{"id":"78cfe0fa-f663-4456-a6cc-b9612ce82124","arxiv_id":"2508.08408","paper_version":2,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":5.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"A feedback-control technique for apparent horizon tracking is introduced for asynchronous parallel binary black hole evolutions.","lead":"The paper presents new methods for tracking black hole horizons during binary black hole merger simulations, designed for asynchronous parallel computing. It describes a feedback control system implemented in the SpECTRE code.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Abstract-only evidence insufficient to support stability and accuracy claims of feedback-controlled horizon tracking.","rationale":"The reader's verdict was UNVERDICTED because only the abstract was reviewed. My stress-test agrees: the most load-bearing concern is not a specific mathematical flaw but the complete absence of evidence for the central claim's key conditions. The abstract promises a feedback control system for horizon tracking in asynchronous parallel evolutions. To believe this works, one needs to know that the controller remains stable and accurate for the intended range of physical configurations, and that the asynchronous task scheduling does not introduce stale-data errors. Neither is addressed in the abstract. A concrete check would be to inspect the full text for numerical experiments, stability analysis, and tests with artificial asynchrony. If those are present and convincing, the claim may hold; if absent, the claim is unsupported. This does not move the verdict from UNVERDICTED because no new evidence has been assessed. The reader's weakest_assumption already identified the stability/accuracy across merger geometries and the asynchronous race-condition risk, so I agree with that assessment.","tokens_in":623,"tokens_out":1199,"duration_ms":15353,"concrete_test":"Download the full text from arXiv and locate the sections describing the feedback control algorithm and its numerical tests. Verify that (1) the controller's gain schedule is either derived from a stability analysis or justified with robustness tests over a parameter sweep; (2) horizon accuracy is measured against a known analytic or high-resolution reference for at least one high-spin (a/M > 0.8) and one asymmetric-mass configuration; and (3) the asynchronous implementation is tested with injected artificial delays to confirm stale data does not degrade accuracy or cause controller instability.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that a feedback control system dynamically tracks apparent horizons in asynchronous parallel BBH evolutions. For this claim to hold, the controller must be stable and accurate across the full range of merger geometries, including high spin and mass asymmetry, without manual tuning. The abstract provides no algorithmic details, no stability argument, and no numerical evidence. In an asynchronous task-based environment, an additional condition is that the controller tolerates stale or out-of-order horizon-finder data; otherwise race conditions could corrupt the tracked horizon. Because the full text is unavailable, these load-bearing conditions are entirely unverified. This is not an assertion of failure, only a precise statement that the abstract leaves the central mechanism unsupported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript, of which only the abstract is available for review, proposes new methods for dynamically tracking apparent horizons in binary black hole (BBH) simulations using a feedback control system, specifically designed for SpECTRE's asynchronous task-based parallel execution. The abstract claims these methods improve speed, efficiency, and accuracy of numerical relativity simulations, and it briefly mentions extension to binary neutron star (BNS) simulations. No equations, algorithmic details, numerical results, or code are provided in the available material.","tokens_in":768,"tokens_out":2171,"duration_ms":28347,"significance":"If the claimed feedback-control horizon tracker were demonstrated to be stable, accurate, and robust in asynchronous parallel evolutions, it would be a valuable contribution to numerical relativity, potentially improving the efficiency of BBH waveform production for next-generation gravitational-wave detectors. However, the abstract alone provides no mechanism, no stability argument, and no numerical evidence. The paper's significance therefore cannot be assessed from the submitted material. No machine-checked proofs, reproducible code, or falsifiable predictions are visible in the abstract.","major_comments":[{"comment":"The central claim — that a feedback control system 'dynamically track[s]' apparent horizons — is unsupported by any control-theoretic stability analysis or convergence argument. The abstract gives no indication of the controller structure, the state being controlled, or the conditions under which tracking is guaranteed. This is load-bearing for the claimed novelty and reliability.","section":"Abstract"},{"comment":"In an asynchronous task-based runtime, the horizon finder may receive stale or out-of-order data. The abstract does not state how the feedback controller or the horizon finder handles such asynchrony, so the claim that the method 'runs correctly and efficiently' in SpECTRE's environment is not yet substantiated.","section":"Abstract"},{"comment":"No numerical results are presented: no waveforms, horizon radii, error measurements, convergence tests, or performance comparisons. Without these, the assertions of speed, efficiency, and accuracy cannot be verified.","section":"Abstract"},{"comment":"The method is stated to work for 'evolutions of BBH mergers' generally. No parameter coverage is mentioned — e.g., spin magnitudes, mass ratios, eccentricity — nor whether manual tuning is required. A feedback controller's performance typically depends on gains and system dynamics, so robustness across the merger parameter space is a central concern that the abstract leaves open.","section":"Abstract"}],"minor_comments":[{"comment":"The phrase 'next-generation detectors will be substantially more accurate' is vague; specifying the relevant error budget or detector characteristics would better motivate the need for faster numerical relativity.","section":"Abstract"},{"comment":"The BNS extension is mentioned in a single sentence but no details are given. Either expand this in the full text or soften the assertion to avoid an unsubstantiated promissory note.","section":"Abstract"},{"comment":"The word 'novel' is used without contextualizing prior horizon finders (e.g., AHFinderDirect, fast-flow methods) or explaining what is new about the feedback-control approach relative to the synchronous algorithms in the literature.","section":"Abstract"}],"recommendation":"uncertain","confidential_remarks":"This report is necessarily based on the abstract alone because no full text was made available. The concerns listed are not accusations of error; they reflect the absence of the evidence needed to judge the paper. If the full manuscript supplies the missing algorithmic details, stability discussion, and numerical validation, the claims may be entirely sound. I recommend obtaining the full text before making a final editorial decision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper addresses a real problem: SpECTRE's asynchronous task-based parallelism breaks the usual synchronous approach to apparent horizon tracking, and the authors propose a feedback control system to fix it. That is a sensible idea and a useful thing to try. The abstract is clearly written and the motivation is honest—next-generation detectors need faster and more accurate waveforms, and horizon tracking is a known bottleneck. I'll take the novelty claim at face value; I don't know of another code doing controlled horizon tracking in an asynchronous environment, so this might genuinely be new.\n\nWhat the abstract doesn't give us is any evidence. No equations, no convergence tests, no stability argument, no error bars. The stress-test note is right that the key risk is whether the controller remains stable and accurate across merger geometries with high spin or mass asymmetry, and whether it tolerates stale or out-of-order horizon-finder data in the asynchronous schedule. Those are load-bearing conditions, and the abstract says nothing about them. That's not a flaw in the paper—it's an abstract—but it means I can't vet the central claim from this document alone.\n\nWhat I can say is that the group is credible and the problem is important. The authors are clearly serious about the asynchronous redesign; they don't just bolt on a parallel horizon finder, they think about the algorithm differently. If the full paper includes a description of the controller, a stability or at least a robustness analysis, and tests across a range of binaries, this would be a solid contribution to numerical relativity. If it doesn't, the referees should send it back. The soft spot I want to flag is exactly the one the stress-test note points to: the interaction between the feedback loop and the asynchronous scheduler. That is the part I'd want to see explained carefully.\n\nMy recommendation: send it to peer review. It's a real method paper from a serious group, and the referees can judge the technical content. I can't verify anything from the abstract, but I don't see a reason to desk reject. I'd bring it to a reading group only after the full text is out, because abstract-level discussion won't get us far.","headline":"A plausible engineering contribution from the SpECTRE group, but the abstract alone can't support the stability and accuracy claims—send to review and let the full paper speak.","tokens_in":1185,"tokens_out":1349,"would_cite":false,"duration_ms":17244,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proposes feedback control systems for tracking apparent horizons in binary black hole simulations under asynchronous task-based parallelism.","keywords":["apparent horizon","horizon tracking","binary black holes","numerical relativity","asynchronous parallelism","task-based parallelism","feedback control","gravitational waves"],"falsifier":"Evolve a binary black hole merger with high mass ratio and high spin in the asynchronous code, tracking the horizon with the proposed feedback method, and compare the horizon location and area at each time step to the same evolution run with a synchronous horizon finder. If the two diverge beyond the discretization tolerance, or if the controller must be manually reset, the central claim is falsified.","tokens_in":582,"feed_emoji":"🕳️","tokens_out":4501,"duration_ms":48441,"temperature":0.7,"pith_summary":"The paper addresses a practical problem in numerical relativity: next-generation gravitational-wave detectors need waveform simulations that are both much more accurate and much faster. The authors argue that making a spectral, task-based simulation code fast requires asynchronous parallelism, but algorithms like horizon tracking were designed for synchronous execution and fail when run that way. Their proposed solution is a feedback control system that dynamically tracks apparent horizons, adjusting the horizon search to the current state of the simulation as tasks complete in arbitrary order. If this works, horizon tracking no longer forces global synchronization, so binary black hole and binary neutron star simulations can run more efficiently on large supercomputers without losing accuracy.","feed_headline":"Feedback control tracks black-hole horizons in async runs","feed_subtitle":"A control loop replaces synchronous horizon searches, letting binary black hole simulations run without global barriers.","key_machinery":"Feedback control system: a loop that uses the most recently computed apparent horizon (or its proxy, such as the coordinate center and radius) as sensor input, applies a control law to set the next search domain, and thereby tracks the horizon continuously as the binary evolves. It carries the argument by replacing the old synchronous, one-shot search with an adaptive loop that survives arbitrary task completion order.","core_discovery":"The central claim is that apparent horizons in binary black hole mergers can be tracked dynamically through a feedback control system designed for asynchronous task-based execution. In contrast to traditional horizon finders that run in fixed lockstep with the evolution, the proposed method treats horizon finding as a control problem: it receives the current estimated horizon location and shape, predicts where to look next, and corrects for drift as the simulation proceeds. The authors state that this allows horizon tracking to run correctly and efficiently in an asynchronous environment, and they sketch how the same approach transfers to binary neutron star simulations.","pith_inferences":["The same feedback design could track other moving regions in computational physics, such as shock fronts or adaptive mesh boundaries, in codes with asynchronous scheduling.","A quantitative stability analysis of the controller—not present in the abstract—would be needed to guarantee behavior for extreme mass ratios or spins; this is the natural next test.","The approach may make waveform production pipelines more autonomous, since manual retuning of the horizon search is replaced by the controller's self-correction."],"forward_implications":["Horizon finding no longer requires a global synchronization point, eliminating a major idle-time bottleneck in task-based parallel evolutions.","Binary black hole waveforms can be produced with longer duration and higher accuracy, matching what next-generation detectors require.","The feedback control approach can be adapted to binary neutron star simulations run with asynchronous parallelism.","Simulation throughput on large supercomputers increases because resources are not waiting on horizon-finder barriers."],"supporting_citations":[],"fun_headline_variants":["Feedback loop tracks horizons in async BBH runs","Async horizon tracking via feedback control system","Control-based horizon tracking for async simulations","Horizon finding reimagined for async parallel evolutions","Feedback control keeps horizon tracks in async mergers"],"cache_read_input_tokens":2816,"weakest_assumption_plain":"The biggest assumption is that the feedback controller stays stable and accurate for every merger geometry—highly spinning, highly asymmetric, very eccentric—without manual retuning or falling back to synchronous execution, and that the asynchronous scheduler never hands the controller stale data.","fun_headline_variants_meta":{"raw":{"variants":["Feedback loop tracks horizons in async BBH runs","Async horizon tracking via feedback control system","Control-based horizon tracking for async simulations","Horizon finding reimagined for async parallel evolutions","Feedback control keeps horizon tracks in async mergers"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000513,"raw_usage":{"total_tokens":2275,"prompt_tokens":634,"completion_tokens":1641,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":378,"completion_tokens_details":{"reasoning_tokens":1573}},"tokens_in":378,"tokens_out":1641,"duration_ms":13597,"temperature":1.0,"reasoning_tokens":1573,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T21:29:50.841128+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Evolve a binary black hole merger with high mass ratio and high spin in the asynchronous code, tracking the horizon with the proposed feedback method, and compare the horizon location and area at each time step to the same evolution run with a synchronous horizon finder. If the two diverge beyond the discretization tolerance, or if the controller must be manually reset, the central claim is falsified.","supporting_citations":[],"review_version":1}