Entity · paper

Follow the Latent Roadmap: Navigating Revocable Decoding for Diffusion LLMs with Anchor Tokens

paperactivefollow-the-latent-roadmap-navigating-revocable-decoding-for-diffusion-llms-with-anchor-tokens-5a71a203·1 events·first seen Jun 16, 2026

Aliases: Follow the Latent Roadmap: Navigating Revocable Decoding for Diffusion LLMs with Anchor Tokens

Co-occurring entities

ASRD

More like this (12)

Adaptive Multi-Step Lookahead Decoding for Diffusion Language Models On-Policy Distillation for LLM Safety: A Routing Approach to Template-Robust Realignment Knowledge Knows, Verbalization Tells: Disentangling Latent Directions for Mathematical Solvability in LLMs Agentic Chain-of-Thought Steering for Efficient and Controllable LLM Reasoning Attractor States Emerge in Multi-Turn LLM Conversations Extending LLM Context via Associative Recurrent Memory Self-Augmenting Retrieval for Diffusion Language Models From Tokens to States: LLMs as a Special Case of World Models and the Continuous Path Beyond Which Models Are Our Models Built On? Auditing Invisible Dependencies in Modern LLMs Continual LLM Upcycling: A Predictor-Gated Bank-Wise Sparsity Training Recipe for Dense-to-Sparse LLMs Can We Break LLMs Out of Self-Loops? Fine-Grained Reasoning Control with Activation Steering Pass the Baton: Trajectory-Relayed On-Policy Distillation

Recent events (1)

5arXiv · cs.CL·Jun 16, 2026·source ↗

ASRD: Training-free anchor-guided revocable decoding for diffusion LLMs improves accuracy and throughput

A new arXiv preprint introduces ASRD (Anchor Supervised Revocable Decoding), a training-free framework for improving decoding quality in diffusion large language models. The method addresses error propagation and local error reinforcement in revocable decoding by separating trusted 'anchor tokens' (identified via temporal consistency) from uncertain candidates, then applying anchor-guided generation and anchor-perturbed verification. Experiments on math and coding benchmarks show up to 6.4% accuracy improvement and 7.2× inference throughput gains over remasking baselines.

Inference Economics ASRD Follow the Latent Roadmap: Navigating Revocable Decoding for Diffusion LLMs with Anchor Tokens