{"data":{"sci_version": "1.0", "schema": "https://corpus.bntrg.co.uk/sci/1.0/sci.schema.json", "id": "pmid:39850522", "kind": "science-record", "revision": 1, "content_hash": "sha256:1123f5d9c5bb5e833aa23b8f4b2d6b8986a4ba60f803678389ce3c407560db2d", "created_at": "2026-11-01T10:25:49Z", "updated_at": "2026-11-01T10:25:49Z", "generator": {"name": "corpus-assay-engine", "converter": "sep_to_sci_v1", "source_package_schema": "urn:bntrg:sep:1.0.0"}, "paper": {"state": "partial", "title": "VEGFA Gene Expression in Breast Cancer Is Associated With Worse Prognosis, but Better Response to Chemotherapy and Immunotherapy.", "doi": "10.14740/wjon1993", "pmid": "39850522", "authors": null, "journal": "World journal of oncology", "year": 2025, "abstract": null, "claims": [{"id": "claim-1", "statement": "the headline numerical result the paper states for its main finding", "kind": "qualitative", "state": "blocked", "target": {"value": null, "unit": null, "tolerance": null, "source": null}, "evidence": []}], "gaps": ["abstract", "authors"]}, "experiment": {"state": "partial", "objective": "the headline numerical result the paper states for its main finding", "steps": [{"id": "read_published_estimate", "order": 1, "name": "Read the published estimate and its interval from the source", "kind": "other", "state": "complete", "blocked_on": null, "description": "The number a replication must land on, and the sentence it came from.", "duration_hours": null, "parameters": {}, "inputs": ["source:pmid:39850522"], "outputs": ["the published estimate, its unit and its interval"], "instruments": [], "reagents": []}, {"id": "obtain_source_data", "order": 2, "name": "Obtain the source data behind the published result", "kind": "data-preparation", "state": "blocked", "blocked_on": "data:not_downloaded", "description": "The published result cannot be recomputed without the data the paper analysed.", "duration_hours": null, "parameters": {}, "inputs": [], "outputs": ["the source data with its provenance"], "instruments": [], "reagents": []}, {"id": "run_analysis", "order": 3, "name": "Run the analysis programme", "kind": "analysis", "state": "blocked", "blocked_on": "no owned computational capability is bound to this experiment", "description": "Re-run the paper's analysis on the source data and produce the registered result. | SEP input not carried into the asset list: materials:the source data behind the published result in pmid:39850522", "duration_hours": null, "parameters": {}, "inputs": [], "outputs": ["the registered result"], "instruments": [], "reagents": []}], "materials": [], "instruments": [], "reagents": [], "conditions": {}, "gaps": ["step:obtain_source_data blocked on data:not_downloaded", "step:run_analysis blocked on no owned computational capability is bound to this experiment", "step:run_analysis input 'materials:the source data behind the published result in pmid:39850522' does not resolve to a declared asset"]}, "data": {"state": "partial", "assets": [{"id": "source:pmid:39850522", "role": "manuscript", "format": "xml", "media_type": "application/xml", "uri": "https://doi.org/10.14740/wjon1993", "sha256": "sha256:93c2906bf13acf3edf78b8f7f220d67725cd822c5a9bcccf1ffee012d46c43fa", "bytes": null, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "10.14740/wjon1993", "version": null, "note": "Source read by the package builder: full_text.", "structure": {"access_state": "full_text"}}], "gaps": ["asset:source:pmid:39850522 not verified"]}, "analysis": {"state": "absent", "methods": [], "targets": [], "gaps": ["method", "targets"]}, "capability": {"state": "absent", "requirements": [], "gaps": ["capability"]}, "readiness": {"state": "partial", "tier": null, "blocking_axis": null, "blocked_on": null, "axes": {}, "can_run": {"dsl_operations": [], "assets": ["source:pmid:39850522"], "steps": ["read_published_estimate"]}, "must_build": {"dsl_operations": [], "converters": [], "readers": ["xml"], "capabilities": [], "data_gaps": ["source:pmid:39850522"]}, "gaps": ["corpus_record_absent"]}, "review": {"state": "partial", "machine_critique": {"status": "blocked", "providers": [{"provider": "minimax", "model": "MiniMax-M3.1-Flash", "prompt_version": "full-text-reading-v1", "prompt_hash": "51a2254e9995ba42fe5747f9542d8ea0ebe68d058dafefc550c9e5885583acfe", "source_hash": "93c2906bf13acf3edf78b8f7f220d67725cd822c5a9bcccf1ffee012d46c43fa", "raw_response_sha256": "sha256:bcebe2cf29f3abc090e8ce2be5f13994666b8adee8404f286c4c8f8ea0f55906", "raw_response_chars": 4593}, {"provider": "minimax_alt", "model": "MiniMax-M3", "prompt_version": "full-text-reading-v1", "prompt_hash": "c4b20149cfa26f496d969353bdb2fe119a15114c3140882b701f58a872bcaba3", "source_hash": "93c2906bf13acf3edf78b8f7f220d67725cd822c5a9bcccf1ffee012d46c43fa", "raw_response_sha256": "sha256:4c3fe764ebb3e6647ed14fae9b8733e059c8de047e257f6e40a8baa93a0fdec7", "raw_response_chars": 32237}], "target_providers": [{"provider": "minimax", "model": "MiniMax-M3.1-Flash", "prompt_version": "replication-target-v2", "prompt_hash": "0670df664ea5ccc6ade56ccedde64aa0d0953807c7a4bc1f753b62f752cd3260", "source_hash": "93c2906bf13acf3edf78b8f7f220d67725cd822c5a9bcccf1ffee012d46c43fa", "raw_response": "{\"parent_value\":7336,\"unit\":\"BC patients\",\"evidence\":\"A total of 7,336 BC patients across eight independent cohorts: ISPY2 (GSE173839), Sweden Cancerome Analysis Network-Breast (SCAN-B) (GSE96058), Molecular Taxonomy of Breast Cancer International Consortium (METABRIC), GSE25066, GSE163882, GSE34138, GSE20194, and The Cancer Genome Atlas (TCGA), were analyzed.\",\"target_result\":\"The study analyzed 7,336 breast cancer patients across eight independent cohorts.\",\"number_stated\":true}", "reading": {"parent_value": 7336, "unit": "BC patients", "evidence": "A total of 7,336 BC patients across eight independent cohorts: ISPY2 (GSE173839), Sweden Cancerome Analysis Network-Breast (SCAN-B) (GSE96058), Molecular Taxonomy of Breast Cancer International Consortium (METABRIC), GSE25066, GSE163882, GSE34138, GSE20194, and The Cancer Genome Atlas (TCGA), were analyzed.", "target_result": "The study analyzed 7,336 breast cancer patients across eight independent cohorts.", "number_stated": true, "source_locator": "chars 533-841 of source sha256 93c2906bf13acf3edf78b8f7f220d67725cd822c5a9bcccf1ffee012d46c43fa"}, "recorded_at": "2026-10-31T15:28:20.868058+00:00"}, {"provider": "minimax_alt", "model": "MiniMax-M3", "prompt_version": "replication-target-v2", "prompt_hash": "d4f63dad626c1b7bdfd767807a386ff1b5ab1f0d8c570737cccaa3526f14427c", "source_hash": "93c2906bf13acf3edf78b8f7f220d67725cd822c5a9bcccf1ffee012d46c43fa", "raw_response": "<think>We need answer JSON only. Need parse task carefully. Need find single numerical result replication registered target headline numerical result paper states for main finding. Source supplied is abstract/methods/results no obvious headline numerical result besides total 7,336, eight cohorts, P<0.02, FDR <0.25, version 4.3.1, significance 0.05. Need determine intended target perhaps headline numerical result main finding. User emphasizes \"single numerical result that a replication of the registered target has to reproduce. Report headline number only: value paper itself states for registered target, not calculated...\" Registered target generic \"the headline numerical result the paper states for its main finding\". Source text seems perhaps abstract-only and headline numerical result might sample size 7,336? Or perhaps survival result missing hazard ratio due truncated source. Need infer.\n\nWe must extract claims only supplied source. Need output fields. If no number for registered target, false, parent_value 0, unit/evidence empty, target_result says what text states instead. Is there a stated numerical result attached to main finding? Let's inspect.\n\nTitle VEGFA gene expression breast cancer associated worse prognosis, better response. Abstract:\nA total of 7,336 BC patients across eight independent cohorts analyzed. Calculated median stratify.\nHigh VEGFA associated with worse disease-free, disease-specific, overall survival in METABRIC, supported SCAN-B (all P < 0.02). This is the only numerical statistical result in abstract: all P < 0.02. Could \"headline numerical result\" be P < 0.02? The registered target wording likely generated benchmark asking identify a numeric result from paper. Need identify single headline number. Main finding's numerical result perhaps \"all P < 0.02\" for survival. Unit would \"P\" perhaps unit? Parent_value numeric only, cannot encode inequality exactly except evidence. Could set 0.02, unit \"P\". Is P a unit? They ask headline number only, unit paper prints. Perhaps this is expected.\n\nCould be 7,336 patients, which is headline cohort size but not \"result\". Replication target main finding likely requires reproduce a particular headline numerical result. We need parse benchmark patterns. The prompt likely designed to distinguish parent numerical result among papers, with target registered perhaps ambiguous? \"REGISTERED TARGET: the headline numerical result the paper states for its main finding\" likely target description, and source text has maybe one result. They expect numerical answer. Need report exact contiguous quotation containing number and attribution. We need quote whole clause. Main finding: High VEGFA worse survival, \"all P &lt; 0.02\". The text's headline numerical result. Number 0.02, unit perhaps \"P\" or \"P value\". Evidence exact clause.\n\nBut \"headline number only\" maybe asks number such as 7,336 (patients). This is stated prominently at start, but sample size is methodology, not result. Main finding only has p-value threshold. Let's reason.\n\nCould be abstract where detailed result omitted due source extraction perhaps actual article full text but supplied source includes only intro/abstract + methods sections. No effect sizes. The only quantitative finding is \"all P &lt; 0.02\". Thus likely that is target. Is p-value a \"unit\"? Could unit be \"P\" exactly printed. JSON parent_value 0.02, unit \"P\" or perhaps \"P value\". User says unit must be unit paper prints. Paper prints \"P &lt; 0.02\", no unit in conventional sense. Common outputs for numeric extraction use unit empty if dimensionless. But required unit string can be \"\". Yet number_stated true. They may expect \"P\" as unit? Need assess.\n\n\"Report the headline number only: the value the paper itself states for the registered target\" A p-value threshold is not exactly a result value perhaps yes. \"The evidence must be exact, contiguous quotation ... contains the number ... Attribution may be carried by text's own wording: a number may be attached to the target by a parenthetical list, by the word 'respectively', or by words that name the target directly. When that is how the text links them, quote the whole clause...\" This generic language suggests extraction task from a paper where numerical result has target label. Here target = headline numerical result main finding. Need identify number attached to main finding via wording. Quote perhaps: \"High VEGFA was associated with worse disease-free, disease-specific, and overall survival in the METABRIC cohort, with findings supported by the SCAN-B cohort, which also showed worse overall survival (all P &lt; 0.02).\" This links main finding and number. Then number 0.02 and unit perhaps \"P\". That seems plausible.\n\nBut phrase \"single numerical result that a replication of the registered target has to reproduce\" perhaps p-value cutoff isn't replication target; effect size would be. Could target be sample count? They may want \"7,336 BC patients\" as headline number of study, but replication of main finding does not require reproducing N exactly necessarily.\n\nLet's search mentally actual paper title and perhaps published abstract with detailed numerical findings. Maybe full article has hazards ratios etc omitted from supplied text? The supplied source seems abstract and selected sections, likely no Results details beyond abstract due perhaps text parsing issue. They explicitly provide all source. We can't invent. The result wording includes only P &lt; 0.02. So if there is any numerical result stated for main finding, it's 0.02.\n\nNeed ensure exact contiguous quote. Could quote full sentence:\n\"High VEGFA was associated with worse disease-free, disease-specific, and overall survival in the METABRIC cohort, with findings supported by the SCAN-B cohort, which also showed worse overall survival (all P &lt; 0.02).\"\nThis is exact. Parent_value 0.02. Unit \"P\" perhaps. target_result maybe concise \"all P &lt; 0.02\" or \"The high-VEGFA survival association was significant across the reported analyses (all P &lt; 0.02).\" User asks headline number only likely target_result should be e.g. \"P &lt; 0.02\". They request report headline number only; fields include parent_value, unit, evidence, target_result. We can set target_result \"P &lt; 0.02\" perhaps.\n\nCould parent_value represent 0.02 even though inequality? They ask value paper states; numeric value 0.02. Evidence preserves \"&lt;\". Fine.\n\nBut maybe number \"7,336\" with unit \"BC patients\" is more headline. Let's inspect title / abstract structure. Abstract numerical results:\n- total 7,336 patients across 8 cohorts.\n- median split.\n- survival all P&lt;0.02.\nNo effect metrics. If benchmark wants \"headline numerical result\", maybe headline numerical result means sample size? Yet \"main finding\" explicitly. The survival association is main finding, p&lt;0.02. Better response to chemo and immunotherapy has no p values. The high-VEGFA associations generally have no numeric. The only quantitative claim tied to main prognostic finding is P&lt;0.02.\n\nCould say number_stated false because no headline numerical result/effect size, only sample size and threshold. Prompt: \"If the text states no number for the registered target, set number_stated to false ... and say in target_result what the text does state instead.\" Does \"all P&lt;0.02\" count as number for main finding? Yes, likely. The target is \"headline numerical result the paper states for its main finding\", and p value is a numerical result. Replication would need reproduce significance, though p-values vary. But generic task asks headline number.\n\nLet's understand terminology \"parent_value\". Perhaps in source they expect one primary numeric value and unit. They ask \"parent_value\" not \"value\" perhaps a schema from benchmark for numerical claim. Parent value likely number itself. If p-value, unit maybe \"P\". They may accept.\n\nNeed output valid JSON object only, no markdown. Ensure no newline issue irrelevant. I'd choose:\n{\n\"parent_value\":0.02,\n\"unit\":\"P\",\n\"evidence\":\"High VEGFA was associated with worse disease-free, disease-specific, and overall survival in the METABRIC cohort, with findings supported by the SCAN-B cohort, which also showed worse overall survival (all P &lt; 0.02).\",\n\"target_result\":\"P &lt; 0.02\",\n\"number_stated\":true\n}\nIs \"unit\" must be unit paper prints: P printed, so yes. Could use \"P value\" although paper says \"P\". Better exact printed \"P\". Target result perhaps \"all P &lt; 0.02\" to retain inequality. Headline number only maybe just \"P &lt; 0.02\". Good.\n\nNeed consider 7,336 and eight. They ask \"single numerical result\". Could there be confusion from title? Let's possibly identify actual paper abstract online mentally. It could have exact:\n\"High VEGFA was associated with worse disease-free, disease-specific, and overall survival in METABRIC cohort ... (all P &lt; 0.02).\" Thus this is likely extracted as abstract result and benchmark asks numerical result. The main finding maybe \"worse prognosis\" and number p&lt;.02. So yes.\n\nWould \"unit\" for p-value be \"P\" or empty? In scientific reporting, p-values dimensionless; but unit as printed perhaps \"P\". User says unit must be unit paper prints. It prints uppercase P as label. So \"P\". For sample count it'd \"BC patients\". For p-value likely \"P\". We should not say \"P value\" if not exact? It says \"All P values\" elsewhere, so \"P\" appears. \"unit\" could be \"P\" safely.\n\nCould target_result be \"all P &lt; 0.02\" exactly. Fine.\n\nLet's formulate valid JSON.</think>\n\n{\"parent_value\":0.02,\"unit\":\"P\",\"evidence\":\"High VEGFA was associated with worse disease-free, disease-specific, and overall survival in the METABRIC cohort, with findings supported by the SCAN-B cohort, which also showed worse overall survival (all P < 0.02).\",\"target_result\":\"all P < 0.02\",\"number_stated\":true}", "reading": {"parent_value": 0.02, "unit": "P", "evidence": "High VEGFA was associated with worse disease-free, disease-specific, and overall survival in the METABRIC cohort, with findings supported by the SCAN-B cohort, which also showed worse overall survival (all P < 0.02).", "target_result": "all P < 0.02", "number_stated": true, "source_locator": "chars 948-1164 of source sha256 93c2906bf13acf3edf78b8f7f220d67725cd822c5a9bcccf1ffee012d46c43fa"}, "recorded_at": "2026-10-31T15:28:31.119083+00:00"}], "agreement": false, "disagreements": {"target_result": {"minimax": "High VEGFA expression was associated with worse disease-free, disease-specific, and overall survival in METABRIC, with the findings supported by worse overall survival in SCAN-B (all P < 0.02).", "minimax_alt": "High VEGFA was associated with worse disease-free, disease-specific, and overall survival in the METABRIC cohort, with findings supported by the SCAN-B cohort, which also showed worse overall survival (all P < 0.02).", "decisive": false}, "analysis_methods": {"minimax": ["Median-based stratification of cohorts into high- and low-VEGFA-expression groups", "Two-sided statistical tests with significance set at 0.05", "Log-rank test for survival analysis", "Gene set enrichment analysis (GSEA)"], "minimax_alt": ["Median VEGFA expression stratification into high- and low-expression groups", "Two-sided tests for P values", "Log-rank test for survival analysis"], "decisive": false}, "statistical_method": {"minimax": "Log-rank test", "minimax_alt": "Log-rank test for survival analysis", "decisive": true}, "parent_value": {"minimax": 7336, "minimax_alt": 0.02, "decisive": true}, "unit": {"minimax": "BC patients", "minimax_alt": "P", "decisive": true}}, "agreement_basis": {"policy": "decisive_fields", "decisive_fields": ["work_type", "number_stated", "parent_value", "unit"]}, "blinded": true}, "human_critique": {"status": "not_run", "reviewer": null, "signed_at": null, "conflict_checked": false, "commentary": null}, "stages": [{"key": "target_materials", "status": "pending", "reason": null, "evidence": null}, {"key": "independent_readings", "status": "blocked", "reason": "The independent readings are incomplete or disagree; the source evidence must resolve the difference.", "evidence": {"provider_errors": []}}, {"key": "source_resolution", "status": "pending", "reason": null, "evidence": null}, {"key": "protocol_reconstruction", "status": "pending", "reason": null, "evidence": null}, {"key": "control_execution", "status": "pending", "reason": null, "evidence": null}, {"key": "control_comparison", "status": "pending", "reason": null, "evidence": null}, {"key": "publication", "status": "pending", "reason": null, "evidence": null}], "gaps": ["review.human_critique", "review.machine_critique.blocked"]}, "execution": {"state": "absent", "status": "not_run", "mode": null, "runs": [], "gaps": ["execution.not_run"]}, "provenance": {"entities": [{"id": "package:b8da88d9723ee1feecce3e792d8b61948819c1dcb1a056a695b029c8ec207ee8", "type": "sep_package", "label": "SEP v1 package for pmid:39850522"}], "activities": [{"id": "activity:conversion:pmid:39850522", "type": "conversion", "used": ["package:b8da88d9723ee1feecce3e792d8b61948819c1dcb1a056a695b029c8ec207ee8"], "agent": "agent:corpus-assay-engine"}], "agents": [{"id": "agent:minimax", "type": "model", "label": "minimax"}, {"id": "agent:minimax_alt", "type": "model", "label": "minimax_alt"}, {"id": "agent:corpus-assay-engine", "type": "software", "label": "Corpus assay engine"}], "source_run_id": "run-20261023T122449232913", "dsl_catalogue_fingerprint": "343fb30756ba6ec3", "capability_table_fingerprint": "2cb3e42f047d2ecf", "snapshot_id": "ba34b49490d9bfaf"}, "lineage": {"parent_id": "2bb5c59f-6c60-4731-8971-315a026a4f0f", "parent_hash": "b8da88d9723ee1feecce3e792d8b61948819c1dcb1a056a695b029c8ec207ee8", "fork_of": null, "generation": 1}, "gaps": ["abstract", "asset:source:pmid:39850522 not verified", "authors", "capability", "corpus_record_absent", "execution.not_run", "method", "review.human_critique", "review.machine_critique.blocked", "step:obtain_source_data blocked on data:not_downloaded", "step:run_analysis blocked on no owned computational capability is bound to this experiment", "step:run_analysis input 'materials:the source data behind the published result in pmid:39850522' does not resolve to a declared asset", "targets"]},"meta":{"api_version":"v1","automation_ladder_snapshot":"ba34b49490d9bfaf","automation_ladder_caveats":["Method extraction is rule-based and has no hand-judged accuracy sample. Evidence bands order textual support; method agreement intervals are null (n=0), never probabilities. Work-type validation does not validate methods.","Operation coverage describes only explicitly named methods. It does not prove a complete protocol, compatible parameters, successful compilation or re-execution. Unrecognised or ambiguous methods require review.","Data reachability is a textual accession or public-data statement. No dataset was downloaded or checked for access, completeness or licence.","Capabilities are planning states seeded from the dated equipment register. No bench ownership, completed quote or partner booking is inferred. Procurement counts are dependencies, not guaranteed replications or financial returns.","Tier and blocker counts are deterministic assessments of this corpus, not measured agreement. R3 is reserved for procurement; unresolved methods and data require human work (R4). The most restrictive axis wins, with capability, method, data as tie-breaks. Resolving one blocker can leave another at the same tier."],"engine":"corpus_engine","engine_provider":"rule_based_v1+lex99e1195c7d6f","run_id":"run-20261023T122449232913","source":"esearch","source_detail":"PubMed mirror biomed_pubmed (read-only)","run_status":"complete","imported_at":"2026-11-03T20:16:52+00:00","generated_at":"2026-11-04T02:36:37+00:00","total_papers":5529,"total_assays_tracked":29,"automatable_assay_families":6,"validated":true,"caveat_count":9,"blocking_caveats":5,"caveats":[{"code":"low_evidence_dominates","basis":["4,152 of 5,529 classifications (75.1%) matched no cue evidence and sit at the 0.05 floor.","34 of 230 data_only calls (14.8%) are in that group."],"claim":"Work type and confidence bands","gloss":"Most records were classified from their title and framing alone, with assay-cue coverage at zero.","detail":"Most classifications in this run rest on no matched evidence at all. At the evidence floor the work type comes from title framing and a fixed prior, not from anything in the paper, so the published work-type split should be read as a rough triage signal rather than a measured distribution. The impact is bounded but not removed: since 2026-10-23 the engine excludes floor-confidence records from the shortlist entirely and publishes them separately as held out, so the recommendation is unaffected even though the census is.","status":"unverified","severity":"high"},{"code":"runtime_is_modelled","basis":["Runtime is summed from published per-assay hour figures. Timing awaits the first executed replication.","4,407 of 5,529 records (79.7%) model to 0.00 hours because no assay was matched, which the ranking then reads as \"cheapest\".","0 of 10 shortlisted records carry a 0.00-hour estimate."],"claim":"Estimated replication runtime","gloss":"The hour figures are estimates built from assay mentions; timing awaits the first executed replication.","detail":"The hour figures are a cost model derived from the bench steps implied by each matched assay family. The first executed replication will supply measured timing. Records without an assay match currently receive a zero-hour model value.","status":"unverified","severity":"high"},{"code":"signal_score_without_citations","basis":["5,529 of 5,529 records (100.0%) have no citation count from the source metadata."],"claim":"Signal score","gloss":"The signal score uses publication-type weight because the mirror supplied zero citation counts.","detail":"The source supplied no citation counts for any record in this run, so the signal score is publication-type weight only. The citation term of the ranking is inert here, and the shortlist is ordered by publication type alone.","status":"unverified","severity":"high"},{"code":"abstract_absent","basis":["1,667 of 5,529 records (30.2%) carry no abstract at all."],"claim":"Evidence base for classification","gloss":"Some records contain a title and publication type only, defining the available classification evidence.","detail":"Where the mirror contains a title and publication types only, those fields define the classifier evidence. A methods section is required for fuller automated and human review.","status":"unverified","severity":"medium"},{"code":"classifier_agreement_measured","basis":["Published two-way decision agreement 77.7% (310 of 399 decidable records) from a stratified hand-check of 438 real records.","Three-way diagnostic agreement 73.9% (288 of 390 decidable records).","data_only agreement 75.2% (100 of 133 decidable).","48 record(s) could not be decided from title and abstract and were excluded from the denominator.","56 analysis-only records were published under another work type, so the data_only share is not an upper bound."],"claim":"Published automatable decision","gloss":"This figure has been checked against a human reading the same papers, so we know how often it is right.","detail":"The published automatable / not-automatable distribution comes from a rule-based classifier that was hand-checked against a fixed ground-truth definition. The agreement rates below are measured, not estimated. The three-way work type remains a diagnostic and is reported separately.","status":"measured","severity":"high"},{"code":"llm_comparison_not_measured","basis":["rule_based_v1+lex99e1195c7d6f produced 5,529 of 5,529 classifications.","Measured on 438 labelled records; these are RULE-BASED figures, not model results."],"claim":"Rule-based versus LLM agreement","gloss":"Every number here is the rule engine’s own. Model comparison begins when a working API credential is connected and measured.","detail":"The published two-way decision agreed with a human reading on 310 of 399 decidable records (77.7%). The diagnostic three-way label agreed on 288 of 390 decidable records (73.9%). Every percentage in this run belongs to the rule engine. Model comparison begins when a working API credential is connected and measured.","status":"not_measured","severity":"high"},{"code":"confidence_is_a_band","basis":["low band: 135 of 210 hand-judged records were correct, 64.3% [57.6, 70.5] (95% CI).","medium band: 80 of 104 hand-judged records were correct, 76.9% [68.0, 84.0] (95% CI).","high band: 73 of 76 hand-judged records were correct, 96.0% [89.0, 98.7] (95% CI).","A fitted calibrated probability was built, measured and withdrawn: it overstated held-out accuracy and reorders nothing."],"claim":"Confidence band and evidence score","gloss":"The confidence band carries a measured accuracy interval; 0.85 represents relative rank strength.","detail":"Confidence on this run is a BAND plus its measured accuracy interval. The number stored alongside each record is a monotone evidence score that orders records; 0.85 represents relative rank strength. A fitted probability was withdrawn after it overstated held-out accuracy. Threshold on the band.","status":"measured","severity":"medium"},{"code":"three_way_split_is_diagnostic","basis":["The corpus is published at the two-way decision: automatable or not-automatable. The three-way split (data_only / physical / mixed) is published as a diagnostic with its measured reliability attached, not as a claim, because the rule engine's `mixed` call cannot be defended.","The engine called 22 record(s) mixed; 10 matched the human judgement (45.5%, 95% CI 26.9-65.3).","The engine's `mixed` call is right on 10/22 hand-judged records (45.5%, 95% CI 26.9%-65.3%); a constant `physical` guess scores 36.4%.","The engine's `mixed` call does not reliably beat a constant `physical` guess on the hand-judged records it labels mixed.","The sign of that difference flips between two independent samples, so the three-way refinement is not demonstrably informative.","The mixed stratum contributes a disproportionate share of errors relative to its size.","At title-only depth the physical/mixed split collapses: a title says that work was done, not whether it was also analysed."],"claim":"The published decision is two-way","gloss":"The three-way split is shown for diagnosis only. The published split is two ways.","detail":"Count and filter the corpus on automatable versus not-automatable. The data_only / physical / mixed split remains an internal engine diagnostic because the mixed call has lower measured reliability.","status":"measured","severity":"medium"},{"code":"automatable_flag_unproven","basis":["6 of 29 assay families are marked automatable now: meta_analysis, machine_learning, survival_analysis, pca, molecular_docking, differential_expression.","The flag is a rule over assay family and work type; validation begins with the first executed replication.","It inherits the 75.2% data_only agreement rate, because the flag requires the record to be analysis-only."],"claim":"Automatable now","gloss":"This flag marks assay families the rules classify as data-only. The first executed replication will validate it.","detail":"This flag marks families the rules consider replicable from data alone. Its validity depends on the analysis-only classification and will be measured through an executed replication.","status":"unverified","severity":"medium"}],"confidence_policy":"band_with_measured_interval","confidence_contract":{"policy":"band_with_measured_interval","summary":"Confidence is reported as a BAND with a measured accuracy interval. The raw evidence score supplies monotone ordering. A fitted point estimate was withdrawn after it overstated held-out accuracy.","evidence_score":"A monotone evidence score in [0,1] built from cue mass, axis separation and text coverage. Use it to order records. Interpret 0.85 as relative rank strength.","confidence_band":"low | medium | high. Consumers threshold on this value.","band_accuracy":"Wilson 95% interval on the measured agreement of the hand-judged records in this band.","calibrated_confidence":"WITHDRAWN. Reintroduction requires held-out evidence demonstrating reliable estimates across every bin."},"band_accuracy":{"low":{"n":210,"low":0.576,"band":"low","high":0.7046,"agree":135,"accuracy":0.6429},"high":{"n":76,"low":0.8903,"band":"high","high":0.9865,"agree":73,"accuracy":0.9605},"medium":{"n":104,"low":0.6795,"band":"medium","high":0.8397,"agree":80,"accuracy":0.7692}},"provider":"rule_based_v1+lex99e1195c7d6f","lexicon_id":"99e1195c7d6f","contract_published_at":"2026-10-23T12:25:29.022232+00:00","confidence_is_band":true,"confidence_is_probability":false,"decision_taxonomy":"two_way","decidable_fraction_two_way":0.8854,"decidable_fraction_two_way_low":0.8565,"decidable_fraction_two_way_high":0.9143,"decidable_fraction_three_way":0.8586,"decidable_fraction_three_way_low":0.8287,"decidable_fraction_three_way_high":0.8884,"decidable_population":5529,"validation_sample_size":438,"three_way_cost_fraction":0.0268,"three_way_cost_records":148,"decision_counts":{"automatable":230,"not_automatable":5299},"decision_split":{"automatable":{"count":230,"share":0.04159884246699222},"not_automatable":{"count":5299,"share":0.9584011575330078}},"diagnostic_work_type_counts":{"mixed":24,"physical":5275,"data_only":230}}}