{"id":"4cbaed1f-8051-44c7-9d2b-4d43a0909a81","arxiv_id":"2606.30993","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":2,"one_line_summary":"A PKG-assisted RSMA scheme with joint trajectory-power-beamforming-compression optimization is developed for multi-user semantic communication in UAV networks and shown in simulations to beat conventional RSMA, NOMA, and SDMA on energy efficiency and semantic accuracy.","lead":"The paper proposes a UAV downlink framework that uses probabilistic knowledge graphs to split semantic data into shared and private parts and applies rate-splitting multiple access to deliver it under tight energy limits. A smart generalist might read it to see how future drone networks could cut energy use while keeping the meaning of transmitted information intact.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"Central performance claims rest on an unvalidated semantic accuracy metric (importance-weighted KG triples) whose correlation with actual reconstruction fidelity is untested.","rationale":"The reader's weakest assumption correctly isolates the least-secured link in the argument chain. Because the paper's headline superiority is simulation-driven and the simulations are defined using the proposed metric, any weakness there directly undermines the central claim without requiring further assumptions about optimization convergence or channel models.","tokens_in":1699,"tokens_out":298,"duration_ms":17221,"concrete_test":"On the simulation outputs already generated for the paper, recompute semantic performance using an alternative metric such as average triple overlap F1 or embedding cosine similarity on the reconstructed KGs; if the ranking versus baselines reverses or the reported gains drop below 5%, the original metric is the source of the claimed advantage.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The strongest claim is that simulations demonstrate superior semantic preservation and energy efficiency versus RSMA/NOMA/SDMA baselines. This rests on the proposed semantic accuracy metric and the assumption that PKGs faithfully decompose UE semantics into shared/private parts. If the weighting scheme is arbitrary or fails to reflect true semantic importance under UAV channel impairments and compression, then reported gains in \"semantic preservation\" become metric-dependent artifacts rather than evidence of framework superiority. No independent validation (e.g., against human judgments or standard semantic similarity scores) is indicated in the abstract or claimed results.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The manuscript proposes a UAV downlink semantic communication framework that employs probabilistic knowledge graphs (PKGs) to model UE semantics and decompose them into shared/private components. It develops a PKG-assisted RSMA transmission scheme, formulates a weighted energy minimization problem jointly optimizing UAV trajectory, power allocation, beamforming, and semantic compression ratio, and solves it via the iterative SWEO algorithm (Lagrangian dual decomposition + successive convex approximation). A semantic accuracy metric based on importance-weighted KG triples is introduced to quantify reconstruction reliability. Extensive simulations are claimed to demonstrate superior energy efficiency, semantic preservation, and performance versus conventional RSMA, NOMA, and SDMA baselines across network parameters.","tokens_in":1860,"tokens_out":625,"duration_ms":23464,"significance":"If the semantic accuracy metric is shown to correlate with actual reconstruction fidelity and the simulation results are made reproducible with proper statistical controls, the framework could advance practical designs for energy-constrained semantic UAV networks by leveraging RSMA's handling of semantic structure. The integration of PKGs for semantic decomposition and the joint optimization of communication/computation/flight energy are potentially useful contributions, though their impact depends on validation of the core metric.","major_comments":[{"comment":"Abstract and Simulation Results section: The central claim that the framework achieves 'superior energy efficiency' and 'enhanced semantic preservation' rests on simulations, yet no details are provided on error bars, baseline implementations (e.g., how NOMA/SDMA are realized with the same PKG decomposition), data exclusion rules, or convergence guarantees of the SWEO algorithm. This renders the performance comparisons uninspectable and load-bearing for the superiority claim.","section":"Abstract / Simulation Results"},{"comment":"Section introducing the semantic accuracy metric: The metric assigns importance-based weights to informative KG triples to quantify reconstruction reliability, but no validation is given against human judgments, standard semantic similarity scores (e.g., BERTScore or KG embedding distances), or ablation under modeled UAV channel impairments and compression. This directly undermines the 'enhanced semantic preservation' claim, as gains may be artifacts of the untested weighting scheme.","section":"Semantic Accuracy Metric"},{"comment":"Problem formulation and assumptions: The framework assumes PKGs accurately capture UE semantics and enable reliable shared/private decomposition for RSMA; however, no sensitivity analysis or comparison to alternative semantic models (e.g., deterministic KGs or embedding-based approaches) is reported to test this under fast-varying UAV channels.","section":"System Model / Problem Formulation"}],"minor_comments":[{"comment":"Notation for the semantic compression ratio and energy-component weights should be clarified with explicit definitions and ranges in the optimization problem statement.","section":"Problem Formulation"},{"comment":"Figure captions for simulation results should include the exact parameter settings (e.g., number of Monte Carlo runs, specific values of free parameters) to aid reproducibility.","section":"Simulation Results"}],"recommendation":"major_revision","confidential_remarks":"The manuscript introduces a custom semantic accuracy metric without external validation, which raises questions about whether the performance gains are robust or metric-specific; this should be addressed before publication in a signal-processing journal."},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the constructive comments on our manuscript. We address each major comment point by point below and will incorporate revisions to improve reproducibility, validation, and robustness as indicated.","responses":[{"response":"We agree that the simulation results require additional details for reproducibility and transparency. In the revised manuscript, we will expand the Simulation Results section to include error bars (mean ± standard deviation over 100 Monte Carlo runs), explicit descriptions of baseline implementations using identical PKG decomposition for NOMA and SDMA, confirmation that no data were excluded, and a dedicated convergence analysis with plots for the SWEO algorithm. These changes will make the performance comparisons fully inspectable while preserving the original simulation outcomes.","revision_made":"yes","referee_comment":"[Abstract / Simulation Results] Abstract and Simulation Results section: The central claim that the framework achieves 'superior energy efficiency' and 'enhanced semantic preservation' rests on simulations, yet no details are provided on error bars, baseline implementations (e.g., how NOMA/SDMA are realized with the same PKG decomposition), data exclusion rules, or convergence guarantees of the SWEO algorithm. This renders the performance comparisons uninspectable and load-bearing for the superiority claim."},{"response":"The importance-weighted KG triple metric is introduced as a structure-aware measure tailored to PKGs. We recognize that external validation would strengthen the claims. In the revision, we will add a subsection providing comparisons of the proposed metric to BERTScore and KG embedding distances on representative reconstructions, along with an ablation study evaluating its sensitivity to UAV channel impairments and varying compression ratios. This will demonstrate correlation with semantic fidelity.","revision_made":"yes","referee_comment":"[Semantic Accuracy Metric] Section introducing the semantic accuracy metric: The metric assigns importance-based weights to informative KG triples to quantify reconstruction reliability, but no validation is given against human judgments, standard semantic similarity scores (e.g., BERTScore or KG embedding distances), or ablation under modeled UAV channel impairments and compression. This directly undermines the 'enhanced semantic preservation' claim, as gains may be artifacts of the untested weighting scheme."},{"response":"PKGs are foundational to the probabilistic shared/private decomposition that RSMA leverages. To address robustness under fast-varying channels, the revised version will include a sensitivity analysis on PKG parameter variations and a new comparison subsection benchmarking against deterministic KG and embedding-based alternatives. These additions will test the assumptions explicitly in the UAV setting.","revision_made":"yes","referee_comment":"[System Model / Problem Formulation] Problem formulation and assumptions: The framework assumes PKGs accurately capture UE semantics and enable reliable shared/private decomposition for RSMA; however, no sensitivity analysis or comparison to alternative semantic models (e.g., deterministic KGs or embedding-based approaches) is reported to test this under fast-varying UAV channels."}],"tokens_in":1531,"tokens_out":611,"duration_ms":28688,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The core new piece is the PKG-assisted RSMA scheme that splits UE semantics into shared and private parts via probabilistic knowledge graphs, then folds that structure into a single optimization over UAV trajectory, power allocation, beamforming, and semantic compression ratio. They solve the resulting non-convex weighted-energy problem with the SWEO algorithm that alternates Lagrangian dual decomposition and successive convex approximation. A secondary addition is the importance-weighted KG-triple metric for semantic accuracy.\n\nThe joint energy model that includes communication, computation, and flight costs is a reasonable practical step for energy-constrained UAV settings, and the use of RSMA to match the shared/private decomposition is a direct fit. The claim of better energy efficiency and semantic preservation than plain RSMA, NOMA, and SDMA across parameter sweeps is the sort of concrete, subfield-useful result that can be checked once the code and parameters are available.\n\nThe soft spot is the semantic accuracy metric. It is defined inside the paper by assigning importance weights to KG triples, yet the abstract gives no test of whether those weights track actual reconstruction quality under channel errors or compression. If the weighting is arbitrary, the reported semantic-preservation gains become metric artifacts rather than independent evidence. Simulation details on baseline implementations, error bars, and PKG construction are also missing from the abstract, so the strength of the numerical claims cannot be judged yet.\n\nThis is for people already working on semantic communications or UAV resource allocation. A reader in that niche can extract the algorithm structure and the specific RSMA-PKG integration even if the metric needs more grounding. The work is coherent enough on its own terms to deserve a serious referee, though the metric validation will probably be the main point of revision.","headline":"The paper combines PKGs with RSMA in a UAV joint-optimization loop and adds a new importance-weighted KG metric, but the performance edge rests on that unvalidated metric.","tokens_in":2327,"tokens_out":428,"would_cite":false,"duration_ms":24816,"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":"A UAV semantic communication system using rate-splitting multiple access and probabilistic knowledge graphs minimizes energy while preserving more semantics than NOMA or SDMA.","keywords":["semantic communication","rate-splitting multiple access","UAV networks","probabilistic knowledge graphs","energy minimization","trajectory optimization","beamforming"],"falsifier":"An experiment or simulation in which the proposed scheme uses more total energy or achieves lower semantic accuracy than conventional RSMA under identical network parameters, channel realizations, and energy weights would falsify the performance claims.","tokens_in":2618,"feed_emoji":"📡","tokens_out":709,"duration_ms":24797,"temperature":0.7,"pith_summary":"The paper develops a downlink framework in which probabilistic knowledge graphs model user equipment semantics and split them into shared and private parts. Rate-splitting multiple access handles these parts to deliver multi-user semantic information under energy limits and fast-changing UAV channels. A joint optimization problem minimizes a weighted sum of communication, computation, and flight energy by tuning trajectory, power allocation, beamforming, and semantic compression ratio. An iterative algorithm combining Lagrangian dual decomposition and successive convex approximation solves the non-convex problem. A new semantic accuracy metric weights knowledge-graph triples by importance to measure reconstruction quality, and simulations show gains over standard RSMA, NOMA, and SDMA.","feed_headline":"Rate-splitting with knowledge graphs cuts UAV semantic energy use","feed_subtitle":"Joint optimization of flight path, power, and compression outperforms NOMA and SDMA while preserving more semantic triples.","key_machinery":"The PKG-assisted RSMA transmission scheme together with the SWEO algorithm that applies Lagrangian dual decomposition and successive convex approximation to jointly optimize UAV trajectory, power, beamforming, and compression ratio.","core_discovery":"Integrating probabilistic knowledge graphs with rate-splitting multiple access enables efficient decomposition and transmission of semantic information in UAV networks; the resulting weighted energy minimization problem is solved by the semantic-aware weighted energy optimization algorithm, yielding better energy efficiency and semantic preservation than conventional multiple-access schemes under the modeled conditions.","pith_inferences":["The same decomposition and optimization approach could be tested on terrestrial base stations where flight energy is absent but computation and transmission costs remain coupled.","If the PKG model is replaced by other semantic representations such as embeddings, the RSMA layer and SWEO solver might still apply with only the accuracy metric changed.","Extending the formulation to include user mobility on the ground would require adding a new term to the weighted energy objective while keeping the RSMA and PKG structure intact."],"forward_implications":["The joint optimization reduces the combined energy cost of flight, transmission, and computation while meeting semantic accuracy targets.","The semantic accuracy metric assigns higher weight to important KG triples, providing a finer measure of reconstruction reliability than unweighted error rates.","The framework outperforms benchmark RSMA, NOMA, and SDMA schemes across varied UAV altitudes, user densities, and energy budgets.","Decomposing semantics into shared and private components via PKGs allows RSMA to allocate common and private rates more efficiently than treating all data as private."],"fun_headline_variants":["PKGs and RSMA minimize UAV semantic energy","UAV semantic RSMA with probabilistic knowledge graphs","Semantic compression optimized in UAV rate splitting","Trajectory power allocation for UAV PKG RSMA"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"Probabilistic knowledge graphs accurately represent user equipment semantics and the importance-weighted KG-triple metric correctly quantifies reconstruction reliability under the assumed channel and energy models.","fun_headline_variants_meta":{"raw":{"variants":["PKGs and RSMA minimize UAV semantic energy","UAV semantic RSMA with probabilistic knowledge graphs","Semantic compression optimized in UAV rate splitting","Trajectory power allocation for UAV PKG RSMA"]},"model":"grok-4.3","cost_usd":0.008136,"raw_usage":{"total_tokens":3685,"prompt_tokens":646,"num_sources_used":0,"completion_tokens":55,"cost_in_usd_ticks":81362000,"prompt_tokens_details":{"text_tokens":646,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":2984,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":646,"tokens_out":55,"duration_ms":38410,"temperature":1.0,"reasoning_tokens":2984,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-01T01:36:24.254195+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"An experiment or simulation in which the proposed scheme uses more total energy or achieves lower semantic accuracy than conventional RSMA under identical network parameters, channel realizations, and energy weights would falsify the performance claims.","supporting_citations":[],"review_version":1}