- Rating
- 3(+1)
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
Tradition CFG applies same guidance scale to every token at every timestep, though different tokens' velocity predictions evolve differently along the sampling trajectory. TrajCFG is a training-free method, which measures the same token's angle of velocity between adjacent timesteps, and maps that angle to a bounded per-token guidance scale, giving weaker guidance where the predicted direction has been stable and stronger guidance where it is still being revised. Main experiments were conducted on Wan series models, and suprpassed most of compared guidance methods on Vbench and Dynamic-Bench with almost no additional computational cost. The authors also conducted experiments on image-to-video, text-to-audio-v text-to-image and image-to-3D, mostly against the base sampler only. Ablations show token-wise allocation beats time-only and sample-wise variants, and that shuffling or reversingthe per-token assignment degrades results, indicating the angle signal carries genuine spatial information rather than merely raising the average scale.
Critical strengths
- Coverage of models, domains. Same mechanism available for many Wan series models, Cosmos-Predict2.5-2B, LTX-2.3, Ovi, Qwen-Image, SD3 Medium and Hunyuan3D 2.0. Including T2V, I2V, T2AV, and Image-to-3D.
- Almost no additional computation cost because CFG already need to compute the velocity.
- Ablation studies shows the angle map carries spatial information rather than acting as a global rescale.
Critical weaknesses
W1. The reported gains are not shown to come from token-wise allocation. Two other explanations remain open: TrajCFG simply runs at a stronger average guidance scale, and the baselines it beats do not adapt the scale per location.
- Higher guidance scale. A token's scale can reach s_max=10.0 against s_base=6.0 (Tab. 7), and gains concentrate on VBench Semantic (72.77 to 75.37, Wan2.1-1.3B) over Quality (83.03 to 84.06, Tab. 1). "Guidance Matters" (arXiv 2602.22570) finds that raising the CFG scale alone competes with most recent guidance methods, since alignment metrics reward large scales. No constant-scale control reported, at the realized mean or at s_max.
- Wrong opponents. The four baselines (CFG-Zero*, CFG-Ctrl, PAG, SRVS; p6) do not allocate per-location scales. SAMG (point-wise) and VAGS (cosine-driven) are described in the paper's own related work (p2), HiGS (arXiv 2509.22300), APG and interval guidance are cited, and none is run. FBG (arXiv 2506.06085) is not cited. Baseline tuning is undocumented while TrajCFG tunes four hyperparameters per backbone.
W2. The rule that sets each token's guidance was never checked on the states the sampler actually visits.
- Fig. 3 needs intermediate states at step k and the two before it. Instead of running the sampler, the authors noise real video latents, "without sampler rollouts" (p4), then sweep CFG on those hand-built states. That one measurement is the whole basis for "small angle, lower guidance; large angle, higher guidance".
- Real sampling compounds its own errors from step to step and can end up far from any real video; noised real video stays near a correct answer throughout. So the experiment tests a regime the method never meets. The authors call the transfer a "hypothesis" (p5) and never check it.
Questions for authors
- What is the guidance scale in practice, averaged over steps and tokens? And if the authors run the CFG at guidance scale of s_max, how much of the gap to TrajCFG is left? Since simply larger guidance scale of CFG might improve these metrics. 2. Were the baselines tuned per backbone the way TrajCFG was, or run at their published defaults?
- The authors may add SAMG / VAGS / HiGS / APG if applicable in T2V / T2I for comparison. Since they are more similar to TrajCFG.
- Does the Fig. 3 effect survive if the states come from actual sampler rollouts instead of noised ground truth? A wider CFG sweep would also help, since some groups look like they bottom out at the edge of the current range.
4 comments
Response to Reviewer xTwH
W1 / Q3. Comparison with SAMG, VAGS, HiGS, and APG.
| T2V, VBench | 1.3B Quality | 1.3B Semantic | 1.3B Total | 5B Quality | 5B Semantic | 5B Total |
|---|---|---|---|---|---|---|
| Base CFG | 83.03 | 72.77 | 80.98 | 82.88 | 74.52 | 81.21 |
| SAMG | 83.40 | 72.71 | 81.26 | 84.11 | 74.15 | 82.11 |
| VAGS-Gen | 83.01 | 73.98 | 81.21 | 82.99 | 75.43 | 81.48 |
| HiGS | 83.68 | 72.68 | 81.48 | 84.01 | 75.23 | 82.25 |
| APG | 82.97 | 73.03 | 80.98 | 83.44 | 75.46 | 81.84 |
| TrajCFG | 84.06 | 75.37 | 82.32 | 84.82 | 77.45 | 83.35 |
| T2I, SD3-Medium, GenEval | Single | Two | Count. | Colors | Pos. | Attr. | Overall |
|---|---|---|---|---|---|---|---|
| Base CFG | 0.9969 | 0.8611 | 0.6125 | 0.8351 | 0.3125 | 0.5650 | 0.7021 |
| SAMG | 0.9969 | 0.8687 | 0.6500 | 0.8457 | 0.3200 | 0.5925 | 0.7123 |
| VAGS-Gen | 0.9969 | 0.8611 | 0.6406 | 0.8511 | 0.3100 | 0.5825 | 0.7070 |
| HiGS | 1.0000 | 0.8611 | 0.6500 | 0.8351 | 0.3225 | 0.5850 | 0.7090 |
| APG | 1.0000 | 0.8712 | 0.6188 | 0.8324 | 0.3225 | 0.5750 | 0.7033 |
| TrajCFG | 1.0000 | 0.8813 | 0.6344 | 0.8484 | 0.3225 | 0.5975 | 0.7140 |
On T2V, TrajCFG leads the strongest added baseline, HiGS, by 0.84 and 1.10 VBench Total. On T2I, it has the best GenEval Overall (+0.0017 over SAMG), although SAMG and HiGS are better on Counting and VAGS-Gen is better on Colors.
Limited-interval guidance is a time-only schedule. We did not run it, but Time-only in Table 3 and the step-wise schedules above are time-only controls derived from TrajCFG's own profile, and token-wise allocation improves on each of them by at least 0.27 Total. The two could be combined, with an interval schedule setting the time profile and TrajCFG allocating within it, but we have not tested this.
FBG sets a state-dependent guidance coefficient during sampling, whereas TrajCFG allocates scales per token from cross-step feedback. We thank the reviewer for the pointer and will cite and discuss FBG in the related work.
W2 / Q4. Fig. 3 on actual sampler states.
We repeated the Fig. 3 protocol on states visited by the sampler (Figure 11). The setup is as follows:
- States. Wan2.2-5B starts from random noise with DAVIS text prompts and runs 10 steps with base CFG.
- Groups. At step 10, tokens are grouped by the angle between their guided velocities at the two preceding steps, using the Fig. 3 bins.
- Intervention. From the same latent, we change the scale of one group while the other groups stay at the default, then finish sampling.
- Metric. These states have no ground truth, so we score the final videos by frame-level CLIP text alignment, averaged over prompts.
Figure 11 shows the same qualitative pattern as Fig. 3. Small-angle tokens prefer guidance below the default. Large-angle tokens benefit from guidance above the default and lose alignment when it is lowered. The transition between these two regimes falls in the 3–6° group, which is largely insensitive in Figure 11. Because only one group is changed at a time, the effects are small. We therefore read both figures qualitatively, as showing that the preferred scale increases with the angle, and not as a calibrated mapping from angle to scale.
The closest test on TrajCFG's own closed-loop trajectories is the Shuffled and Reversed ablations in Table 3: replaying its scale maps with shuffled or reversed token assignments, from the same initial noise, lowers Total by 0.46–0.77.
Thank you for the detailed response. My concerns have been well addressed, and I will raise my score.
Dear Reviewer xTwH,
We are delighted that our responses have addressed your concerns and that you now support the acceptance of our paper.
Thank you for your thoughtful and constructive feedback, which has helped us clarify the paper’s contributions and strengthen its presentation.
Best regards, The Authors
Response to Reviewer xTwH
We thank the reviewer for the specific controls and comparisons requested. New figures are on the last page (p. 24) of the revised appendix.
W1 / Q1. Realized guidance scale and constant-CFG controls.
Averaged over samples, tokens, and all 40 steps on VBench, TrajCFG's applied scale is 5.65 on Wan2.1-1.3B and 3.92 on Wan2.2-5B. Both are below the default scales of 6 and 5.
Figure 12 shows the per-step mean. It is close to 10 when adaptation starts, falls below the default by about step 10, and stays near smin for most of the remaining steps. Only a few early steps approach smax, so TrajCFG mostly lowers guidance rather than raising it. Within each step, sites with large historical angles still receive more guidance than settled sites. The allocation changes the ranking across tokens, not the mean level.
The step-wise rows apply, at each step, a single scale to all tokens: the token maximum or the token mean of TrajCFG's map at that step.
We attribute this remaining gap to token-wise allocation, in agreement with the Time-only and Sample-wise ablations in Table 3. TrajCFG's improvement thus has two parts: the time profile that its feedback produces, and the allocation across tokens. We will report both explicitly.
W1 / Q2. Baseline tuning.
All baselines, including those added below, used the method-specific defaults from their papers at the backbone's default CFG. We did not run a per-backbone search for them.
For TrajCFG, Table 7 lists the per-backbone settings. The history spans and fusion weights (0.8/0.2) are fixed across all backbones, and ρlo takes only two values (0.90 or 0.95), so the main per-backbone choice is [smin,smax]. We acknowledge this asymmetry. The controls above show that changing vanilla CFG's scale, or its per-step schedule, does not reproduce the gain.