{"id":"d6599355-8e2d-44da-bba0-972e4ae678d0","arxiv_id":"2606.07796","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":6,"one_line_summary":"Amplitude-based quantum-inspired beliefs fed to PPO cut simulated IoV cumulative damage and damage variance versus classical Bayesian belief under adaptive attackers.","lead":"This paper proposes Q-BIRD, a PPO defender that tracks hidden attacker intent with complex amplitude beliefs instead of classical Bayesian probabilities in partially observable vehicle networks. If the simulation gains hold, amplitude beliefs could stabilize adaptive cyber defense when attackers probe and evade.","discovery_kind":"new_application","skeptic_critique":{"model":"grok-4.5","headline":"Abstract/body numerical and platform conflict undermines the sole-cause ablation claim for amplitude belief.","rationale":"The reader correctly flags the underspecified fixed U_t and the abstract/body metric and platform contradictions as the soft spot under the central ablation claim. Those inconsistencies are load-bearing: if the numbers and environment are not stable, one cannot credit amplitude dynamics for the gains. No stronger internal mathematical contradiction is needed; the evidence package itself does not yet support the sole-cause statement. CONDITIONAL remains the right verdict—publishable after cleaned, single-platform results, full U_t specification, multi-seed statistics, and honest non-zero ASR—rather than ACCEPT or REJECT. The concrete test above would settle whether the concern lands.","tokens_in":16708,"tokens_out":644,"duration_ms":5631,"concrete_test":"Reconcile abstract vs Table I: re-run the exact ablation of §VI-D.1 (identical PPO, reward, attacker, env; only B vs U_t) under one declared platform, publish the full definition of U_t (matrix construction from s and a), multi-seed means/stds, and non-zero ASR. If the −60%/−90% gains and ASR=0 disappear or shrink below the abstract’s milder numbers once U_t is fixed and seeds are reported, the sole-cause claim fails.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The strongest claim attributes all gains (Table I: mean damage 27.495 vs 69.348, variance 3.636 vs 37.054, ASR 0.000, survival 1.000) solely to replacing Bayesian B with fixed amplitude update ψ_{t+1}=U_t ψ_t + Born rule, under identical PPO/architecture/attacker. That causal attribution cannot be trusted while the paper simultaneously reports incompatible numbers and platforms: abstract claims SUMO-OMNeT++/Veins with mean damage 28.0±3.0 vs 36.0±5.5, variance 6.0±1.5 vs 12.0±2.8, ASR 0.05±0.02, survival 0.96±0.02; body Table I and §VI-E report a CICIoV-23-calibrated custom simulator with perfect ASR/survival and much larger effect sizes; §VI opening still says quantitative results are deferred. U_t itself remains unspecified (how it is built from (s_{t+1},a_def) is never given; §VI-G admits it is fixed/hand-designed). Without a single consistent experimental record and a fully defined operator, the ablation cannot isolate “quantum-inspired belief” as the cause of the reported gains.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper formulates IoV cyber defense as a partially observable sequential attacker–defender problem and proposes Q-BIRD: a PPO defender whose input is augmented by an amplitude-based belief state ψ_t ∈ ℂ^H over hidden attacker intent, updated by a linear map ψ_{t+1}=U_t ψ_t with Born-rule probabilities b(θ_i)=|ψ(i)|². The central empirical claim is an ablation holding architecture, PPO training, reward, attacker model, and environment fixed: replacing classical Bayesian belief with the amplitude update yields large gains (Table I: mean cumulative damage 27.495 vs 69.348, variance 3.636 vs 37.054, ASR 0.000, survival 1.000). Supporting analyses include strategy-shift robustness, cost–security trade-offs, and post-hoc SHAP/LIME/Grad-CAM attribution arguing that belief features drive decisions when classical belief collapses.","tokens_in":17034,"tokens_out":1416,"duration_ms":15220,"significance":"If the ablation result holds under a fully specified, reproducible belief operator and a single consistent experimental record, the work would be a useful contribution to adaptive IoV defense: it treats the belief model as a designable component rather than a fixed Bayesian submodule, reports an honest negative result (random policy beating classical-belief PPO), and pairs RL defense with multi-method explainability. The formulation of long-horizon damage-plus-cost minimization under adaptive hidden intent is well motivated for V2X. Those strengths are currently undercut by internal numerical/platform conflicts and an underspecified U_t, so the claimed causal role of “quantum-inspired” belief is not yet established at journal standard.","major_comments":[{"comment":"Abstract vs body experimental record is inconsistent and load-bearing for the sole-cause claim. The abstract reports SUMO–OMNeT++/Veins results (mean damage 36.0±5.5→28.0±3.0, variance 12.0±2.8→6.0±1.5, ASR 0.05±0.02, survival 0.96±0.02, plus PDR/latency/throughput). §VI-E and Table I report a CICIoV-23-calibrated custom simulator with much larger effect sizes (69.348→27.495 mean, 37.054→3.636 variance, ASR 0.000, survival 1.000). §VI opening still states that quantitative results are “deferred to a subsequent revision.” Until one platform, one metric set, and one set of numbers are used throughout, the ablation cannot support the claim that amplitude belief alone causes the reported gains.","section":null},{"comment":"§V-C, Eqs. (13)–(14) and Algorithm 1: the amplitude operator U_t is never fully specified. The text says U_t encodes the impact of s_{t+1} and a_def_t, but does not give its construction (closed form, parameterization, dependence on observations, phase rules, or how interference is produced). §VI-G admits U_t is fixed/hand-designed rather than learned. Without a complete definition of U_t, the ablation (classical B vs amplitude update) cannot isolate “quantum-inspired belief” as the causal factor; the operator could act as an arbitrary regularizer. A reproducible definition of U_t (or a learned unitary-like map with training details) is required for the central claim.","section":null},{"comment":"§VI-B.3–4 and Table I: ASR and survival depend on an author-chosen damage threshold θ that is not stated numerically, and perfect ASR=0 / survival=1.000 on 80 test episodes is reported without confidence intervals or sensitivity to θ. Combined with forced attacker strategy shifts every 50 steps (§VI-C) and a fixed four-intent space, the headline stability gains may be sensitive to these free design choices. Report θ, sensitivity of ASR/survival to θ, and results without forced periodic shifts so the robustness claim can be assessed.","section":null},{"comment":"§VI-A / abstract communication metrics: packet delivery ratio, latency, throughput, and service availability appear only in the abstract (and are not tabulated or analyzed in §VI). Either provide the co-simulation protocol and tables that produce those numbers, or remove them from the abstract so claims match the evaluated environment.","section":null}],"minor_comments":[{"comment":"Title/abstract use “Amplitude-Belief” / Q-BIRD while the body title is “Belief-Space Quantum-Inspired Reinforcement Learning…”; align naming across front matter and body.","section":null},{"comment":"§V-H: complexity argument with H=4 is fine, but state explicitly that H is fixed by design and discuss scaling if intent cardinality grows.","section":null},{"comment":"Fig. 5–10 captions and §VI-F: SHAP magnitudes (e.g., ±10^4–10^5) are hard to interpret without stating whether logits are raw or scaled; add units or normalization notes.","section":null},{"comment":"References [12], [20], [21] and related work: ensure year/venue consistency and that comparisons to Guo et al. ASR=0.500 note platform differences more carefully in the main text, not only in a table footnote.","section":null},{"comment":"Notation: b_t is used both for classical simplex beliefs and for Born-rule probabilities extracted from ψ_t; a short notational distinction would reduce confusion in §V-C–E.","section":null},{"comment":"Typos/grammar: e.g., “whic states” (§V-C), “commonly and widely reported metrics” (abstract of body), and occasional missing spaces around equations.","section":null}],"recommendation":"major_revision","confidential_remarks":"The manuscript reads like a draft in which an abstract (SUMO/Veins numbers) and a body (custom simulator + Table I) were not reconciled, and §VI still contains a “results deferred” sentence. That is fixable but currently blocks trust in the ablation. Novelty relative to quantum-inspired RL surveys is incremental; the contribution is the IoV defense application and the belief-operator ablation, not a new quantum algorithm. Scope is appropriate for a security/ITS venue if experiments are made consistent and U_t is fully specified. I would not recommend accept or minor revision until those two load-bearing issues are resolved."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The one thing to know: this is a clean engineering idea—swap classical Bayesian intent belief for a fixed amplitude/Born-rule update inside PPO for partially observable IoV defense—and the ablation framing (same network, same PPO, same attacker, only the belief operator changes) is the right way to test it. The reported variance collapse is the interesting part if it holds.\n\nWhat is new is not quantum-inspired RL itself (Zhang et al. already survey that) or IoV RL defense (Guo, Chen, etc.). It is the concrete pairing of amplitude belief with cost-aware sequential mitigation under an adaptive four-intent attacker, plus the negative result that random can beat classical-belief PPO when the Bayesian posterior collapses under evasion. The formalization of ψ, U_t, and Born extraction is clear enough for a methods paper, the threat model is realistic, and the SHAP/LIME/Grad-CAM section is more careful than most RL-security work.\n\nThe soft spots are real and load-bearing. The abstract claims SUMO-OMNeT++/Veins with modest gains (damage 36→28, var 12→6, ASR 0.05, survival 0.96). The body Table I and §VI-E report a CICIoV-23-calibrated custom simulator with much larger effects and perfect ASR/survival 0/1. Section VI still says quantitative results are deferred. U_t is never fully specified from (s, a_def) and is admitted to be fixed/hand-designed, so the causal claim that “amplitude belief alone” produced the gains cannot be isolated. Free parameters (damage table, costs, threshold θ, forced shift period) are many; no multi-seed stats or code. Those are fixable, not fatal, but they currently undercut the strongest claim.\n\nThis is for people working on adaptive cyber defense and belief modules under partial observability, not for quantum information or general RL theory. It deserves a serious referee once the numbers and platform are made consistent, U_t is written down, and ASR is honest. I would not desk-reject it; I would send it out with a clear demand for a single experimental record. Worth engaging after revision, not before.","headline":"Useful ablation idea for amplitude belief in IoV defense, but abstract/body metric and platform conflicts make the sole-cause claim untrustworthy as written.","tokens_in":17706,"tokens_out":542,"would_cite":false,"duration_ms":5559,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"Amplitude-based belief states cut IoV defense damage variance by about 10× versus classical Bayesian updates under the same PPO policy.","keywords":["Internet of Vehicles","reinforcement learning","quantum-inspired belief","partial observability","intrusion detection","amplitude belief","PPO","adaptive attacker"],"falsifier":"Re-run the identical ablation (same PPO, reward, four-intent adaptive attacker, 600+80 episodes) after replacing the hand-designed U_t with either a pure classical Bayesian update or a randomly initialized learned unitary; if the large variance and ASR gaps disappear, the causal attribution to amplitude belief fails.","tokens_in":17538,"feed_emoji":"🛡️","tokens_out":956,"duration_ms":9066,"temperature":0.7,"pith_summary":"The paper argues that classical Bayesian belief tracking of a hidden attacker intent collapses under the deceptive, probing, and evasion traffic that adaptive adversaries produce in Internet-of-Vehicles networks, feeding overconfident and unstable inputs into a reinforcement-learning defender. It replaces that belief model with a quantum-inspired amplitude state: a normalized complex vector that is updated by a linear operator and only later turned into probabilities by taking squared magnitudes (Born rule). The resulting belief vector is concatenated with observable traffic features and given to a Proximal Policy Optimization agent that chooses cost-aware actions (monitor, alert, throttle, isolate). Holding architecture, reward, attacker model and environment fixed, the amplitude formulation is claimed to be the sole source of large measured gains: roughly 60 percent lower mean cumulative damage, roughly 90 percent lower damage variance, attack success rate driven to zero, and survival probability of one on held-out test episodes. Explainability tools (SHAP, LIME, Grad-CAM) are used to show that the belief features dominate decisions precisely when the classical posterior collapses. The practical claim is that non-classical uncertainty representations can stabilize sequential cyber defense in partially observable V2X settings without quantum hardware.","feed_headline":"Amplitude beliefs cut IoV damage variance ~10× vs Bayesian PPO","feed_subtitle":"Same policy, same attacker; only the uncertainty model changes, and attack success falls to zero.","key_machinery":"Quantum-inspired amplitude belief: a complex vector ψ_t updated by a linear operator ψ_{t+1}=U_t ψ_t and converted to intent probabilities by the Born rule b(θ_i)=|ψ(i)|²; this vector is concatenated with the observable state and supplied to the PPO policy.","core_discovery":"When a PPO defender is given an amplitude-based belief over four hidden attacker intents instead of a classical Bayesian probability vector, and every other component of the training loop is held fixed, mean cumulative damage falls from 69.348 to 27.495 and damage variance falls from 37.054 to 3.636, while attack success rate drops to 0.000 and survival rises to 1.000 on 80 test episodes.","pith_inferences":["If the amplitude operator were made learnable rather than fixed, the same architecture might track continuous or higher-dimensional intent spaces that the current four-state model cannot represent.","The finding that a random policy sometimes beats PPO-with-Bayesian-belief suggests classical belief collapse can be actively harmful; similar diagnostics may be useful in other POMDP security settings.","Because the method reports both security and V2X communication metrics (PDR, latency, throughput), it invites joint evaluation of defense actions against service-level agreements rather than security scores alone."],"forward_implications":["Defenders can keep uncertainty distributed across intent hypotheses during evasion windows instead of collapsing to an overconfident posterior.","Damage variance becomes a first-class security metric: low-variance policies are harder for adaptive attackers to probe and exploit.","The same amplitude-belief interface can be swapped into other partially observable RL defense loops without changing the policy architecture or requiring quantum hardware.","Explainability tools can be used to verify that belief features, not raw traffic spikes, drive mitigation decisions under strategy shifts."],"fun_headline_variants":["Amplitude beliefs cut IoV damage variance ~10× vs Bayesian PPO","Amplitude belief drops IoV damage variance 37→3.6 fixed PPO loop","Amplitude-belief PPO cuts mean IoV damage 69→27, ASR to 0","Same PPO same attacker: amplitude belief yields near-zero ASR","Amplitude beliefs over 4 intents cut V2X damage variance ~10×"],"cache_read_input_tokens":128,"weakest_assumption_plain":"The claim that the fixed, hand-designed linear amplitude operator is a faithful model of belief evolution under real V2X deception, rather than a simulator-specific regularizer that happens to help with four discrete intents.","fun_headline_variants_meta":{"raw":{"variants":["Amplitude beliefs cut IoV damage variance ~10× vs Bayesian PPO","Amplitude belief drops IoV damage variance 37→3.6 fixed PPO loop","Amplitude-belief PPO cuts mean IoV damage 69→27, ASR to 0","Same PPO same attacker: amplitude belief yields near-zero ASR","Amplitude beliefs over 4 intents cut V2X damage variance ~10×"]},"model":"grok-4.5","effort":"low","cost_usd":0.004472,"raw_usage":{"total_tokens":1406,"prompt_tokens":895,"num_sources_used":0,"completion_tokens":102,"cost_in_usd_ticks":44720000,"prompt_tokens_details":{"text_tokens":895,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":409,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":895,"tokens_out":102,"duration_ms":3838,"temperature":1.0,"reasoning_tokens":409,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-14T18:16:46.760844+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Re-run the identical ablation (same PPO, reward, four-intent adaptive attacker, 600+80 episodes) after replacing the hand-designed U_t with either a pure classical Bayesian update or a randomly initialized learned unitary; if the large variance and ASR gaps disappear, the causal attribution to amplitude belief fails.","supporting_citations":[],"review_version":2}