- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper allocates inference compute during diffusion decoding through repeated trajectory expansion, look-ahead verification, and SSP resampling. Scheduled child–parent potentials telescope to a terminal reward tilt, and majority voting selects the final answer. Experiments cover six benchmarks with LLaDA and additional evaluations with Dream, showing gains over single-trajectory decoding and task-dependent advantages over output-level sampling.
Critical strengths
-
The method combines look-ahead scoring, scheduled weighting, and low-variance resampling into a practical inference procedure requiring no model training or external reward model by default. The fixed-schedule telescoping construction provides a coherent account of its intended target.
-
The empirical gains are useful, particularly on ARC-Challenge and TruthfulQA: at the default budget, S³ exceeds BoK by 3.07 and 4.68 percentage points, respectively. Code-generation and Dream results broaden the evidence. The analysis also reports limitations, including the small advantage on GSM8K and cases where PRM cost favors terminal selection.
Critical weaknesses
-
The central diversity diagnostics contradict their interpretation. Appendix E.2 states that SSP retains all eight founder lineages throughout 64 steps, but Figure 10 shows the SSP curves declining to approximately two and three surviving lineages. The numerical descriptions of Figure 9 also disagree with its bars. These discrepancies weaken the evidence supporting diversity preservation as a mechanism behind the gains. Moreover, the fixed-frontier resampling comparison does not establish its contribution to final accuracy, especially when systematic and stratified resampling also match SSP's reported offspring-count variance.
-
I could not reconstruct the evaluated procedure consistently from the stated configurations. Section 3.1 defines , whereas Table 11 fixes with default . Algorithm 1 votes over survivors, but Table 19 describes surviving particles despite the default . These may be notation or reporting errors, but they concern the decoding schedule and final candidate population and need resolution before the matched-compute comparisons can be assessed confidently.
Questions for authors
-
Please reconcile Figures 9–10 with Appendix E.2 and specify the logging, aggregation, and genealogy-tracking procedure. Were alternative resamplers evaluated by replaying fixed frontiers or within complete searches? How does SSP compare with systematic or stratified resampling in final accuracy under otherwise matched settings?
-
What exact decoding schedule and final voting population produced the reported results? Please resolve the configuration discrepancies above and explain the forward-pass counts and reuse of look-ahead logits.
-
How does Lemma 1 establish decreasing ESS with nonuniform parent offsets? For , child scores , and corresponding parent scores , Eq. (9) gives ESS increasing from approximately at to at . This appears to invalidate the stated monotonicity argument for Algorithm 2, although it does not affect the default fixed schedule.
2 comments
Thank you for the clarifications. They resolve my questions about the schedule and ESS. However, the submitted experiments used a different weighting rule from Eq. (6), and the reported advantage over BoK does not hold under matched selection. This weakens the empirical support for the paper, so I will maintain score. I have no further questions.
Response to Reviewer EYwn
Thank you for the close reading of the appendices. The new runs and the accuracy results are in the response to Reviewer dCQ4 (W2). Here, S3 means S3 (505 NFE), unless we state another NFE.
W1 and Q1. Figures 9–10 and Appendix E.2. Figure 10 shows the behavior of SSP: the mean number of lineages falls from 8.0 at step 0 to 4.41 at step 32 and 2.14 at step 56. Appendix E.2 does not agree with this, and we will correct it with Figure 10 as the reference.
Table R5. Replay on identical logged weights (MATH-500, N=8, b=2).
We apply each resampler to the same logged weights and track the founder of each particle. As you noted, this replay does not measure accuracy. In complete searches (Table R4, response to Reviewer AhDy), SSP does not differ significantly from systematic or stratified resampling [1]. Multinomial resampling is worse on ARC-C and TruthfulQA under majority vote. Thus, the evidence supports low-variance resampling as a group, but not SSP in particular.
W2 and Q2. Schedule and voting population. The schedule follows Algorithm 1 and Table 11: L=128, K=64, T=64, so 2 blocks of 32 steps, with 2 positions for each step. We will use this convention in Section 3.1.
The submitted tables mix votes over the N survivors and over the Nb frontier. The new results vote over the Nb frontier for all methods. A vote over the survivors changes the majority accuracy of S3 by at most 0.34 pp.
NFE and reuse of logits. S3 (505 NFE) scores each child: 1+63×8=505 passes, 1 for the root and 63×8 for the children. Survivors reuse cached logits at the next step.
The submitted S3 runs used the stand-alone weight of Remark 2, not Eq. (6) of Algorithm 1. Two children of one parent received the same look-ahead score, so these runs used 256 distinct NFE, not 512. We call this setting S3 (256 NFE), and we will report both counts. The telescoping argument (Proposition 2) does not apply to these runs. It applies only to S3 (505 NFE).
Q3. Lemma 1. Your example is correct: in general, ESS is not monotone in Δ. We will replace Lemma 1 with this precise statement. Let ℓk(Δ)=ok+Δsk with offset ok=τt(sk−skpar), weights wk∝eℓk, and w~k∝wk2. Let sˉ=∑kwksk and s~=∑kw~ksk.
Algorithm 2 needs a feasible initial point, ESS(0)≥ηNb, and then keeps ESS(lo)≥ηNb. Thus, it returns a feasible increment. It returns the largest feasible increment only if ESS is monotone over the full interval [0,τ−τt]. Item 3 at every Δ in this interval is one sufficient case. If ESS is not monotone, it can return a smaller feasible increment. The fixed schedule in our experiments does not use Algorithm 2 or this lemma.
Telescoping and performance. Proposition 2 is an algebraic identity for the target of Eq. (6). It does not cover the submitted S3 (256 NFE) runs. It is not an accuracy guarantee for S3 (see Reviewer AhDy, Minor 2).
BoK margins in your summary. These margins compare BoK under arg-max with S3 under majority vote (see Reviewer dCQ4, W2). With the same rule, S3 does not beat BoK.
[1] Srinivasan. FOCS 2001. https://doi.org/10.1109/SFCS.2001.959935