{"id":"5281fbd7-cb69-4ceb-be26-f1747af4875d","arxiv_id":"2606.14223","paper_version":3,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":3.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Overview paper advocates advancing 6G ISAC to event-level sensing for behavioral semantics and intent prediction in IoV and LAE scenarios.","lead":"The paper argues that 6G integrated sensing and communications must shift from sensing discrete target parameters to event-level sensing that tracks continuous behaviors and predicts intent. This overview targets applications in vehicle networks and low-altitude systems to support more intelligent network services.","discovery_kind":"review","skeptic_critique":{"model":"grok-4.3","headline":"No rigorous technical distinction between target-level and event-level sensing is provided, leaving the necessity claim ungrounded","rationale":"The reader's weakest assumption directly matches the load-bearing point. Because the manuscript is a survey without new derivations, experiments, or formal verification, the absence of a precise technical demarcation does not alter the existing UNVERDICTED assessment.","tokens_in":1731,"tokens_out":283,"duration_ms":15356,"concrete_test":"From the section on fundamental concepts and sensing types, extract any equation or pseudocode that defines an event-level state model and its mapping to intent/behavioral output; if none exists or if it reduces to standard continuous-time tracking without an added semantic layer, recompute the motivation section's claims under the null that no new capability is introduced.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires that target-level sensing is inherently limited to fragmented snapshots without behavioral semantics, while continuous-time event-level modeling enables intent prediction. The paper introduces these as distinct paradigms in its overview of concepts and techniques but supplies no formal state-space definition, objective function, or algorithmic separation showing how event-level sensing produces semantic outputs that standard ISAC tracking (e.g., extended Kalman or particle filters on continuous trajectories) cannot. Without this, the asserted gap and the required advancement remain descriptive rather than demonstrated.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The manuscript claims that existing ISAC studies are limited to target-level sensing, which supplies only fragmented physical snapshots lacking behavioral semantics for intent interpretation, and that 6G ISAC must advance to event-level sensing via continuous-time state modeling to enable persistent recognition and prediction of target intent. It offers an overview of fundamental concepts, sensing types, scenarios, enabling techniques (waveform design, state estimation/tracking, event recognition), IoV/LAE applications, and future research directions.","tokens_in":1818,"tokens_out":392,"duration_ms":13782,"significance":"If the distinction can be formalized, the overview could help steer ISAC research toward semantically richer, intent-aware sensing for intelligent networks. The paper's value is in its synthesis of representative scenarios, applications, and cross-domain techniques rather than new derivations or experiments.","major_comments":[{"comment":"Introduction and fundamental concepts section: the claim that target-level sensing inherently lacks behavioral semantic capability while event-level sensing enables intent prediction rests on a narrative contrast, without a formal state-space definition, objective function, or explicit algorithmic separation from standard continuous-trajectory methods (e.g., extended Kalman or particle filters) already used in ISAC tracking.","section":"Introduction and fundamental concepts"},{"comment":"Key enabling techniques section (target state estimation and tracking; event recognition): no concrete example, reference, or derivation is supplied showing how event recognition produces semantic outputs that cannot be obtained by augmenting existing ISAC trackers with semantic post-processing, leaving the asserted necessity of the new paradigm ungrounded.","section":"Key enabling techniques"}],"minor_comments":[{"comment":"The term 'intelligent service engine' is used without a precise definition or citation to prior literature.","section":"Introduction"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the constructive feedback on our overview paper. We address the major comments point by point below, noting that the manuscript synthesizes concepts and scenarios rather than deriving new formalisms.","responses":[{"response":"We agree that the distinction is presented through conceptual narrative and scenarios rather than a formal state-space definition or objective function. As an overview, the section motivates the paradigm by highlighting limitations of fragmented physical snapshots versus persistent behavioral semantics. Standard trackers provide physical states but do not inherently yield intent prediction without additional semantic layers. We will revise the introduction to include a high-level state-space representation distinguishing physical estimation from event-level semantic modeling.","revision_made":"yes","referee_comment":"[Introduction and fundamental concepts] Introduction and fundamental concepts section: the claim that target-level sensing inherently lacks behavioral semantic capability while event-level sensing enables intent prediction rests on a narrative contrast, without a formal state-space definition, objective function, or explicit algorithmic separation from standard continuous-trajectory methods (e.g., extended Kalman or particle filters) already used in ISAC tracking."},{"response":"The section provides an overview of techniques with citations but does not include a specific derivation or example of necessity, consistent with the paper's role as a synthesis rather than a technical proposal. We acknowledge that a concrete illustration of how event recognition integrates into continuous-time modeling (versus post-processing) would better ground the distinction. We will add a brief conceptual example and reference in the revision.","revision_made":"yes","referee_comment":"[Key enabling techniques] Key enabling techniques section (target state estimation and tracking; event recognition): no concrete example, reference, or derivation is supplied showing how event recognition produces semantic outputs that cannot be obtained by augmenting existing ISAC trackers with semantic post-processing, leaving the asserted necessity of the new paradigm ungrounded."}],"tokens_in":1358,"tokens_out":403,"duration_ms":21121,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The paper's core move is to argue that target-level ISAC only gives snapshots and cannot handle intent or behavioral semantics, so the field needs event-level sensing that works on continuous-time states. It then reviews waveform design, state estimation, event recognition, and applications in IoV and LAE, plus some future directions.\n\nWhat it does is organize a set of existing techniques and scenarios under one label. That can help readers who want a single place to see how sensing might feed higher-level services.\n\nThe soft spot is the central claim. The text treats the gap between target-level and event-level sensing as self-evident, yet it gives no state-space model, objective function, or comparison showing that standard trackers (Kalman, particle filters on trajectories) cannot already produce the semantic outputs claimed to require a new paradigm. Without that, the necessity argument stays descriptive.\n\nThe paper is for people who follow 6G ISAC surveys and want an overview of where the community might head next. It is not carrying new derivations or measurements, so it will not change technical work directly. A serious editor could still send it to referees to check the coverage and citation balance, since the topic is timely even if the framing needs tightening.","headline":"This is a survey that names event-level sensing as the next step for 6G ISAC but supplies no formal separation from existing continuous tracking methods.","tokens_in":2309,"tokens_out":325,"would_cite":false,"duration_ms":9943,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"6G ISAC must advance from isolated target snapshots to continuous event-level sensing to interpret behavioral intent.","keywords":["ISAC","6G","event-level sensing","target-level sensing","environmental understanding","intent recognition","IoV","LAE"],"falsifier":"A demonstration that post-processed target-level data already achieves comparable accuracy in predicting target intent and behavior in IoV or LAE scenarios would undermine the need for the proposed shift.","tokens_in":2624,"feed_emoji":"📡","tokens_out":585,"duration_ms":12044,"temperature":0.7,"pith_summary":"The paper argues that existing integrated sensing and communications work in 6G provides only fragmented physical measurements of targets, which cannot support recognition of ongoing behaviors or intentions. It proposes shifting to event-level sensing that tracks states across continuous time to enable persistent prediction of semantic meaning. This change matters for mission-critical uses such as vehicle networks and low-altitude operations, where deeper understanding would let networks act as intelligent service engines rather than mere data collectors. The review outlines enabling techniques in waveform design, tracking, and recognition, then maps them to concrete scenarios.","feed_headline":"6G ISAC needs event-level sensing to read target intent","feed_subtitle":"Continuous-time modeling replaces fragmented snapshots so networks can predict behaviors in vehicles and low-altitude systems.","key_machinery":"Event-level sensing, which models continuous-time states to support recognition of behavioral semantics and intent.","core_discovery":"To bridge the gap between raw physical measurements and meaningful environmental understanding, ISAC in 6G must advance toward event-level sensing, which models continuous-time states to enable persistent recognition and prediction of target intent and behavioral semantics, overcoming the limitations of target-level sensing that supplies only fragmented snapshots.","pith_inferences":["Event-level sensing could require new data models that combine physical measurements with learned behavioral patterns over time.","Success would imply tighter coupling between sensing outputs and network decision layers for real-time adaptation.","Testing in controlled IoV or drone environments could quantify the accuracy gain in intent prediction compared with snapshot methods."],"forward_implications":["Waveform design, target state estimation, and event recognition techniques become central to realizing continuous sensing.","Applications in Internet of vehicles and low-altitude economy gain the ability to enhance downstream operational functions with semantic event information.","6G networks can evolve toward proactive, intent-aware services instead of reactive parameter estimation.","Future research must address integration across sensing types and scenarios to support intelligent evolution."],"fun_headline_variants":["6G ISAC shifts to event-level sensing for intent prediction","Event-level sensing models states to read 6G target behaviors","Beyond targets: ISAC event sensing for 6G semantic insight","Continuous-time ISAC enables behavioral recognition in 6G","6G ISAC adopts event-level approach for deeper understanding"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"Target-level sensing inherently lacks the ability to interpret behavioral intent, and event-level modeling can supply that capability in practical networks.","fun_headline_variants_meta":{"raw":{"variants":["6G ISAC shifts to event-level sensing for intent prediction","Event-level sensing models states to read 6G target behaviors","Beyond targets: ISAC event sensing for 6G semantic insight","Continuous-time ISAC enables behavioral recognition in 6G","6G ISAC adopts event-level approach for deeper understanding"]},"model":"grok-4.3","cost_usd":0.004987,"raw_usage":{"total_tokens":2436,"prompt_tokens":667,"num_sources_used":0,"completion_tokens":83,"cost_in_usd_ticks":49874500,"prompt_tokens_details":{"text_tokens":667,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":1686,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":667,"tokens_out":83,"duration_ms":14237,"temperature":1.0,"reasoning_tokens":1686,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-27T05:07:51.183583+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"A demonstration that post-processed target-level data already achieves comparable accuracy in predicting target intent and behavior in IoV or LAE scenarios would undermine the need for the proposed shift.","supporting_citations":[],"review_version":1}