- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper decomposes few-step dLLM decoding loss into marginal drift (per-position posterior gap to a fine-step teacher) and dependency error (incompatibility among tokens committed in the same step under conditional independence). PC-LoRA freezes the backbone and aligns marginals with mixed-budget on-policy teacher KL. Counterfactual parent reveals then supply pairwise dependency targets; a lightweight head predicts them from one student forward pass, the graph is compressed into a maximum spanning tree rooted at the highest-confidence candidate, and a correction head applies gated parent-conditioned logit residuals, with exactly one backbone forward per step. The appendix provides an exact KL decomposition and the Chow-Liu projection. At 64 NFE, TreePC beats vanilla, Fast-dLLM and CD4LM on GSM8K, MATH-500, HumanEval and MBPP.
Critical strengths
The decomposition maps one-to-one onto the method: splits marginal calibration from residual dependence, and the Chow-Liu connection justifies a mutual-information-weighted MST rather than an ad hoc graph. Counterfactual parent reveals are a computable supervision signal. Mechanism-level metrics are thorough (marginal KL, top-1 agreement, Spearman, edge overlap, conditional KL reduction) with local-tree and random-tree controls.
Critical weaknesses
- Every efficiency number is an estimate, and latency is never compared against competing methods. All three RQ4 subfigures are labeled "Estimated"; the text states that per-step cost was measured only on GSM8K and HumanEval and multiplied by NFE, with task-family proxies for MATH-500 and MBPP, so four benchmarks of bars rest on two measurements. The extrapolation ignores EOS early stopping, the dependence of auxiliary cost on , and whether the vanilla side uses caching. The per-step latency, throughput and peak memory promised in Section 4.2 appear nowhere. The latency comparison is against vanilla-256/512 only; Fast-dLLM and CD4LM never enter a latency figure, although the former carries a block KV cache so equal NFE does not mean equal wall-clock, and the dLLM-Cache line appears only in related work.
- The end-to-end evidence is a single operating point, with no ablation. Of nine nominal budgets only has a numeric table; , 8 and 16 appear only as axis ticks, although the conclusion itself notes that tree construction and correction add measurable latency at very small budgets. The word "ablation" never appears: the tree against a full graph, the root rule, the gate and global scale, the ranking loss, counterfactual samples and top- are none of them isolated, even though the paper calls the ranking term especially important. The generation length is never stated, so NFE cannot be converted into the per-step commit count , on which both tree coverage and the auxiliary cost depend. There are also no seeds or confidence intervals, while several Table 2 gaps are 1-3 points.
- Most of the gain comes from standard distillation. From RQ1 and Table 2, Dream GSM8K at 64 NFE gives vanilla 49.7, PC-only about 67.4, TreePC 72.5, so the titular tree correction adds about 5.1 points, and 1.7 on LLaDA. Neither abstract nor conclusion conveys this.
- Training availability is unmatched, its cost is unreported, and the closest competitors are absent. TreePC collects trajectories on each benchmark's training distribution and trains three parameter groups, while Fast-dLLM is training-free and it is unstated whether CD4LM was retrained under matched data and compute. Counterfactual labeling runs one teacher forward per sampled parent for every candidate pair, and the four-stage pipeline regenerates on-policy states, yet no figure is given for that total cost. CoDD and Adaptive Parallel Decoding, which the paper lists as the same family, are never compared, and two controls are missing: PC-LoRA plus independent decoding plus a confidence threshold, and a plain LoRA SFT at the same data budget.
- The fixed-budget setting is never justified. Per-position confidence is free in text decoding, so discarding adaptivity needs an argument, and two obvious middle grounds go undiscussed: a confidence threshold with a step cap, and attaching the dependency correction to an adaptive scheduler, which the paper's own description allows. This also affects how Table 2 reads: forced to 64 NFE, Fast-dLLM falls below plain vanilla few-step on LLaDA MATH-500 and HumanEval (21.7 vs 26.3, 28.2 vs 34.0), which indicates it is being run outside its design point; the table supports "a confidence threshold fails at 64 NFE" rather than "TreePC beats Fast-dLLM". Separately, a tree keeps of the pairs, half at but about 3% at , so coverage is lowest exactly where the budget is tightest, and Figure 3(b)'s 75-92% does not say at which it was measured.
Questions for authors
- Can you provide measured end-to-end wall-clock on all four benchmarks (with the actual step counts under EOS early stopping), broken down into backbone, Dependency Head, MST and Correction Head, and place Fast-dLLM and CD4LM in the same latency figure?
- What is the generation length ? Please report quality, realized NFE, latency and the auxiliary share at , 8 and 16.
- Please add component ablations (tree versus full graph, root rule, gate and global scale, ranking loss, , top-), and place vanilla, PC-only and full TreePC in one table so the net contribution of the tree correction is directly readable.
- What are Fast-dLLM's full quality-versus-realized-NFE and wall-clock curves at its own thresholds (0.9, 0.8, 0.7)? What does a fixed budget buy over a threshold with a step cap, and what happens if the dependency correction is attached to an adaptive scheduler instead?
- What is the total cost of counterfactual labeling and the four-stage pipeline, in teacher forward passes and GPU hours? With "PC-LoRA + independent decoding + confidence threshold" and a same-data LoRA SFT as controls, how much lead remains, and how does TreePC compare with CoDD and Adaptive Parallel Decoding?
- What are the distributions of and tree depth, and the value and trigger rate of the fallback threshold ? How does the fraction of pairs covered by the tree vary with , and at which was Figure 3(b) measured?