{"paper":{"id":"2609.00014.pdf","title":"Behaviorally Grounded User Profiles from the Wild for Personalized Alignment and Multi-Perspective Reasoning","authors":["Yuxuan Li","Victor Zhong","Ehsan Kamalloo"],"arxiv_id":"","url":"","source_kind":"pdf","ir_version":"1.0"},"generated_at":"2026-09-14T04:01:24.305601+00:00","counts":{"error":0,"warning":0,"info":0},"findings":[],"top_findings":[],"ir_summary":{"claims":20,"contributions":3,"method_components":10,"experiments":13,"baselines":6,"datasets":14,"metrics":8,"tables":8,"figures":4,"citations":43,"equations":0,"quantitative_results":468,"has_repository":false},"ir_health":{"observations":[{"code":"ablation-labels-match-nothing","severity":"warning","message":"None of the experiment ablation labels match any method component.","detail":"Labels found: 'Open-Ended -> Random', 'w/o profiles (high-temperature sampling)', 'Open-Ended -> Raw Posts'. Either the components were extracted with different wording, or these labels are baselines rather than ablations — in which case RL004 has no evidence that anything was ablated."},{"code":"compile-warning","severity":"note","message":"references: the reference list carries no entry numbers; citation keys are enumeration order [1]..[N], not labels printed in the paper, so a key may not match the [N] used in the body","detail":""}],"errors":0,"warnings":1,"trustworthy":true},"diagnostics":{"source":"/home/oceann/workspace/research-linter/eval/papers/2609.00014.pdf","rules_run":["RL001","RL002","RL003","RL004"],"rules_skipped":[{"rule":"RL005","reason":"not-applicable"}],"rule_errors":{},"rule_findings":{"RL001":0,"RL002":0,"RL003":0,"RL004":0},"rule_seconds":{"RL001":0.01,"RL002":0.9,"RL003":7.85,"RL004":0.05},"dropped_no_evidence":0,"demoted_low_confidence":0,"ir_health":{"observations":[{"code":"ablation-labels-match-nothing","severity":"warning","message":"None of the experiment ablation labels match any method component.","detail":"Labels found: 'Open-Ended -> Random', 'w/o profiles (high-temperature sampling)', 'Open-Ended -> Raw Posts'. Either the components were extracted with different wording, or these labels are baselines rather than ablations — in which case RL004 has no evidence that anything was ablated."},{"code":"compile-warning","severity":"note","message":"references: the reference list carries no entry numbers; citation keys are enumeration order [1]..[N], not labels printed in the paper, so a key may not match the [N] used in the body","detail":""}],"errors":0,"warnings":1,"trustworthy":true},"notes":["RL001 funnel arithmetic: sentences_with_from=7 endpoint_pairs=3 claims_compared=0 emitted=0 [dropped: not-a-measurement=0]","RL001 funnel delta-column: structured_tables=8 with_exactly_one_delta_column=0 rows_checked=0 emitted=0","RL001 funnel cross-location: quant=468 prose=137 prose_with_metric=27 prose_with_dataset=70 prose_with_metric_and_dataset=23 distinct_buckets=3 emitted=0 [dropped: not-a-measurement=1 not-in-quote=9]","RL001 funnel identifier-drift: settings_checked=2 sections_stating_one_value=1 emitted=0","RL002 citation entailment: attempted 14 of 14 ranked cited claims (37 in-text markers from 43 bibliography entries); 1 cited works could not be identified, 8 had no retrievable text, 0 model quotes failed verbatim verification, 0 were topically unrelated to the retrieved text.","RL002 candidate ranking (why the top 14 were chosen): [33] [introduction, names an artefact (OPPU), title/sentence overlap near zero]; [37] [introduction, title/sentence overlap near zero, no DOI/arXiv id]; [32] [introduction, title/sentence overlap near zero, no DOI/arXiv id]; [31] [introduction, title/sentence overlap near zero, no DOI/arXiv id]; [24] [introduction, title/sentence overlap near zero, no DOI/arXiv id]; [40] [introduction, title/sentence overlap near zero, no DOI/arXiv id]; [30] [introduction, title/sentence overlap near zero]; [10] [introduction, title/sentence overlap near zero]","RL002 label distribution over 5 classified citations: SUPPORTED=5","RL002 base-rate self-check: not tripped (0/5 = 0% NOT_SUPPORTED, threshold 40%).","RL002 attribution check (case B): 3 candidates name an artefact, 3 of them made the top-N and were checked against their cited source","RL002: 4 citations skipped because the sentence reports this paper's own results (the citation there is a comparison pointer, not evidence for the sentence)","RL002: 2 markers skipped because the claim is jointly attributed to several works","RL002: 4 enumerated-citation sentences skipped (they cite >=3 works, so the claim is about the set and cannot be refuted by any single cited work)","RL002: 1 cited works unresolvable (first 3: [{\"key\": \"[27]\", \"title\": \"Mauve: Measuring the gap between neural text and human text using divergence frontiers\", \"why\": \"no work record matched the bibliography title\"}])","RL003: 3 queries -> 116 unique works (arXiv search 50, category listings 0, scholarly graph 66 [crossref 66], 0 graph works matched to an arXiv id); source errors during retrieval: arXiv 0, graph 0","RL003: parsed the reference list of 7/12 comparable papers (5 unreadable) — that is the PeerUsage denominator","RL003: enriched 0 of 0 candidates with their own abstract (0 unavailable, 0 over budget); 0 dropped as published after this paper","RL003: scholarly graph for this run = crossref","RL003 funnel (342 candidates): cited_only_from_related_work -74, peer_usage_below_quorum -76, survey_or_artifact -39, title_names_no_method -153 || kept: citation_context_unknown +38 || survived-cheap-gates 0 -> with-abstract 0 -> comparable 0 -> above-threshold-0.65 0","RL003: 0 comparable candidates all scored below 0.65 — nothing to report for this paper","RL004 funnel: components=10 contribution_backed=3 candidates=1 ablations_found=7 matched_by_ablation=1 unmatched=0 experiments=13","RL004 component M1 'Behaviorally Grounded Profile Extraction': covered (filtered earlier)","RL004 component M2 'LLM Attributes Extraction': covered (filtered earlier)","RL004 component M3 'Removing duplicates, conflicts & Summarise': covered (filtered earlier)","RL004 component M4 'Personalized Alignment via SFT': covered (candidate)","RL004 component M5 'Test-time Multi-Perspective Reasoning': covered (filtered earlier)","RL004 component M6 'Mixture of Perspectives': covered (filtered earlier)","RL004 component M7 'Aggregator': covered (filtered earlier)","RL004 component M8 'Full-parameter SFT': excluded:not-a-claimed-contribution","RL004 component M9 'QLoRA': excluded:not-a-claimed-contribution","RL004 component M10 'GPT-OSS-120B': excluded:not-a-claimed-contribution","RL004 funnel: ranked=0 emitted=0 suppressed_by_cap=0 dropped_unverified_quote=0"],"missing_core_rules":[],"seconds":30.46,"llm":{"calls":0,"cache_hits":7,"errors":0,"prompt_chars":0},"sources":{"arxiv":{"requests":3,"errors":0,"cache_hits":44,"available":1},"openalex":{"requests":1,"errors":0,"cache_hits":0,"available":0,"unavailable_reason":"OpenAlex free daily budget exhausted (resets midnight UTC)"},"crossref":{"requests":0,"errors":0,"cache_hits":17,"available":1},"s2":{"requests":0,"errors":0,"cache_hits":0,"available":1},"short_circuited":[]},"cache":{"hits":68,"misses":4}},"compile_warnings":["references: the reference list carries no entry numbers; citation keys are enumeration order [1]..[N], not labels printed in the paper, so a key may not match the [N] used in the body"]}