arxiv.org/pdf/2609.16372v1 https://arxiv.org/pdf/2609.01449v1 (this one is quite mindblowing cause they noise they the actual input space every time but it still works) https://arxiv.org/pdf/2609.11801

Paper and discussion history. A mention is not an endorsement.

Register Tokens for Bounded-State Reasoning in Diffusion Language Models · HackerPapers discussions

Register Tokens for Bounded-State Reasoning in Diffusion Language Models — mentioned by E-Reverance