{"id":"8ea55feb-9e09-4d31-ad9a-87778079b248","arxiv_id":"2605.14900","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"COREKG uses sensitivity-based importance sampling from coreset theory to generate personalized KG summaries that achieve higher query accuracy and structural coverage than prior methods while using only a tiny fraction of the original graph.","lead":"The paper presents COREKG, which adapts coreset theory to build personalized knowledge graph summaries by sampling triples according to user-specific query sensitivity scores. A smart generalist might read it to see how large graph data can be reduced for efficient, user-tailored question answering and visualization.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"No significant objection identified","rationale":"Reader correctly flagged the sensitivity-to-bounded-error step as the load-bearing assumption and noted the abstract-only limitation. No additional internal flaw is visible from the given material.","tokens_in":1766,"tokens_out":221,"duration_ms":16894,"concrete_test":"Re-read the full manuscript (methods + appendix) and extract the precise definition of sensitivity, the coreset size bound, and any theorem stating the (1+ε) guarantee; confirm whether the proof invokes a standard coreset result for the query class used or supplies a new derivation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The abstract states that sensitivity-based importance sampling produces a coreset with bounded approximation error for a user-specific query workload. Without access to the full methods, proofs, or sensitivity definition, no internal inconsistency or unsupported step can be isolated from the provided text. The claim is a direct adaptation of known coreset techniques; whether the adaptation succeeds is an empirical and formal question left open by the abstract alone.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The manuscript presents COREKG, which adapts coreset theory to personalized KG summarization. For a dataset and user-specific query workload, it constructs per-user summaries by sampling triples via sensitivity-based importance sampling, claiming the resulting subset approximates the full KG with bounded approximation error. Summaries are built independently per user from sensitivity scores measuring triple importance w.r.t. the workload. Evaluation on Freebase, WikiData, and DBpedia reports higher query-answering accuracy and structural coverage than GLIMPSE, PPR, iSummary, PEGASUS, and APEX² while using only a tiny fraction of the original graph.","tokens_in":1821,"tokens_out":321,"duration_ms":25720,"significance":"If the bounded-error guarantee can be established and the empirical gains are reproducible under detailed protocols, the work would supply a principled, workload-aware method for reducing KG size while preserving query utility. This could benefit storage-constrained and latency-sensitive KG applications such as question answering and visualization.","major_comments":[{"comment":"Abstract: the central claim that the sampled subset 'approximates the characteristics of the full dataset with bounded approximation error' is asserted without any derivation, explicit sensitivity definition, algorithm, or error bound; this is load-bearing for the contribution and cannot be assessed from the given text.","section":"Abstract"}],"minor_comments":[{"comment":"The abstract contains the LaTeX fragment 'APEX$^2$' which should be corrected to APEX² for readability.","section":null}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the constructive feedback. The single major comment highlights a valid issue with the abstract's presentation of the core theoretical claim. We address it directly below.","responses":[{"response":"We agree the abstract is too terse on this point. The full manuscript (Sections 3–4) defines sensitivity scores as the maximum influence of each triple on query answers under the workload, presents the importance-sampling algorithm, and derives the (1+ε) approximation bound via standard coreset analysis. To make the contribution assessable from the abstract, we will revise it to include a one-sentence reference to the sensitivity definition and the fact that the error bound follows from coreset theory (with a pointer to the relevant sections).","revision_made":"yes","referee_comment":"[Abstract] Abstract: the central claim that the sampled subset 'approximates the characteristics of the full dataset with bounded approximation error' is asserted without any derivation, explicit sensitivity definition, algorithm, or error bound; this is load-bearing for the contribution and cannot be assessed from the given text."}],"tokens_in":1352,"tokens_out":246,"duration_ms":14323,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main takeaway is that COREKG uses sensitivity-based importance sampling on triples to create separate summaries for each user based on their query workload, with the abstract asserting a bounded approximation error and stronger results than GLIMPSE, PPR, iSummary, PEGASUS, and APEX² on Freebase, WikiData, and DBpedia.\n\nWhat stands out as new is the explicit per-user construction rather than a single global summary. The motivation around reduced storage and faster queries for personalized needs is clear and practical.\n\nThe approach itself is a straightforward transfer of an established sampling idea to this setting, which is fine if the adaptation works. The claim that workload-derived sensitivity scores produce a coreset with controlled error is the central technical step.\n\nThe soft spot is that the abstract states the bounded error and superior performance without showing the sensitivity formula, the coreset algorithm, or any derivation. That makes it impossible to check whether the guarantees actually carry over to KG queries or if the sampling introduces bias for certain query types. The evaluation is described only at a high level, so it is hard to judge whether the workload generation, baseline implementations, or accuracy metrics are set up in a way that supports the comparison.\n\nThis paper is aimed at researchers in knowledge graph management and data summarization who need smaller, query-aware subsets. A reader looking for a practical sampling method in that subfield would find the direction useful once the details are filled in.\n\nI would send it to peer review. The problem is real and the technique is a reasonable fit, but the formal and experimental claims need checking by referees who can see the full methods and proofs.","headline":"The paper adapts sensitivity sampling from coresets to build per-user KG summaries and claims better accuracy than baselines, but the abstract leaves the error bound and sensitivity definition unshown.","tokens_in":2288,"tokens_out":414,"would_cite":false,"duration_ms":22842,"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":"Sensitivity-based importance sampling produces personalized knowledge graph summaries that approximate full datasets with bounded error for user queries.","keywords":["knowledge graph summarization","coreset construction","personalized summarization","sensitivity sampling","query workload","approximation error"],"falsifier":"Measure the actual query-answering error on the sampled subset for the target workload and check whether it exceeds the error bound predicted by the sensitivity sampling procedure.","tokens_in":2666,"feed_emoji":"🕸️","tokens_out":604,"duration_ms":20846,"temperature":0.7,"pith_summary":"The paper adapts coreset theory to create small, user-specific summaries of large knowledge graphs. For any dataset and a given user's query workload, it samples triples according to sensitivity scores that capture each triple's importance to that workload. The resulting subset is guaranteed to approximate the full graph's characteristics up to a bounded error. A reader would care because the summaries require far less storage and deliver faster queries while preserving accuracy tailored to individual users.","feed_headline":"Sensitivity sampling yields tiny personalized KG summaries","feed_subtitle":"For each user's query workload the method selects a small triple subset that keeps query accuracy within a provable bound.","key_machinery":"Sensitivity-based importance sampling that constructs a coreset for a user query workload by weighting triples according to their contribution to answering queries in the workload.","core_discovery":"For a given dataset and a user-specific query workload, the approach samples a relevant subset of triples using sensitivity-based importance sampling. The subset approximates the characteristics of the full dataset with bounded approximation error. Sensitivity scores are defined to measure the importance of each triple with respect to the user's query workload, and these scores drive the coreset construction algorithm. Summaries are constructed independently for each user based on their query behaviour.","pith_inferences":["The same sampling procedure could be applied to streaming or frequently updated graphs by recomputing sensitivities only on changed triples.","Summaries built for similar users might be merged to reduce redundant storage while still respecting individual workloads.","The bounded-error guarantee could be tested by evaluating performance on queries drawn from a distribution close to but distinct from the original workload."],"forward_implications":["The sampled subset preserves query-answering accuracy with bounded error for the given workload.","Structural coverage of the original graph is retained in the summary.","Storage size and query runtime drop substantially compared with the full graph.","The method yields higher query accuracy and coverage than GLIMPSE, PPR, iSummary, PEGASUS and APEX² on Freebase, WikiData and DBpedia."],"fun_headline_variants":["Sensitivity sampling creates personalized KG coresets","User queries guide compact KG summary coresets","Importance sampling yields bounded KG summaries","Coreset method personalizes KG summaries per workload"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"Sensitivity scores defined with respect to a user's query workload can be used to construct a coreset whose approximation error remains bounded for that workload.","fun_headline_variants_meta":{"raw":{"variants":["Sensitivity sampling creates personalized KG coresets","User queries guide compact KG summary coresets","Importance sampling yields bounded KG summaries","Coreset method personalizes KG summaries per workload"]},"model":"grok-4.3","cost_usd":0.003054,"raw_usage":{"total_tokens":1689,"prompt_tokens":720,"num_sources_used":0,"completion_tokens":51,"cost_in_usd_ticks":30537000,"prompt_tokens_details":{"text_tokens":720,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":918,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":720,"tokens_out":51,"duration_ms":7975,"temperature":1.0,"reasoning_tokens":918,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-30T20:29:34.969763+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Measure the actual query-answering error on the sampled subset for the target workload and check whether it exceeds the error bound predicted by the sensitivity sampling procedure.","supporting_citations":[],"review_version":1}