AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Residual Context Diffusion Language Models

Yuezhou Hu, Harman Singh, Monishwaran Maheswaran, Haocheng Xi, Coleman Richard Charles Hooper, Jintao Zhang, Aditya Tomar, Michael W. Mahoney, Sewon Min, Mehrdad Farajtabar, Kurt Keutzer, Amir Gholami and 1 more

ICML 2026 · Large language models › Language, safety and interpretability · approach: Parallel and early-exit decoding · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add residual context diffusionSDAR-8B-b64 SeqD · AIME24accuracy (%): 7.08 -> 18.75Improved
add residual context diffusionSDAR-8B-b64 SeqD · AIME25accuracy (%): 9.79 -> 16.04Improved
add residual context diffusionSDAR-8B-b32 SeqD · MATH500accuracy (%): 65.80 -> 76.20Improved
add residual context diffusionLLaDA SeqD · MinervaMathaccuracy (%): 31.10 -> 37.00Improved
add residual context diffusionLLaDA SeqD · GSM8Kaccuracy (%): 75.74 -> 78.09Improved
add residual context diffusionSDAR-4B-b32 SeqD · GSM8Kaccuracy (%): 81.73 -> 84.91Improved
train baseline longer (8 epochs) instead of adding RCD (ablation)SDAR-8B-b64 SeqD · MATH500accuracy (%): 68.00 -> 74.40Mixed
Dirichlet init instead of 1.7B reference-model init (ablation)RCD SDAR-8B-b64 · MATH500accuracy (%): 74.2 -> 77.4Improved
zero init instead of 4B reference-model init (ablation)RCD SDAR-8B-b64 · MATH500accuracy (%): 71.8 -> 75.6Improved
add RCD residual vs Loopholing hidden-state injection (ablation)SDAR-4B-b64 · GSM8Kaccuracy (%): NA -> 86.35Improved
add normalized-entropy weighting of residual (ablation)RCD interpolation · SDAR-8B-b64 untrained draftPareto frontier accuracy vs tokens per stepImproved
add RCD under matched throughput (Fastdllm)LLaDA SeqD · GSM8Kaccuracy (%): 76.12 -> 78.70Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy