- Rating
- 1
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
TETHER formulates few-shot motor-imagery EEG adaptation as target-conditioned, fixed-budget source-trial retrieval. A source-supervised encoder embeds covariance/tangent-space and bandpower features. Local semantic and cross-domain consensus hyperedges organize the source repository; a small labeled target support set retrieves and attaches to relevant hyperedges. Two HGNN branches, a relation gate, and a trial scorer are trained using source-only pseudo-target episodes and a differentiable weighted-ridge utility objective. At deployment, exact class-balanced selection provides a source subset, and a selected-source-fitted tangent/bandpower logistic classifier is trained with target support. Evaluation covers four-class BNCI2014-001 LOSO and binary cross-dataset LODO on BNCI-LR, Cho2017, and Lee2019-MI. The paper emphasizes adaptation accuracy, small selected-source budgets, and sub-second online processing.
This assessment concerns the 22-page manuscript and the supplied TETHER_code archive. It includes static inspection, a passing subset of 18 supplied tests, and limited synthetic checks of core retrieval functions, not reproduction of the reported EEG experiments. The archive explicitly omits recordings, checkpoints, and experimental outputs.
Critical strengths
1. A coherent task-specific retrieval formulation with an implemented learning mechanism. Reusing source-side organization while conditioning selection on a small labeled support set addresses a concrete adaptation problem. The local/consensus relations and downstream-utility supervision form a connected design rather than unrelated feature modules. The inspected implementation contains the stated normalized hypergraph propagation and differentiable ridge solve. Limited synthetic checks confirm agreement with the propagation equation, gradient flow through the utility mechanism, and feasible exact class-balanced selection; these checks do not validate benchmark accuracy.
2. Useful data-isolation provisions and informative controlled ablations. The final classifier fits its tangent reference and scalers on selected sources, and its fitting API excludes real target-query data. The retriever removes the pseudo-target domain from episode graphs/candidates while transparently allowing its earlier source-encoder pretraining. The release provides nested-support/trial-ID checks and documents seed-within-subject aggregation. Table 3 and its runner use the same final decoder for the main method and structural/utility ablations. The README is unusually explicit about comparison and preprocessing limitations. These are meaningful strengths, even though the principal comparisons and efficiency claims remain unresolved.
Critical weaknesses
1. The principal comparisons change the downstream model and adaptation weighting, so they do not establish the claimed benefit of source-trial retrieval.
The README explicitly states that traditional selector baselines use EEGNet with a fixed per-target-sample weight of 2, whereas the final TETHER results use tangent/bandpower features and logistic regression with 50% of the total training-weight mass assigned to target support (README.md, lines 279–295). This is confirmed by tether/lodo_training.py, lines 41–76, and tether/tangent_decoder.py, lines 83–112 and 124–177. The documented two-stage binary workflow first generates selections and intermediate EEGNet scores, then uses scripts/run_tangent_binary.py for the reported TETHER classifier; the README explicitly says the intermediate TETHER scores are not the reported results (lines 92–101).
For balanced binary B=500, the neural baselines' nominal target-weight fractions are 3.85%, 7.41%, and 13.79% at 5/10/20 shots per class, rather than 50%. These are nominal sample-weight fractions, not an assertion about exact gradient contributions under batch-normalized losses. Nevertheless, representation, classifier, optimization, and target emphasis all change together with selection. Thus, the large gaps in Tables 1–2 and Figure 4 cannot be attributed specifically to higher-order retrieval. Complete-system comparisons can be useful, but should be labeled and interpreted as such; the README acknowledges that they do not isolate the selector. The decisive missing comparison is to pass random, similarity-based, pairwise, and TETHER-selected subsets through the same final feature map/classifier, target-weight policy, and development budget. Native end-to-end methods such as RSSDA can remain additional system-level references.
The binary All-source reference is also inconsistent between manuscript and implementation. Appendix C.1 (p. 17) says it uses all source trials plus a fixed target reference support set. In the supplied runner, tether/lodo_bnci_seed0.py, lines 455–476, passes an empty target-index array and explicitly records baseline_protocol="source_only" and target_trials_used=0. README.md, lines 288–290, confirms this is not an all-source version of the target-adapted classifier. Consequently, comparisons with that row do not establish that selecting 2.35% of sources improves on equivalently adapted all-source learning. The percentage itself is a correct selected-subset ratio, but it is not the total historical data used to learn the encoder, graph, and retriever. An all-source reference using the same support and final decoder is needed; the source-only reference should be retained under an accurate label. The support definition underlying the shot-independent four-class All-source row also needs to be made explicit.
Table 3 provides more relevant within-pipeline evidence and should not be dismissed. However, removing HGNN propagation changes Macro BA from 0.5196/0.5407/0.5781 to 0.4948/0.5266/0.5618, substantially smaller changes than the headline cross-system gaps. Their importance should be assessed with paired subject-level effects and uncertainty, not inferred from marginal standard deviations. The no-utility ablation deliberately removes both utility training and the learned scorer, as disclosed in the paper, and is therefore not a one-factor isolation of a loss term. A non-hypergraph or pairwise retriever with the same frozen encoder and utility supervision would better distinguish the specific structural contribution. There is also a surrogate/deployment gap: training evaluates soft sample weights through ridge regression in frozen embeddings, whereas deployment uses hard selections and refitted tangent features with logistic regression. This is not an algebraic error, but evidence that the surrogate ranks subsets usefully for the final classifier would strengthen the central utility claim.
2. The efficiency and preprocessing claims are not aligned with the supplied measurement and execution paths.
Runtime boundaries. Section 4.2 and Appendix E.2 define the approximately 0.64-second online cost as source retrieval plus target adaptation, excluding query forward inference. In scripts/run_tangent_binary.py, selections are read and checked before timing (lines 398–434); the timer begins at line 439, includes feature-map/classifier fitting and query prediction, and ends at line 448. The recorded training_seconds therefore excludes retrieval and includes query inference. Baseline runtime fields include validation-stage and final-stage selection/training (tether/lodo_bnci_seed0.py, lines 949–960), and the compared classifiers are different. README.md, lines 307–318, explicitly warns that these fields cannot be compared directly as end-to-end online latency. This does not prove that Table 11 was computed from those exact fields, but the supplied timing path does not substantiate its advertised boundary or the approximately 9x retrieval/adaptation speedup. The table-producing script and measured phase-level logs, matched cache conditions, hardware/synchronization details, and the offline amortization denominator are needed.
Four-class preprocessing. Appendix D (p. 18) describes a common fourth-order zero-phase Butterworth pipeline and 0.5–3.0-second, 625-sample epochs. The documented four-class final configuration uses tmin=0.5, tmax=4.0 (configs/bnci2014_001_final.yaml, lines 20–29). Its extraction path uses raw.filter(method="fir", phase="zero") and passes those epoch limits directly to MNE (tether/data.py, lines 153–222). Thus the supplied four-class entry point does not implement the universally stated window/filter combination. The binary preprocessing path does implement the stated 0.5–3.0-second crop; this discrepancy should not be generalized to every task. The authors should identify the exact configuration/cache version underlying each table and reconcile filtering, referencing, trial normalization, and source-statistic fitting.
Signal-level partition dependence. For Cho2017, complete continuous class recordings are filtered/resampled before epochs receive reconstructed run labels (tether/lodo_data.py, lines 110–152 and 258–307); support/query partitioning occurs subsequently in tether/lodo_splits.py, lines 39–58. README.md, lines 297–305, acknowledges that the bidirectional preprocessing can couple signals across a partition boundary and that its effect has not been measured. Trial-ID disjointness is a useful check, but does not by itself establish strict signal-level independence. I have not established the size of this effect or demonstrated performance inflation. Partition-wise preprocessing or a targeted boundary-dependence check is needed to resolve this limited risk, without assuming that the complete test set contaminated training.
Finally, the code/report discrepancies cannot be resolved from aggregate means alone. Rechecking 170 displayed recall-to-Macro-BA rows found agreement within rounding, which is a positive internal consistency check; it does not verify the subject-level predictions, standard deviations, or Holm-adjusted tests. Result-matched configurations, prediction/selection outputs, and the statistics/timing scripts would allow the main comparisons to be audited. Omission of raw outputs is not by itself a reason for rejection; their importance here follows from the concrete protocol mismatches.
Overall assessment. I recommend rejection in the current form. The main concern is not an intrinsically invalid hypergraph algorithm or implausibly high accuracy: the current evidence does not separate source selection from different downstream learners and adaptation weights, and the advertised efficiency/protocol definitions are not established by the supplied paths. The archive's transparency and working core functions are appreciated. I found no direct evidence of fabricated data or deliberate target-query-label use in model fitting. Matched-decoder comparisons and reconciled, independently auditable execution/timing records could materially change the assessment.
Questions for authors
1. Matched-selector evaluation. Can the existing random, similarity-based, pairwise, and TETHER-selected subsets be decoded with exactly the same selected-source-fitted tangent/bandpower map, logistic classifier, target mass, and tuning policy? Please provide paired subject-level changes. Which Tables 1–2/Figure 4 comparisons were intended as complete-system rather than selector-only comparisons?
2. All-source information budget. Why does Appendix C.1 describe target reference support for binary All-source while the runner passes an empty target set? Please identify the result-producing configuration and report all-source with the same k-shot support and final decoder. What support set produced the single four-class All-source reference?
3. Timing provenance. Which script and measured records produced Table 11? Please separate cached/source-side preparation, support encoding, source retrieval, decoder fitting, and query prediction, with identical measurement boundaries for competing methods. What offline cost and number of reuses define the amortized values, and are validation/development costs handled equivalently?
4. Preprocessing correspondence and isolation. Did the four-class results use the documented 0.5–4.0-second FIR configuration or the manuscript's 0.5–3.0-second Butterworth pipeline? Please provide configuration hashes, epoch shapes, and preprocessing manifests. For Cho2017, has partition-wise processing or a signal-boundary dependence test been performed?
5. Structural contribution and result traceability. Please provide paired effects for the existing ablations and, if available, a retriever with matched utility supervision but no hypergraph/pairwise-only structure. How well does the soft ridge utility correspond to performance of the hard-selected logistic decoder? Please identify the subject–seed prediction outputs and aggregation/statistics code underlying the principal means, uncertainty, and significance annotations. These requests concern the reported claims and existing datasets, not a new application domain.
10 comments
Response to Reviewer HAqy [2/N]
[Q1] Matched-selector evaluation.
Thank you for your question. We have added four-class LOSO experiments on BNCI2014-001 and provide the following clarifications.
1. Scope of interpretation of the original comparisons
The original Tables 1–2 and Figure 4 were intended to evaluate the overall performance of the complete TETHER framework and existing transfer methods under their respective experimental protocols. For the conventional selection baselines that use a common downstream EEGNet pipeline (Random, Subject-level Top-k + Random, Bandpower, Riemannian, Pairwise, MTS, and MEIS), the results can be used to compare the performance of different source-data selection strategies within that decoding pipeline. RSSDA is evaluated using its specialized adaptation pipeline. All-source and Subject-level Top-k + All Trials serve as reference methods unconstrained by a fixed source-data budget, providing performance references when all available source trials are used. Thus, the original experiments primarily support comparisons of the overall performance of different complete methods and their respective selection strategies. The additional experiments with common downstream decoding further control the feature mapping, final classifier, and target support sample weights to analyze the contribution of the source-trial selection strategy itself more directly.
2. Additional comparisons under a common downstream protocol
All methods use the same source-data budget, target support set, query set, and random seeds. The final Tangent Space / Bandpower feature mappings are fitted using the source trials selected by each method. All methods use Logistic Regression, the same proportion of total weight assigned to target support samples, and parameter configurations fixed in advance. Tables 1–2 report the comparisons for B=500 and B=1000, respectively. TETHER achieves the highest mean Macro BA in all evaluated 5-, 10-, and 20-shot settings. These results show that, under common downstream decoding conditions, TETHER's target-conditioned source-trial selection continues to provide favorable overall adaptation performance, offering further evidence for the effectiveness of its retrieval strategy.
Table 1 Retrieval comparisons with matched downstream decoders: four-class LOSO on BNCI2014-001, B = 500
| Shot | Method | Left Hand | Right Hand | Both Feet | Tongue | Macro BA |
|---|---|---|---|---|---|---|
| 5 | Class-balanced Random-B | 0.4438 ± 0.1918 | 0.5494 ± 0.1803 | 0.4086 ± 0.1518 | 0.4988 ± 0.2902 | 0.4752 ± 0.1418 |
| 5 | Bandpower Top-B | 0.3886 ± 0.2050 | 0.5324 ± 0.2155 | 0.3892 ± 0.1551 | 0.5296 ± 0.2800 | 0.4600 ± 0.1385 |
| 5 | Riemannian Top-B | 0.4404 ± 0.2535 | 0.5389 ± 0.2281 | 0.4457 ± 0.1947 | 0.5481 ± 0.2907 | 0.4933 ± 0.1595 |
| 5 | Pairwise kNN Graph Diffusion | 0.4481 ± 0.2121 | 0.4596 ± 0.1987 | 0.4623 ± 0.2473 | 0.4735 ± 0.3088 | 0.4609 ± 0.1514 |
| 5 | TETHER | 0.4830 ± 0.1417 | 0.5654 ± 0.1610 | 0.4691 ± 0.1358 | 0.5608 ± 0.2231 | 0.5196 ± 0.1425 |
| 10 | Class-balanced Random-B | 0.4972 ± 0.2284 | 0.5506 ± 0.2266 | 0.4627 ± 0.1950 | 0.5423 ± 0.2601 | 0.5132 ± 0.1556 |
| 10 | Bandpower Top-B | 0.4503 ± 0.2553 | 0.5660 ± 0.2152 | 0.4287 ± 0.1871 | 0.5528 ± 0.3023 | 0.4995 ± 0.1545 |
| 10 | Riemannian Top-B | 0.4836 ± 0.2686 | 0.5466 ± 0.2484 | 0.5046 ± 0.2168 | 0.5821 ± 0.2721 | 0.5292 ± 0.1617 |
| 10 | Pairwise kNN Graph Diffusion | 0.4873 ± 0.2410 | 0.4904 ± 0.2247 | 0.5136 ± 0.2594 | 0.5204 ± 0.2968 | 0.5029 ± 0.1551 |
| 10 | TETHER | 0.5457 ± 0.1892 | 0.5627 ± 0.1960 | 0.4849 ± 0.1558 | 0.5694 ± 0.2433 | 0.5407 ± 0.1440 |
| 20 | Class-balanced Random-B | 0.5049 ± 0.2153 | 0.5753 ± 0.2434 | 0.5404 ± 0.2001 | 0.5840 ± 0.2651 | 0.5512 ± 0.1533 |
| 20 | Bandpower Top-B | 0.4590 ± 0.2703 | 0.5852 ± 0.2283 | 0.5244 ± 0.2327 | 0.5932 ± 0.2837 | 0.5404 ± 0.1568 |
| 20 | Riemannian Top-B | 0.4821 ± 0.2605 | 0.5633 ± 0.2701 | 0.5769 ± 0.2440 | 0.6154 ± 0.2511 | 0.5594 ± 0.1631 |
| 20 | Pairwise kNN Graph Diffusion | 0.5056 ± 0.2090 | 0.5485 ± 0.2144 | 0.5420 ± 0.2285 | 0.5861 ± 0.2639 | 0.5455 ± 0.1493 |
| 20 | TETHER | 0.5485 ± 0.2103 | 0.5722 ± 0.2319 | 0.5673 ± 0.1898 | 0.6244 ± 0.2292 | 0.5781 ± 0.1301 |
Response to Reviewer HAqy [3/N]
Table 2 Retrieval comparisons with matched downstream decoders: four-class LOSO on BNCI2014-001, B = 1000
| Shot | Method | Left Hand | Right Hand | Both Feet | Tongue | Macro BA |
|---|---|---|---|---|---|---|
| 5 | Class-balanced Random-B | 0.4454 ± 0.2204 | 0.5614 ± 0.2124 | 0.4253 ± 0.1694 | 0.4830 ± 0.3031 | 0.4788 ± 0.1513 |
| 5 | Bandpower Top-B | 0.4136 ± 0.2406 | 0.5123 ± 0.2465 | 0.4198 ± 0.1789 | 0.5509 ± 0.2751 | 0.4742 ± 0.1558 |
| 5 | Riemannian Top-B | 0.4549 ± 0.2803 | 0.5198 ± 0.2595 | 0.4654 ± 0.2222 | 0.5182 ± 0.3047 | 0.4896 ± 0.1689 |
| 5 | Pairwise kNN Graph Diffusion | 0.4534 ± 0.2217 | 0.4883 ± 0.2240 | 0.4821 ± 0.2370 | 0.4602 ± 0.3082 | 0.4710 ± 0.1619 |
| 5 | TETHER | 0.5287 ± 0.1997 | 0.5127 ± 0.2101 | 0.4747 ± 0.1568 | 0.5769 ± 0.2218 | 0.5233 ± 0.1344 |
| 10 | Class-balanced Random-B | 0.4812 ± 0.2242 | 0.5645 ± 0.2429 | 0.4750 ± 0.2055 | 0.5373 ± 0.2480 | 0.5145 ± 0.1521 |
| 10 | Bandpower Top-B | 0.4682 ± 0.2572 | 0.5389 ± 0.2575 | 0.4676 ± 0.2061 | 0.5534 ± 0.2860 | 0.5070 ± 0.1629 |
| 10 | Riemannian Top-B | 0.4818 ± 0.2835 | 0.5327 ± 0.2628 | 0.5059 ± 0.2423 | 0.5571 ± 0.2899 | 0.5194 ± 0.1695 |
| 10 | Pairwise kNN Graph Diffusion | 0.4849 ± 0.2432 | 0.5130 ± 0.2300 | 0.5117 ± 0.2488 | 0.5244 ± 0.2952 | 0.5085 ± 0.1666 |
| 10 | TETHER | 0.5537 ± 0.2084 | 0.5463 ± 0.2337 | 0.5191 ± 0.1976 | 0.6111 ± 0.2515 | 0.5576 ± 0.1373 |
| 20 | Class-balanced Random-B | 0.5019 ± 0.2137 | 0.5787 ± 0.2457 | 0.5543 ± 0.1996 | 0.5830 ± 0.2469 | 0.5545 ± 0.1529 |
| 20 | Bandpower Top-B | 0.4852 ± 0.2728 | 0.5481 ± 0.2613 | 0.5457 ± 0.2552 | 0.5932 ± 0.2593 | 0.5431 ± 0.1613 |
| 20 | Riemannian Top-B | 0.4864 ± 0.2719 | 0.5636 ± 0.2779 | 0.5772 ± 0.2604 | 0.5920 ± 0.2597 | 0.5548 ± 0.1623 |
| 20 | Pairwise kNN Graph Diffusion | 0.5191 ± 0.2275 | 0.5611 ± 0.2335 | 0.5556 ± 0.2290 | 0.5731 ± 0.2520 | 0.5522 ± 0.1552 |
| 20 | TETHER | 0.5775 ± 0.2036 | 0.5716 ± 0.2271 | 0.5753 ± 0.2338 | 0.6873 ± 0.1970 | 0.6029 ± 0.1393 |
After controlling for the downstream learner, TETHER continues to achieve the highest mean Macro BA under the tested conditions, supporting the overall effectiveness of its retrieval pipeline.
Response to Reviewer HAqy [1/N]
We sincerely appreciate your careful and technically detailed review, as well as your positive comments on our data-isolation safeguards and controlled ablations. Your questions helped us identify aspects of the experimental scope, implementation, and result traceability that required clearer explanation and validation. In response, we have added new analyses and revised the manuscript accordingly. Below, we address each point in turn and welcome any further questions or suggestions.
Thank you for your question, and we sincerely apologize for the confusion. The original experiments primarily evaluated the overall performance of the complete TETHER framework relative to existing methods. To further clarify the independent contributions of the retrieval strategy and its structural design, we have added retrieval comparisons under a common downstream decoding protocol (Q1), further clarified the target support set and information budget used by the All-source reference method (Q2), and further examined the roles of hypergraph structure and utility supervision through subject-level paired analyses of the existing ablation configurations, comparisons of different retrievers under the same utility supervision, and an analysis of the correspondence between the soft training surrogate and final decoding with hard selection (Q5). These additional analyses strengthen the original experimental evidence by addressing fairness in downstream evaluation, the contribution of retrieval structure, and consistency between training and deployment. The additional experiments further show that, under common downstream decoding conditions, TETHER maintains high mean Macro BA across the evaluated source-data budgets and support-set sizes. Meanwhile, the retriever comparisons under the same utility supervision provide more direct evidence for analyzing the additional value of the dual-relation hypergraph relative to unstructured retrieval and simple relation modeling. These results enable us to distinguish more carefully among full-system performance, the retriever's contribution, and the contribution of the structural design.
Thank you for carefully examining this part of the work. To address the correspondence between the original timing records and the actual execution pipeline, we have reviewed the relevant execution entry points and retained records, established a staged timing protocol that distinguishes source-side preparation, target-side online calibration, and query inference, and explicitly described the scope of caching, the cost of building offline resources, and their actual reuse counts (Q3). To address the inconsistency in preprocessing protocols, we have rechecked the raw GDF data and processing configurations for the four-class task, corrected the descriptions of the filters, time windows, and epoch data shapes actually used, and added a boundary-dependence check for continuous-signal preprocessing in Cho2017 (Q4). In addition, we have further organized the prediction outputs at the subject and random-seed levels and their correspondence with metric computation, statistical aggregation, and significance testing to improve the traceability and verifiability of the reported results (Q5).
These checks clarify the actual measurement scope of the efficiency metrics in the paper, distinguishing online calibration time, query inference time, and source-side offline preparation costs. The preprocessing checks identify and correct inaccuracies in the original description of the four-class processing pipeline and quantify cross-boundary dependence in the original continuous-signal processing for Cho2017. Together with the organization of prediction outputs and statistical analysis records, this additional work further clarifies how the experimental results were obtained and the scope to which they apply, avoiding interpretations of runtime efficiency, data isolation, and statistical conclusions that extend beyond the available evidence.