{"id":"fc966a46-bba0-448c-b9af-ed141f58b987","arxiv_id":"2605.22942","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":3.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"An auxiliary MLP for explicit world-to-image projection is added to a fusion transformer to boost buoy association performance on the MaCVi 2026 challenge test set.","lead":"This paper adds a small MLP called QueryMLP to a DETR transformer baseline to predict buoy pixel locations in images from chart data and IMU readings, then appends those predictions to the model's queries. The change aims to ease geometric learning and improves scores on a 2026 maritime vision challenge.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.3","headline":"No ablation or baseline comparison isolates whether QueryMLP predictions improve performance","rationale":"The reader's weakest assumption (accuracy and utility of QueryMLP outputs) is exactly the untested link required by the strongest claim. The absence of any isolating experiment keeps the verdict at UNVERDICTED; the proposed test directly measures whether that assumption holds.","tokens_in":1644,"tokens_out":284,"duration_ms":13263,"concrete_test":"Retrain or evaluate the identical DETR decoder twice on the held-out test set—once with the QueryMLP coordinates appended to the query vector and once without—then compare Overall, F1 and mIoU; a difference smaller than the run-to-run variance falsifies the claim that the spatial prior is load-bearing.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires that the appended QueryMLP pixel predictions supply a useful spatial prior that reduces the transformer's geometric load. The manuscript states only the final leaderboard numbers (Overall 0.7386) for the modified system and does not report (a) the unmodified baseline score, (b) an ablation that removes the appended coordinates, or (c) any pixel-level accuracy of the QueryMLP itself. Without these quantities it is impossible to verify that the modification is responsible for any gain or that the predictions are accurate enough not to inject noise.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper proposes a modification to the DETR-based fusion transformer baseline for the MaCVi 2026 Vision-to-Chart buoy association challenge. It introduces a dedicated QueryMLP, trained on chart measurements and IMU orientation data, to explicitly predict each buoy's waterline contact point in the image; these predicted pixel coordinates are appended to the baseline decoder query vectors to supply a direct spatial prior and reduce the transformer's implicit geometric projection burden. The modified system reports an Overall score of 0.7386 (F1 = 0.8055, mIoU = 0.6718) on the challenge held-out test set, placing second on the leaderboard.","tokens_in":1750,"tokens_out":559,"duration_ms":17348,"significance":"If the central claim holds, the work demonstrates a practical method for injecting an explicit learned world-to-image mapping into transformer decoders for geometric data-association tasks, which could generalize to other vision-to-chart or sensor-fusion settings. The competitive held-out performance suggests engineering utility, but the absence of isolating experiments prevents a clear assessment of whether the added component drives the result or merely accompanies other unstated changes.","major_comments":[{"comment":"Abstract and Results: the central claim that 'appending the predicted pixel coordinates from the QueryMLP ... reduces the geometric reasoning burden on the transformer decoder' cannot be evaluated because the manuscript reports only the final leaderboard numbers and provides neither (a) the unmodified baseline score, (b) an ablation that removes the appended coordinates, nor (c) any pixel-level accuracy or error statistics for the QueryMLP itself.","section":"Abstract / Results"},{"comment":"The soundness of attributing gains to the QueryMLP rests on the untested assumption that its predictions are sufficiently accurate to act as a useful prior rather than noise; without reported training details, validation metrics on the MLP, or comparison to the baseline decoder alone, this assumption remains unverified and load-bearing for the contribution.","section":"Abstract"}],"minor_comments":[{"comment":"The abstract states the approach is 'lightweight' but supplies no architecture diagram, layer counts, or training hyperparameters for QueryMLP, making reproducibility difficult.","section":"Abstract"},{"comment":"No error analysis or failure-case discussion is mentioned, which would help readers understand when the added spatial prior helps or harms association.","section":null}],"recommendation":"major_revision","confidential_remarks":"This appears to be a concise challenge-report rather than a full methods paper; the journal may wish to consider whether the scope and depth align with its typical standards for technical contributions."},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the comments. We provide point-by-point responses below.","responses":[{"response":"We acknowledge that isolating experiments are absent from the manuscript. The presented work focuses on the performance achieved by the modified system on the challenge test set. As the baseline was not re-implemented or evaluated by us, we cannot provide the requested comparisons. The second-place ranking serves as an indirect indicator of effectiveness. We will revise the text to avoid over-attributing the result to the QueryMLP without direct evidence.","revision_made":"partial","referee_comment":"[Abstract / Results] Abstract and Results: the central claim that 'appending the predicted pixel coordinates from the QueryMLP ... reduces the geometric reasoning burden on the transformer decoder' cannot be evaluated because the manuscript reports only the final leaderboard numbers and provides neither (a) the unmodified baseline score, (b) an ablation that removes the appended coordinates, nor (c) any pixel-level accuracy or error statistics for the QueryMLP itself."},{"response":"The manuscript does not report training details or validation metrics for the QueryMLP, as the emphasis was on the overall association performance. We accept that this leaves the assumption unverified. We will add a description of how the QueryMLP was trained in the revised version.","revision_made":"partial","referee_comment":"[Abstract] The soundness of attributing gains to the QueryMLP rests on the untested assumption that its predictions are sufficiently accurate to act as a useful prior rather than noise; without reported training details, validation metrics on the MLP, or comparison to the baseline decoder alone, this assumption remains unverified and load-bearing for the contribution."}],"tokens_in":1365,"tokens_out":402,"duration_ms":21621,"standing_objections":["unmodified baseline score on the test set","ablation removing the appended coordinates","pixel-level accuracy statistics for the QueryMLP"]},"desk_editor":{"model":"grok-4.3","letter":"The paper's core move is to train a small MLP on chart measurements and IMU data to output the expected image pixel for each buoy's waterline contact point, then concatenate those two numbers onto the existing world-space query vector fed to the DETR decoder. The stated goal is to hand the transformer an explicit spatial prior so it spends less capacity on learning the projection. The system reaches 0.7386 overall on the held-out test set and places second on the leaderboard. That is a concrete engineering step for this exact task, and the idea of injecting a geometric hint directly into the query is reasonable given how DETR-style models handle set prediction. The final numbers are at least competitive for the challenge. The weakness is straightforward and load-bearing: the manuscript gives no score for the unmodified baseline, no ablation that removes the appended coordinates, and no separate accuracy figure for the QueryMLP predictions themselves. Without those quantities it is not possible to tell whether the MLP is responsible for any of the reported performance or whether it is simply neutral or even slightly harmful. The stress-test note correctly flags this gap. The work is narrowly scoped to one maritime data-association challenge and does not test the approach on other datasets or tasks. Readers already competing in the MaCVi challenge might pick up the QueryMLP trick as a quick addition to try. Everyone else will find little to take away because the experiments do not isolate the claimed benefit. I would not bring this to a reading group, would not cite it, and would not send it for peer review until the missing controls are added.","headline":"This adds a QueryMLP to predict pixel locations and append them to DETR queries for a buoy challenge, but reports only the final score with no baseline or ablation to show the addition helps.","tokens_in":2198,"tokens_out":398,"would_cite":false,"duration_ms":16083,"reading_group":"no","serious_thinker":"yes","would_accept_peer_review":false},"rs_alignment":{"model":"grok-4.3","evidence":[],"headline":"Standard CV engineering (QueryMLP spatial prior in DETR) with zero overlap to RS forcing chain","alignment":"orthogonal","rationale":"The paper's central machinery is an empirical MLP (6→128→128→128→2) trained with SmoothL1 on chart+IMU features to predict waterline pixel coordinates, appended to DETR decoder queries. This is ordinary supervised learning for geometric projection in maritime vision; it invokes no recognition cost J(x), no φ-ladder, no 8-tick periodicity, no parameter-free constant derivations, and no distinction-to-spacetime forcing. RS has no theorems about transformer query augmentation or learned world-to-image mappings.","tokens_in":42000,"confidence":"high","tokens_out":158,"duration_ms":4692,"cache_read_input_tokens":128,"cache_creation_input_tokens":0},"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Appending QueryMLP-predicted pixel coordinates to decoder queries eases geometric projection for buoy association.","keywords":["buoy association","vision-to-chart","QueryMLP","DETR","world-to-image projection","data fusion","transformer decoder"],"falsifier":"Retraining the baseline without the appended pixel coordinates and checking whether the test-set Overall score falls materially below 0.7386.","tokens_in":2547,"feed_emoji":"🛟","tokens_out":594,"duration_ms":15108,"temperature":0.7,"pith_summary":"This paper demonstrates that a lightweight MLP can be trained to map chart measurements and IMU orientation directly to image pixel locations for each buoy. Appending those predicted coordinates to the existing decoder query vectors supplies an explicit spatial prior, so the transformer no longer has to discover the full world-to-image mapping on its own. The modification yields an overall score of 0.7386 on the held-out test set, placing second in the challenge. A reader would care because it shows a practical way to inject geometric knowledge into fusion transformers without redesigning the architecture.","feed_headline":"Pixel prior from QueryMLP lifts buoy association to 0.7386","feed_subtitle":"Appending learned image coordinates reduces geometric load on the decoder for second-place result","key_machinery":"QueryMLP, a dedicated MLP that explicitly predicts the buoy's waterline contact point in the image from chart measurements and IMU orientation data.","core_discovery":"The central claim is that training a dedicated QueryMLP to predict the buoy's waterline contact point in the image from chart measurements and IMU orientation data, then appending these coordinates to the baseline decoder query vector, supplies a direct spatial prior per buoy and thereby reduces the geometric reasoning burden on the transformer decoder.","pith_inferences":["The same explicit-pixel injection could be tested on other DETR-style association tasks that involve known 3-D to 2-D mappings.","If the MLP outputs are treated as soft priors rather than hard coordinates, the decoder might learn to down-weight them when they conflict with image evidence.","The technique presupposes reliable IMU orientation; performance would need re-evaluation on platforms lacking such sensors."],"forward_implications":["Decoder queries receive an explicit pixel-location prior for each buoy.","The transformer decoder faces a lighter geometric-projection task.","The method reaches an Overall score of 0.7386, F1 of 0.8055 and mIoU of 0.6718 on the held-out test set.","The approach places second on the challenge leaderboard."],"fun_headline_variants":["QueryMLP adds waterline pixel predictions to decoder queries","Learned image coordinates provide spatial prior per buoy","QueryMLP reduces geometric burden via explicit pixel coords","Waterline contact prediction appended for 0.7386 association score"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The QueryMLP produces pixel predictions accurate enough to aid the transformer without introducing new errors or requiring extensive additional training.","fun_headline_variants_meta":{"raw":{"variants":["QueryMLP adds waterline pixel predictions to decoder queries","Learned image coordinates provide spatial prior per buoy","QueryMLP reduces geometric burden via explicit pixel coords","Waterline contact prediction appended for 0.7386 association score"]},"model":"grok-4.3","cost_usd":0.00812,"raw_usage":{"total_tokens":3649,"prompt_tokens":587,"num_sources_used":0,"completion_tokens":62,"cost_in_usd_ticks":81199500,"prompt_tokens_details":{"text_tokens":587,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":3000,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":587,"tokens_out":62,"duration_ms":16052,"temperature":1.0,"reasoning_tokens":3000,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-05-25T05:47:40.295248+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Retraining the baseline without the appended pixel coordinates and checking whether the test-set Overall score falls materially below 0.7386.","supporting_citations":[],"review_version":1}