{"ID":5935774,"CreatedAt":"2026-07-07T01:22:02.77346169Z","UpdatedAt":"2026-07-07T02:10:06.972658124Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.03236","arxiv_id":"2607.03236","title":"TACG: Trajectory-Aware Commit Gating for Diffusion Language Model Decoding","abstract":"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.","short_abstract":"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 r...","url_abs":"https://arxiv.org/abs/2607.03236","url_pdf":"https://arxiv.org/pdf/2607.03236v1","authors":"[\"Chengcheng Wang\",\"Tingzhang Luo\",\"Wenhao Li\",\"Jianyuan Guo\",\"Chang Xu\"]","published":"2026-07-03T11:45:21Z","proceeding":"cs.CL","tasks":"[\"cs.CL\"]","methods":"[\"Diffusion Model\",\"Large Language Model\",\"Language Model\"]","has_code":false,"code_links":[{"ID":613932,"CreatedAt":"2026-07-07T01:22:02.77346169Z","UpdatedAt":"2026-07-07T01:22:02.77346169Z","DeletedAt":null,"paper_id":5935774,"paper_url":"https://arxiv.org/abs/2607.03236","paper_title":"TACG: Trajectory-Aware Commit Gating for Diffusion Language Model Decoding","repo_url":"https://github.com/Clarence-CV/TACG-DLLM","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}