mlsanityData to 14 Oct 2026

TACG: Trajectory-Aware Commit Gating for Diffusion Language Model Decoding

arXiv 2607.03236 ↗First posted 3 Jul 2026cs.CLcited by 3

On the map: Diffusion models → Discrete and text diffusion

Abstract (arXiv)

Diffusion language models (DLLMs) generate text by iteratively denoising masked positions, exposing a trajectory of predictive distributions rather than a single instantaneous belief. Most existing decoders ignore this trajectory and commit tokens from the current snapshot alone, conflating confidence with commitment readiness: a transient top-1 peak under incomplete context can be locked in, while candidates with consistent cross-step support are delayed. We propose Trajectory-Aware Commit Gating (TACG), a training-free gate-level decoder that anchors token identities to the base posterior and uses trajectory-aware signals only to decide whether the current proposal is ready to commit. TACG combines Temporal Implicit Logits Guidance (TILG), which keeps an exponential moving average of past logits as a self-reference and contrasts the current logits against this reference in natural-parameter space, with a History Gate (HG) that enforces short-term proposal persistence before commitment. Together with a capped extra-promotion budget, these components yield a stability-constrained commit rule without auxiliary networks or extra forward passes. We evaluate TACG on LLaDA, Dream, and LLaDA2-Mini across code (HumanEval, MBPP) and math (GSM8K, MATH500) benchmarks; it typically improves or preserves accuracy while reducing denoising steps and increasing tokens per forward (TPF). The code is publicly available at https://github.com/Clarence-CV/TACG-DLLM.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Placed in Discrete and text diffusion, one of 14 directions in Diffusion models. Builds on 46 of its 50 references that are in the map. First posted on arXiv 3 Jul 2026.

Builds on

46 of 50 references.

Closest ancestors · 19

Same direction, newest first

  1. 2026Attention-Based Sampler for Diffusion Language Models18 Mar
  2. 2026The Flexibility Trap: Rethinking the Value of Arbitrary Order in Diffusion Language Models21 Jan
  3. 2025d-TreeRPO: Towards More Reliable Policy Optimization for Diffusion Language Models10 Dec
  4. 2025KLASS: KL-Guided Fast Inference in Masked Diffusion Models7 Nov
  5. 2025Lookahead Unmasking Elicits Accurate Decoding in Diffusion Language Models4 Nov
  6. 2025Improving Reasoning for Diffusion Language Models via Group Diffusion Policy Optimization9 Oct
  7. 2025Advancing Reasoning in Diffusion Language Models with Denoising Process Rewards2 Oct
  8. 2025Dream-Coder 7B: An Open Diffusion Language Model for Code1 Sep
  9. 2025Diffusion Language Models Know the Answer Before Decoding27 Aug
  10. 2025Dream 7B: Diffusion Large Language Models21 Aug
  11. 2025wd1: Weighted Policy Optimization for Reasoning in Diffusion Language Models7 Jul
  12. 2025DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation25 Jun
7 more
  1. 2025LongLLaDA: Unlocking Long Context Capabilities in Diffusion LLMs17 Jun
  2. 2025Dimple: Discrete Diffusion Multimodal Large Language Model with Parallel Decoding22 May
  3. 2025Large Language Diffusion Models14 Feb
  4. 2024Masked Diffusion Models are Secretly Time-Agnostic Masked Models and Exploit Inaccurate Categorical Sampling4 Sep
  5. 2024Simple and Effective Masked Diffusion Language Models11 Jun
  6. 2024Simplified and Generalized Masked Diffusion for Discrete Data6 Jun
  7. 2023Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution25 Oct

Other references · 27

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  4. 2020Denoising Diffusion Implicit Modelscited by 9,781
  5. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  6. 2022Classifier-Free Diffusion Guidancecited by 5,847
  7. 2023Adding Conditional Control to Text-to-Image Diffusion Modelscited by 5,503
  8. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  9. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,356
  10. 2021Program Synthesis with Large Language Modelscited by 3,637
  11. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,300
  12. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,719
15 more
  1. 2025Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decodingcited by 420
  2. 2025LLaDA 1.5: Variance-Reduced Preference Optimization for Large Language Diffusion Modelscited by 285
  3. 2025LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuningcited by 148
  4. 2025Accelerating Diffusion LLMs via Adaptive Parallel Decodingcited by 112
  5. 2026d3LLM: Ultra-Fast Diffusion LLM using Pseudo-Trajectory Distillationcited by 37
  6. 2025Accelerating Diffusion LLM Inference via Local Determinism Propagationcited by 21
  7. 2025LoPA: Scaling dLLM Inference via Lookahead Parallel Decodingcited by 17
  8. 2025UltraLLaDA: Scaling the Context Length to 128K for Diffusion Large Language Modelscited by 13
  9. 2025SDAR-VL: Stable and Efficient Block-wise Diffusion for Vision-Language Understandingcited by 13
  10. 2026Stable-DiffCoder: Pushing the Frontier of Code Diffusion Large Language Modelcited by 13
  11. 2026MMaDA-VLA: Large Diffusion Vision-Language-Action Model with Unified Multi-Modal Instruction and Generationcited by 10
  12. 2026DLLM-Searcher: Adapting Diffusion Large Language Model for Search Agentscited by 8
  13. 2026Streaming-dLLM: Accelerating Diffusion LLMs via Suffix Pruning and Dynamic Decodingcited by 7
  14. 2026DLLM Agent: See Farther, Run Fastercited by 6
  15. 2026Mosaic: Unlocking Long-Context Inference for Diffusion LLMs via Global Memory Planning and Dynamic Peak Tamingcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026From Truncation to Commitment: Persistent Context in Uniform Discrete Diffusioncited by 1
  2. 2026Commit Locally, Exit Globally: Coordinating Adaptive Sampling and Early Exit in Diffusion Language Modelscited by 0
  3. 2026TRACESTEER: Trace Driven Activation Steering for Hallucination Mitigation in Diffusion Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Don't Commit Alone: Joint Token Commitment in Diffusion Language Models
  2. 2026Efficient Diffusion LLMs via Temporal-Spatial Parallel Decoding and Confidence Extrapolation
  3. 2026Self-Distilled Trajectory-Aware Boltzmann Modeling: Bridging the Training-Inference Discrepancy in Diffusion Language Models
  4. 2026On Trajectory-Aware Training for Masked Diffusion Language Models
  5. 2026CORA-Diff: Confidence-Oriented Residual Acceptance for Efficient Diffusion Language Model Inference
  6. 2026Who Should Lead Decoding Now? Tracking Reliable Trajectories for Ensembling Masked Diffusion Language Models
  7. 2026Deferred Commitment Decoding for Diffusion Language Models
  8. 2026Polestar: Drift-Aware Cache Calibration and Token Commitment for Efficient Inference of Diffusion LLMs