- Rating
- 3(+1)
- Confidence
- 1(−1)
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
RINO is a self-supervised jet-representation method that applies DINO/iBOT-style teacher-student training across views obtained by stopping exclusive kT clustering at different numbers of subjets. It pretrains on QCD jets from JETCLASS, finetunes on labeled JETNET top-versus-QCD data, and evaluates on JETCLASS top jets and other unseen signal topologies. The paper reports improved OOD top-tagging, label efficiency, cross-topology transfer, and smaller representation-space domain discrepancies than several controlled-backbone SSL baselines.
Critical strengths
- Using physically ordered kT coarse-graining as the view construction is intuitive.
- The paper includes useful controlled comparisons, like Tables 1 and 3, Section 4.5.
Critical weaknesses
- I noticed RINO’s teacher settings were selected using JETCLASS OOD top-tagging accuracy, the headline metric. Can the authors identify a separate tuning split or report the tuned result on an untouched test set.
- Evaluation events are disjoint from pretraining, but all SSL methods pretrain on unlabeled JETCLASS QCD. I would like to ask the authors to qualify the OOD claim as transfer with access to target-domain background data.
- Baselines inherit published pretraining settings while RINO receives target-based tuning; reported deviations cover fine-tuning seeds. It would be more appropriate to assess the comparison under comparable tuning opportunities and independent pretraining runs.
Questions for authors
- Exactly which JETCLASS events and partitions are used for QCD pretraining, hyperparameter selection, probes, and final evaluation; are any final-evaluation QCD events present during pretraining?
- Can all headline results be recomputed on an untouched target test set after teacher temperature and EMA momentum are selected without using JETCLASS target labels?
- Why is the protocol described as target-label-free OOD evaluation when Appendix A.17 explicitly uses target-domain top-tagging accuracy to select RINO hyperparameters?
- What happens when every baseline receives the same target-validation access and hyperparameter-search budget as RINO?
- How many independently pretrained backbones were used per method, and how much variance remains when SSL pretraining, not only downstream finetuning, is repeated?
- For the ablation controls the clustering objective and nominal view scales. Can matched-count or information-retention controls further distinguish energy-ordered physical structure from differences in what each clustering algorithm preserves?
5 comments
Reply to Reviewer 1hwG (Part 2 of 3)
3. Selection on the Headline Metric, and Independent Pretraining Runs
I noticed RINO's teacher settings were selected using JETCLASS OOD top-tagging accuracy, the headline metric. Can the authors identify a separate tuning split or report the tuned result on an untouched test set.
Can all headline results be recomputed on an untouched target test set after teacher temperature and EMA momentum are selected without using JETCLASS target labels?
Why is the protocol described as target-label-free OOD evaluation when Appendix A.17 explicitly uses target-domain top-tagging accuracy to select RINO hyperparameters?
How many independently pretrained backbones were used per method, and how much variance remains when SSL pretraining, not only downstream finetuning, is repeated?
We would like to demonstrate that the teacher-setting tuning is not necessary. RINO retrained with DINOv2's default teacher setting yields over three independent pretraining runs, against for the tuned setting. JetClass labels are therefore not needed to select the teacher settings, and the main results no longer use any teacher setting selected on JetClass.
Why we remove the selection instead of moving it. In the paper's protocol, JetClass is a proxy for real collider data (Sec. 4.1, A.13), so it should be treated as an unlabeled dataset, with its labels entering only for evaluation. First, a separate JetClass tuning split would come from the same distribution as the test set, so selecting on it would still rely on labels that do not exist in practice. Second, real data carry no labels on which one can select a set of hyperparameters in hindsight. Therefore, we do not select the teacher settings at all: the teacher temperature and the EMA momentum follow DINOv2's default teacher setting (EMA momentum on a cosine schedule from 0.994 to 1, with teacher temperature warmed up linearly from 0.04 to 0.07), fixed before any of these runs is evaluated, with everything else unchanged. No teacher-setting choice then uses the JetClass test split.
The target-label-free protocol. RINO's training is target-label-free: no JetClass label enters pretraining or finetuning. We agree that JetClass labels should not also have been used to choose the teacher settings (A.17); with DINOv2's default teacher setting they are not. The caption of the deployment-workflow figure (Fig. 7 in the submission, Fig. 8 in the revised PDF) will say that "target-label-free" refers to the training stages.
The headline metric, recomputed. Table 1 uses one pretrained backbone per SSL method; its finetuned-head covers only that backbone's 10 finetuning seeds (Sec. 4.1). The table below gives Table 1's MLP-head accuracy (finetuned on JetNet, evaluated on JetClass top vs. QCD) over repeated, independent pretraining runs for RINO and five baselines (at least three runs per method; at least five finetuning seeds per run). Each entry is the mean over pretraining runs. The first is the standard deviation between pretraining runs, and the second is the spread between finetuning seeds within a run (the standard deviation over each run's finetuning seeds, combined over the runs as a root mean square).
| Method | JetClass top vs. QCD accuracy ( pretraining finetuning) |
|---|---|
| JetCLR | |
| OmniJet- | |
| MPMv1 | |
| MPMv2 | |
| JetCLR-scale | |
| RINO (DINOv2's default teacher setting) | |
| RINO (tuned setting, A.17) |
Two observations support the statement above:
- The two teacher settings agree: their means differ by 0.3 percentage points, well within the run-to-run spread of either (1.5 to 2.1 percentage points).
- The printed 0.847 (Table 1) lies within one standard deviation of the tuned setting's multi-run mean.
With either teacher setting and the MLP head, RINO remains above every baseline's multi-run mean.
The revised PDF will update Tables 1 and 2 in full over independent pretraining runs.
Reply to Reviewer 1hwG (Part 3 of 3)
4. Equal Tuning Access
What happens when every baseline receives the same target-validation access and hyperparameter-search budget as RINO?
An equal-budget search for every baseline does not fit this discussion window, so we answer from two sides.
RINO without the selection. With DINOv2's default teacher setting, RINO's teacher settings involve no choice on JetClass, and RINO remains above every baseline's multi-run mean with the MLP head (3 above).
Baselines with the selection. The appendix already reports checks of this kind, which also serve as extra information for future work. The finetuning weight decay and dropout of the supervised model, MPMv1, MPMv2 and RINO are varied in A.7 (Table 10): Table 1 uses the shared default, and the "Best" column is selected on JetClass accuracy. In Table 10, RINO's worst setting (0.843) exceeds the best setting of the other three methods (MPMv2, 0.810). One choice does enter Table 1: OmniJet- is read out by mean pooling over all tokens, which outperforms the last-token alternative (A.17). This is in line with the default classification head of the OmniJet- code, which also pools over all tokens (by a sum). We also vary a main pretraining hyperparameter of OmniJet-: tokenizer codebooks of 256, 2,048 and 8,192 codes (8,192 is the codebook size of the original OmniJet- model) yield , and (one pretraining run each). These are inside the range of the individual runs with our 512-code tokenizer ( to ; mean , table in 3), so we see no dependence on the codebook size, and all stay below RINO's mean with either teacher setting.
5. Matched-Count and Information-Retention Controls
Can matched-count or information-retention controls further distinguish energy-ordered physical structure from differences in what each clustering algorithm preserves?
Information retention. We run an information-retention control study with the classifier of the new experiment in our reply to Reviewer ANRi (Section 1.2), with the same network and training settings: a small permutation-invariant network that sees one view alone, from each subjet's transverse momentum fraction and direction. For each clustering and each , it separates top from QCD jets (JetClass test jets with at least 17 constituents, 100k per class; test AUC over three seeds, standard deviation at most ). The clustering only builds the pretraining views, so the control is on those views.
| Subjets | 2 | 3 | 4 | 6 | 8 | 16 |
|---|---|---|---|---|---|---|
| 0.914 | 0.963 | 0.961 | 0.959 | 0.961 | 0.963 | |
| Cambridge/Aachen | 0.774 | 0.883 | 0.927 | 0.955 | 0.961 | 0.960 |
| anti- | 0.639 | 0.738 | 0.804 | 0.871 | 0.903 | 0.946 |
(All particles: 0.969.) The clusterings differ in what their coarse views preserve: at two to four subjets, views carry more of what separates top from QCD jets than Cambridge/Aachen or anti- views. Cambridge/Aachen views catch up from on, and anti- views only approach them at . Two subjets cannot resolve the three prongs of a top-quark decay, which likely explains why is lower even for . However, the algorithm still outperforms the other two at .
Matched count. The views are compared at the same throughout. In about 90% of jets with more than 16 constituents, all three algorithms keep every particle. We repeat the control on these jets alone:
| Subjets | 2 | 3 | 4 | 6 | 8 | 16 |
|---|---|---|---|---|---|---|
| 0.915 | 0.962 | 0.960 | 0.959 | 0.959 | 0.963 | |
| Cambridge/Aachen | 0.789 | 0.892 | 0.931 | 0.957 | 0.963 | 0.964 |
| anti- | 0.651 | 0.748 | 0.811 | 0.878 | 0.906 | 0.946 |
Every entry is within of the table above, so the differences are in how the particles are grouped, not in which are kept.
What this shows. The order at coarse is the order of the transfer accuracies in Sec. 4.5 (, , and for , Cambridge/Aachen, and anti- respectively; for , across multi-seed pretraining runs with DINOv2's default teacher setting, now in the revised PDF), and the experiment in our reply to Reviewer ANRi (Section 1.2) finds the same clustering to carry the least trace of hadronization at coarse . Therefore, the control does distinguish the energy-ordered clustering: its coarse views keep most of the top vs. QCD separation and carry the least hadronization and detector detail.
We would like to thank the reviewer again for the constructive questions, which help make the protocol and its limits explicit.
Reply to Reviewer 1hwG (Part 1 of 3)
We thank the reviewer for the close reading of our protocol and the constructive questions. We have run new experiments to address them, and respond below, grouped by topic. Notably, the teacher-setting selection on JetClass is not needed: RINO retrained with DINOv2's default teacher setting [1] achieves an accuracy of 0.843±0.021 over three independent pretraining runs, against 0.840±0.015 for the tuned setting. We share the key results first. The complete tables with the other entries will be updated in the revised PDF.
1. Partitions, and Overlap Between Pretraining and Evaluation
Partitions and overlap. None of the final-evaluation QCD jets is used to pretrain or finetune the models. The partitions are files of the official JetClass splits (
train_100M,val_5M,test_20M): every SSL method pretrains on the 10M unlabeled QCD jets oftrain_100M, with the 500k QCD jets ofval_5Mas the pretraining validation set, and all finetuned taggers are evaluated on top and QCD jets fromtest_20M, which holds 2M of each. To reduce the computational burden, the k-NN and linear probes of the pretrained backbones use a random subsample of 200k of the test-split samples, fit and scored by cross-validation. A.4 gives the pretraining size and states the disjointness, but we will revise A.4 to explicitly state the splits, the pretraining validation set, and the probe subsample.Hyperparameter selection. The reported model's teacher settings are indeed selected on the JetClass accuracy of this evaluation split in the current submission (A.17), which we agree should not enter any decision. As we show in Section 3, this selection is not needed: with DINOv2's default teacher setting, set before any of these runs is evaluated, RINO yields 0.843±0.021 on the same test split, against 0.840±0.015 for the tuned setting. Their means differ by 0.3 percentage points, well within the spread between pretraining runs of either one.
2. Access to Target-Domain Background
We agree and will qualify the claim accordingly in Sec. 1 and Sec. 4.1 of the revised PDF: the OOD result is a transfer in which pretraining has access to unlabeled background (QCD) jets from the target domain. This matches how the method is used in practice: the tagger is deployed on collider data, which are unlabeled, and self-supervised pretraining can use them because it needs no labels. In our benchmark, the unlabeled JetClass QCD jets are a proxy for QCD-dominated real data (Sec. 4.1). However, no JetClass label enters the pretraining or finetuning stage; the signal class never enters pretraining; and none of the 2M evaluation QCD jets appears in pretraining (1 above).
We also test RINO without this access. The model is pretrained instead on Aspen Open Jets [2], real jets from the CMS 2016 Open Data with no JetClass jets at all, and then finetuned on JetNet. RINO's accuracy is 0.813±0.021 on JetClass top vs. QCD over independent pretraining runs, against 0.840±0.015 when pretrained on JetClass QCD (± over pretraining runs in both). Without JetClass jets in pretraining, RINO's accuracy is about 3 percentage points lower, but it still transfers.
[1] M. Oquab, et al. DINOv2: Learning Robust Visual Features without Supervision. Trans. Mach. Learn. Res., 2024. arXiv:2304.07193.
[2] O. Amram, et al. Aspen Open Jets: unlocking LHC data for foundation models in particle physics. Mach. Learn.: Sci. Technol., 6:030601, 2025. arXiv:2412.10504.