{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:K65TFAVEVNBNHRWTEY3FC2XXKF","short_pith_number":"pith:K65TFAVE","schema_version":"1.0","canonical_sha256":"57bb3282a4ab42d3c6d32636516af7515b9ce6859383fbfec2ee771b9de4b4c9","source":{"kind":"arxiv","id":"2605.29400","version":1},"attestation_state":"computed","paper":{"title":"Architecture-Sensitive Supervised Fine-Tuning for Screen-Conditioned Action Prediction: A PiSAR Benchmark","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.HC"],"primary_cat":"cs.AI","authors_text":"Abhishek Vyas, Rahul Bissa, Yash Jain","submitted_at":"2026-05-28T05:49:36Z","abstract_excerpt":"We benchmark three supervised fine-tuned models against frontier zero-shot baselines on a 661-row held-out slice of PiSAR (Persona, intent, Screen, Action, Rationale), a 12,929-tuple corpus of screen-anchored behavioural rationales curated from public app-store reviews, Pew American Trends Panel demographics, and the OPeRA shopper traces. Every model, frontier or fine-tuned, is evaluated on the same 661-row slice with the same scoring pipeline. Two findings. First, frontier zero-shot baselines (Claude Opus 4.7 and GPT-5.5) reach sem_sim 0.459 and 0.482 respectively; a fine-tuned Qwen3-VL-8B-In"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.29400","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-28T05:49:36Z","cross_cats_sorted":["cs.CL","cs.HC"],"title_canon_sha256":"9ea2e3607b727ed3cc970d26092015ef11e576989efb3ca04159453dfffd66ae","abstract_canon_sha256":"59a760223dc49cc8a0941e79482a6b3aaeb671cf2aa885b941e92db530249d60"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:37.098598Z","signature_b64":"4kivnC1Kne8VRbHzLPai9q8M49SXXvYNQ0TtBJuHuDcdCvL5oBPtcYxuz4z6YsYPkvqgoM2a5bHsetDHX03PDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"57bb3282a4ab42d3c6d32636516af7515b9ce6859383fbfec2ee771b9de4b4c9","last_reissued_at":"2026-05-29T01:05:37.097700Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:37.097700Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Architecture-Sensitive Supervised Fine-Tuning for Screen-Conditioned Action Prediction: A PiSAR Benchmark","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.HC"],"primary_cat":"cs.AI","authors_text":"Abhishek Vyas, Rahul Bissa, Yash Jain","submitted_at":"2026-05-28T05:49:36Z","abstract_excerpt":"We benchmark three supervised fine-tuned models against frontier zero-shot baselines on a 661-row held-out slice of PiSAR (Persona, intent, Screen, Action, Rationale), a 12,929-tuple corpus of screen-anchored behavioural rationales curated from public app-store reviews, Pew American Trends Panel demographics, and the OPeRA shopper traces. Every model, frontier or fine-tuned, is evaluated on the same 661-row slice with the same scoring pipeline. Two findings. First, frontier zero-shot baselines (Claude Opus 4.7 and GPT-5.5) reach sem_sim 0.459 and 0.482 respectively; a fine-tuned Qwen3-VL-8B-In"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29400","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.29400/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.29400","created_at":"2026-05-29T01:05:37.097868+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29400v1","created_at":"2026-05-29T01:05:37.097868+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29400","created_at":"2026-05-29T01:05:37.097868+00:00"},{"alias_kind":"pith_short_12","alias_value":"K65TFAVEVNBN","created_at":"2026-05-29T01:05:37.097868+00:00"},{"alias_kind":"pith_short_16","alias_value":"K65TFAVEVNBNHRWT","created_at":"2026-05-29T01:05:37.097868+00:00"},{"alias_kind":"pith_short_8","alias_value":"K65TFAVE","created_at":"2026-05-29T01:05:37.097868+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/K65TFAVEVNBNHRWTEY3FC2XXKF","json":"https://pith.science/pith/K65TFAVEVNBNHRWTEY3FC2XXKF.json","graph_json":"https://pith.science/api/pith-number/K65TFAVEVNBNHRWTEY3FC2XXKF/graph.json","events_json":"https://pith.science/api/pith-number/K65TFAVEVNBNHRWTEY3FC2XXKF/events.json","paper":"https://pith.science/paper/K65TFAVE"},"agent_actions":{"view_html":"https://pith.science/pith/K65TFAVEVNBNHRWTEY3FC2XXKF","download_json":"https://pith.science/pith/K65TFAVEVNBNHRWTEY3FC2XXKF.json","view_paper":"https://pith.science/paper/K65TFAVE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29400&json=true","fetch_graph":"https://pith.science/api/pith-number/K65TFAVEVNBNHRWTEY3FC2XXKF/graph.json","fetch_events":"https://pith.science/api/pith-number/K65TFAVEVNBNHRWTEY3FC2XXKF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/K65TFAVEVNBNHRWTEY3FC2XXKF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/K65TFAVEVNBNHRWTEY3FC2XXKF/action/storage_attestation","attest_author":"https://pith.science/pith/K65TFAVEVNBNHRWTEY3FC2XXKF/action/author_attestation","sign_citation":"https://pith.science/pith/K65TFAVEVNBNHRWTEY3FC2XXKF/action/citation_signature","submit_replication":"https://pith.science/pith/K65TFAVEVNBNHRWTEY3FC2XXKF/action/replication_record"}},"created_at":"2026-05-29T01:05:37.097868+00:00","updated_at":"2026-05-29T01:05:37.097868+00:00"}