paper-with-me

홈 › Papers

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

2026-07-03 · Chengcheng Wang, Tingzhang Luo, Wenhao Li, Jianyuan Guo, Chang Xu 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.

📄 PDF Abstract BibTeX arXiv:2607.03236

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

SLIM-RL: Risk-Budgeted Random-Masking RL for Diffusion LLMs Without Trajectory Slicing

2026-06-30 · Ruikang Zhao, Zhenting Wang, Han Gao, Ligong Han arxiv

Reinforcement learning for diffusion large language models (dLLMs) has largely moved to trajectory-aware methods. The current state of the art, TraceRL, holds that random masking is mismatched with the model's inference …

Reinforcement Learning

DiG-Plan: Mitigating Early Commitment for Tool-Graph Planning via Diffusion Guidance

2026-06-04 · Yansi Li, Zhuosheng Zhang arxiv

Generating executable tool plans requires selecting appropriate subsets from tool libraries, a combinatorial search problem with an exponentially large solution space. However, we identify a critical misalignment in pred…

Context-Aware Initialization for Reducing Generative Path Length in Diffusion Language Models

2025-12-22 · Tongyuan Miao, Gary Huang, Kai Jun Han, Annie Jiang arxiv

Diffusion Large Language Models (DLLMs) enable fully parallel token decoding but often remain impractical at inference time due to the many denoising iterations required to refine an information-free, fully masked initia…

Read the Trace, Steer the Path: Trajectory-Aware Reinforcement Learning for Diffusion Language Models

2026-06-03 · Anant Khandelwal, Manish Gupta arxiv

Diffusion large language models (dLLMs) generate responses by iteratively unmasking and revising many positions in parallel. This process leaves a rich denoising trace depicting which tokens become confident, which remai…

Reinforcement Learning

BAG: Budget-Aware Gating for Diffusion Caching

2026-08-10 · Tong Zhao, Mingkun Lei, Yucheng Han, Chi Zhang arxiv

Diffusion caching is a lightweight strategy that accelerates Diffusion Transformers (DiTs) by reusing intermediate features across denoising steps, but existing paradigms face a fundamental trade-off: online heuristics l…