{"id":"e0e9939c-8694-4a49-9ee6-ef5c3a122fae","arxiv_id":"2605.24714","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":5.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Formulates AoI minimization in ISAC as MDP with threshold policy for single source and restless bandit with Whittle indices for multi-source, with truncation error bounds.","lead":"The paper optimizes age of information in an integrated sensing and communication system where a base station chooses sensing, communication, or joint modes each slot to minimize a cost combining information age and action costs. A smart generalist might read it to see how standard decision tools can structure policies for dual-function wireless systems that must balance freshness and resource use.","discovery_kind":"new_application","skeptic_critique":{"model":"grok-4.3","headline":"No significant objection identified","rationale":"The reader's weakest assumption is the modeling premise itself; once that premise is granted, the structural claim follows from standard MDP arguments that the paper asserts it has carried out. Because the query supplies only the abstract-level description, no concrete gap in the proof steps can be exhibited, so the appropriate action is to leave the UNVERDICTED verdict unchanged.","tokens_in":1808,"tokens_out":299,"duration_ms":25517,"concrete_test":"Extract the precise definition of 'ordered threshold structure' and the induction hypotheses used in the value-iteration argument (likely in the single-source section); verify that the one-step cost and transition kernel satisfy the required supermodularity / stochastic monotonicity conditions for every pair of actions.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is a structural result (ordered threshold policy) for an infinite-state discounted MDP whose state is the pair of AoIs. The modeling assumptions (three mutually exclusive modes with fixed success probabilities and additive costs, Markovian evolution fully determined by the two AoI coordinates) are standard for AoI problems and are sufficient for the usual inductive arguments on value-function monotonicity/submodularity that establish threshold structure. The subsequent truncation step is accompanied by an explicit error bound, which does not affect the existence proof for the infinite-state policy. No internal inconsistency or missing monotonicity condition is visible in the stated claim.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper studies AoI optimization in an ISAC system with three discrete-time modes (sensing, communication, joint) having fixed success probabilities and costs. For the single-source case it formulates a discounted infinite-horizon MDP whose state is the pair of AoIs, proves that the optimal stationary policy has an ordered threshold structure, and supplies a state-space truncation together with an explicit error bound that determines the required truncation size. For the multi-source case it casts the problem as a restless multi-armed bandit, derives both a Whittle-index policy (when indexability holds) and an approximate Whittle-index policy (when it does not), and presents numerical illustrations of the threshold structure and policy performance.","tokens_in":1920,"tokens_out":516,"duration_ms":13452,"significance":"If the threshold-structure proof and the truncation error bound hold, the work supplies a concrete structural result and a computationally tractable approximation for an infinite-state MDP arising in ISAC, which is a useful addition to the AoI literature. The explicit analytic determination of truncation size and the extension of Whittle indexing to the non-indexable regime are strengths that enhance practical applicability. These elements, together with the standard but carefully applied MDP and restless-bandit machinery, give the manuscript a solid technical foundation.","major_comments":[],"minor_comments":[{"comment":"The precise definition of the 'ordered threshold structure' (e.g., the partial order on the two-dimensional AoI state space and the monotonicity direction of the switching curve) should be stated explicitly in the single-source formulation section rather than left implicit in the proof.","section":null},{"comment":"The transition probabilities and immediate costs for each of the three modes are described qualitatively; writing the explicit four-tuple (p_s, c_s, p_c, c_c, p_j, c_j) and the resulting AoI update rules in a single displayed equation would improve verifiability of the MDP.","section":null},{"comment":"In the multi-source section the condition that distinguishes the indexable regime from the non-indexable regime is stated but not accompanied by a simple, checkable criterion on the per-source parameters; adding such a criterion would clarify when each policy is applicable.","section":null},{"comment":"Numerical figures would benefit from error bars or multiple random seeds to confirm that the reported performance gap between the approximate Whittle policy and the exact Whittle policy is statistically stable.","section":null}],"recommendation":"minor_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the positive summary, significance assessment, and recommendation of minor revision. No major comments appear in the report, so we have no specific points to address point-by-point. We will incorporate any minor editorial suggestions in the revised version.","responses":[],"tokens_in":1343,"tokens_out":60,"duration_ms":13344,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The core contribution is the structural result for the single-source case: they set up a discounted infinite-horizon MDP whose state is the pair of AoIs, model the three mutually exclusive modes with fixed success probabilities and additive costs, and show the optimal policy is monotone in an ordered-threshold sense. They also derive a concrete bound on the error introduced by truncating the infinite state space, which tells you how far out you need to go for a target accuracy. That combination is not just a re-labeling of prior AoI work; the two-dimensional state and the explicit truncation analysis are the parts that add something usable.\n\nThe modeling assumptions line up with standard AoI practice and are sufficient for the usual value-function monotonicity arguments, so the threshold claim looks reachable. The truncation bound is a practical plus because it avoids hand-waving about approximation quality. For the multi-source scheduling they move to restless bandits, supply both a Whittle index policy and an approximate version, and separate the indexable and non-indexable regimes. The numerics are said to confirm that the approximation stays competitive even outside the guaranteed regime.\n\nThe weakest part is that the multi-source analysis leans on the usual indexability conditions without showing that the ISAC cost structure automatically satisfies them; when they fall back to the approximate policy the justification is mostly numerical. The three-mode channel is a reasonable abstraction but treats the modes as strictly exclusive, which may understate hardware realities where sensing and communication can partially overlap. Nothing in the abstract suggests circularity or post-hoc fitting.\n\nThis is a focused incremental paper aimed at people already working on AoI scheduling inside ISAC or similar dual-function systems. It is technically grounded enough to merit referee time rather than a desk reject, even if the multi-source part needs tightening.","headline":"The paper proves an ordered threshold structure for the two-dimensional AoI MDP in a three-mode ISAC channel and supplies an explicit truncation error bound; the multi-source Whittle extension is more routine.","tokens_in":2427,"tokens_out":444,"would_cite":false,"duration_ms":18555,"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":"In single-source ISAC systems the optimal AoI policy has an ordered threshold structure on its two-dimensional state.","keywords":["age of information","integrated sensing and communication","Markov decision process","threshold policy","restless multi-armed bandit","Whittle index"],"falsifier":"A value-iteration computation on a sufficiently large finite truncation that produces an optimal policy whose action regions violate the claimed ordered threshold ordering for at least one pair of AoI values.","tokens_in":2704,"feed_emoji":"📡","tokens_out":702,"duration_ms":17627,"temperature":0.7,"pith_summary":"The paper studies how a base station should choose among sensing, communication, and joint modes to keep status information fresh at a monitor while paying different costs for each choice. It models the single-source case as an infinite-horizon discounted MDP whose state is the pair of ages of information at the monitor and at the sensor. The central result is a proof that the optimal stationary policy is monotone in a specific ordered-threshold sense across this two-dimensional state space. The authors also derive an analytic truncation of the state space that keeps the optimality gap below any prescribed level. For multiple sources the scheduling task is recast as a restless bandit and solved with both exact and approximate Whittle-index policies.","feed_headline":"ISAC systems admit threshold-structured AoI policies","feed_subtitle":"Single-source optimal policy orders decisions by two age values; truncation error is bounded analytically and multi-source scheduling uses W","key_machinery":"The two-dimensional AoI Markov decision process whose optimal stationary policy is proved to possess an ordered threshold structure.","core_discovery":"For the single source scenario, we formulate the problem as a Markov decision process with a two-dimensional AoI state and prove that the optimal stationary policy admits an ordered threshold structure in the AoI state space. Since the AoI evolves over an infinite space, we truncate the state space to reduce complexity and rigorously bound the resulting error. The analysis analytically determines the truncation size needed to keep the error below a given threshold. For the multi-source scenario, we formulate the scheduling problem as a restless multi-armed bandit and develop both a Whittle index policy and an approximate Whittle index policy.","pith_inferences":["The threshold structure may permit a low-memory lookup-table implementation on resource-limited base stations.","The same MDP formulation could be reused to study continuous-time or energy-harvesting variants by changing only the transition probabilities.","If the three-mode assumption is relaxed to allow mode-dependent reliability that varies with current AoI, the threshold property would have to be re-proved."],"forward_implications":["The optimal policy can be computed by searching only over candidate threshold pairs rather than over the full policy space.","The truncation size required for any target error can be calculated in closed form before running the algorithm.","In the multi-source case the Whittle-index policy is optimal when indexability holds and remains competitive when it does not.","The same structural result immediately yields a simple online scheduler once the thresholds are tabulated."],"fun_headline_variants":["Ordered thresholds optimize AoI in single-source ISAC","Truncation bounds AoI error for ISAC MDP","Whittle policies schedule multi-source ISAC updates","Two-age MDP proves threshold structure in ISAC"],"cache_read_input_tokens":64,"weakest_assumption_plain":"The system can be modeled as a discrete-time process with exactly three mutually exclusive modes whose success probabilities and costs are fixed constants independent of the current ages.","fun_headline_variants_meta":{"raw":{"variants":["Ordered thresholds optimize AoI in single-source ISAC","Truncation bounds AoI error for ISAC MDP","Whittle policies schedule multi-source ISAC updates","Two-age MDP proves threshold structure in ISAC"]},"model":"grok-4.3","cost_usd":0.003259,"raw_usage":{"total_tokens":1789,"prompt_tokens":756,"num_sources_used":0,"completion_tokens":54,"cost_in_usd_ticks":32587000,"prompt_tokens_details":{"text_tokens":756,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":979,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":756,"tokens_out":54,"duration_ms":8553,"temperature":1.0,"reasoning_tokens":979,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-30T11:51:34.716174+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"A value-iteration computation on a sufficiently large finite truncation that produces an optimal policy whose action regions violate the claimed ordered threshold ordering for at least one pair of AoI values.","supporting_citations":[],"review_version":1}