{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:N5GYWTFUQPBE7UY42KJXRLIUGR","short_pith_number":"pith:N5GYWTFU","canonical_record":{"source":{"id":"2605.14205","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T00:01:11Z","cross_cats_sorted":[],"title_canon_sha256":"3d0850380e19e90bf30715c20989772fef818895df33e1f9a1c7527cd3d6f2d0","abstract_canon_sha256":"6c48b9c860951119510e3289ae4339f983bda13e19d0d5cf16be4dad608717bc"},"schema_version":"1.0"},"canonical_sha256":"6f4d8b4cb483c24fd31cd29378ad14346ad1ebea3b3304867e188455b4c058fb","source":{"kind":"arxiv","id":"2605.14205","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14205","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14205v1","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14205","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"pith_short_12","alias_value":"N5GYWTFUQPBE","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"N5GYWTFUQPBE7UY4","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"N5GYWTFU","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:N5GYWTFUQPBE7UY42KJXRLIUGR","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14205","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T00:01:11Z","cross_cats_sorted":[],"title_canon_sha256":"3d0850380e19e90bf30715c20989772fef818895df33e1f9a1c7527cd3d6f2d0","abstract_canon_sha256":"6c48b9c860951119510e3289ae4339f983bda13e19d0d5cf16be4dad608717bc"},"schema_version":"1.0"},"canonical_sha256":"6f4d8b4cb483c24fd31cd29378ad14346ad1ebea3b3304867e188455b4c058fb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:11.002885Z","signature_b64":"ZLfMZ6QD5EnetMqklUtz1zANCZV8rMzVj6g3EEECvAhExtC2PYHONVCQAHvQXZgH6x1uZyujgHqG92YmPRPRBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6f4d8b4cb483c24fd31cd29378ad14346ad1ebea3b3304867e188455b4c058fb","last_reissued_at":"2026-05-17T23:39:11.002190Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:11.002190Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14205","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8YEEUlBuWbiLBmJB3rhjN1v+bVj1diQ/l/R65Ccl26XAwH1OwHafSNmm68RrIeHvu+58mRkaYjpC4W2hfZUZCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-02T18:44:55.401604Z"},"content_sha256":"8bca8612131eaca63c649afc1d77ccc1f801dec251269b8e53cc71155a238109","schema_version":"1.0","event_id":"sha256:8bca8612131eaca63c649afc1d77ccc1f801dec251269b8e53cc71155a238109"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:N5GYWTFUQPBE7UY42KJXRLIUGR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SimPersona: Learning Discrete Buyer Personas from Raw Clickstreams for Grounded E-Commerce Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"SimPersona learns discrete buyer types from clickstreams to let LLM agents simulate diverse real buyer populations in e-commerce.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Alberto Castelo, Han Li, Lingyun Wang, Shuang Xie, Ted Chaiwachirasak, Zahra Zanjani Foumani","submitted_at":"2026-05-14T00:01:11Z","abstract_excerpt":"LLM-based web agents can navigate live storefronts, yet they often collapse to a single \"average buyer\" policy, failing to capture the heterogeneous and distributional nature of real buyer populations. Existing personalization methods rely on hand-crafted prompt-based personas that are brittle, difficult to scale, context-inefficient, and unable to faithfully represent population-level behavior. We introduce SimPersona, a novel framework that learns discrete buyer types from historical traffic and exposes them to LLM-based web agents as compact persona tokens. Given raw clickstreams, a behavio"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Evaluated on 8.37M buyers across 42 held-out live storefronts, SimPersona achieves 78% conversion-rate alignment with real buyers, exhibits interpretable behavioral variation across buyer types, and outperforms a baseline with 8× more parameters on goal-oriented shopping tasks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The discrete buyer types learned from historical clickstreams will transfer effectively to LLM agent behavior in new live interactions without major distribution shift or loss of fidelity.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SimPersona uses VQ-VAE to induce discrete buyer types from clickstreams, maps them to LLM persona tokens, and fine-tunes agents to achieve 78% conversion-rate alignment with real buyers across 42 storefronts.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"SimPersona learns discrete buyer types from clickstreams to let LLM agents simulate diverse real buyer populations in e-commerce.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"6f93ca10aae8a5fb546895a22ac5058ad7b46ddda69fc8f530bf60f25ff45f24"},"source":{"id":"2605.14205","kind":"arxiv","version":1},"verdict":{"id":"5e6cb387-d2e2-478c-888c-176bc54fb474","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T02:52:59.932218Z","strongest_claim":"Evaluated on 8.37M buyers across 42 held-out live storefronts, SimPersona achieves 78% conversion-rate alignment with real buyers, exhibits interpretable behavioral variation across buyer types, and outperforms a baseline with 8× more parameters on goal-oriented shopping tasks.","one_line_summary":"SimPersona uses VQ-VAE to induce discrete buyer types from clickstreams, maps them to LLM persona tokens, and fine-tunes agents to achieve 78% conversion-rate alignment with real buyers across 42 storefronts.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The discrete buyer types learned from historical clickstreams will transfer effectively to LLM agent behavior in new live interactions without major distribution shift or loss of fidelity.","pith_extraction_headline":"SimPersona learns discrete buyer types from clickstreams to let LLM agents simulate diverse real buyer populations in e-commerce."},"references":{"count":35,"sample":[{"doi":"","year":2007,"title":"k-means++: The advantages of careful seeding","work_id":"ffbfe61b-cb77-4b64-a921-fa4713e96fee","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1974,"title":"A dendrite method for cluster analysis.Communications in Statistics – Theory and Methods, 3(1):1–27, 1974","work_id":"bfbe7651-0e14-4401-9191-be17bc96c872","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Beyond demographics: Aligning role-playing llm-based agents using human belief networks","work_id":"4dade001-b288-440f-b9cd-3a9c10e96919","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1988,"title":"Lawrence Erlbaum Associates, 2 edition","work_id":"d4f908c1-d99f-4777-b721-828c5e0d4b1b","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Mind2web: Towards a generalist agent for the web","work_id":"45afb2f7-1446-4115-a5fb-463dbdd1a2e0","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":35,"snapshot_sha256":"d81f7cc9d98c22e7140e88f6ddd82c964a319e3991ec09f9b1d6a8e7f1e17087","internal_anchors":6},"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"},"verdict_id":"5e6cb387-d2e2-478c-888c-176bc54fb474"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"s3lqLH2Jh4qEWhuyKi40FVpCzNqoMLjWYMUpq4jphIF5PPWcfJzlzd+f0cg05QXjZ5WaGZcmxwVXaoebF6rtCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-02T18:44:55.402237Z"},"content_sha256":"ea008c4d753eadced32169191ecd6dcc89d0b1e649f95e1fbef53d3a73514dfd","schema_version":"1.0","event_id":"sha256:ea008c4d753eadced32169191ecd6dcc89d0b1e649f95e1fbef53d3a73514dfd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/N5GYWTFUQPBE7UY42KJXRLIUGR/bundle.json","state_url":"https://pith.science/pith/N5GYWTFUQPBE7UY42KJXRLIUGR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/N5GYWTFUQPBE7UY42KJXRLIUGR/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-02T18:44:55Z","links":{"resolver":"https://pith.science/pith/N5GYWTFUQPBE7UY42KJXRLIUGR","bundle":"https://pith.science/pith/N5GYWTFUQPBE7UY42KJXRLIUGR/bundle.json","state":"https://pith.science/pith/N5GYWTFUQPBE7UY42KJXRLIUGR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/N5GYWTFUQPBE7UY42KJXRLIUGR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:N5GYWTFUQPBE7UY42KJXRLIUGR","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"6c48b9c860951119510e3289ae4339f983bda13e19d0d5cf16be4dad608717bc","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T00:01:11Z","title_canon_sha256":"3d0850380e19e90bf30715c20989772fef818895df33e1f9a1c7527cd3d6f2d0"},"schema_version":"1.0","source":{"id":"2605.14205","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14205","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14205v1","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14205","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"pith_short_12","alias_value":"N5GYWTFUQPBE","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"N5GYWTFUQPBE7UY4","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"N5GYWTFU","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:ea008c4d753eadced32169191ecd6dcc89d0b1e649f95e1fbef53d3a73514dfd","target":"graph","created_at":"2026-05-17T23:39:11Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Evaluated on 8.37M buyers across 42 held-out live storefronts, SimPersona achieves 78% conversion-rate alignment with real buyers, exhibits interpretable behavioral variation across buyer types, and outperforms a baseline with 8× more parameters on goal-oriented shopping tasks."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The discrete buyer types learned from historical clickstreams will transfer effectively to LLM agent behavior in new live interactions without major distribution shift or loss of fidelity."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"SimPersona uses VQ-VAE to induce discrete buyer types from clickstreams, maps them to LLM persona tokens, and fine-tunes agents to achieve 78% conversion-rate alignment with real buyers across 42 storefronts."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"SimPersona learns discrete buyer types from clickstreams to let LLM agents simulate diverse real buyer populations in e-commerce."}],"snapshot_sha256":"6f93ca10aae8a5fb546895a22ac5058ad7b46ddda69fc8f530bf60f25ff45f24"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"LLM-based web agents can navigate live storefronts, yet they often collapse to a single \"average buyer\" policy, failing to capture the heterogeneous and distributional nature of real buyer populations. Existing personalization methods rely on hand-crafted prompt-based personas that are brittle, difficult to scale, context-inefficient, and unable to faithfully represent population-level behavior. We introduce SimPersona, a novel framework that learns discrete buyer types from historical traffic and exposes them to LLM-based web agents as compact persona tokens. Given raw clickstreams, a behavio","authors_text":"Alberto Castelo, Han Li, Lingyun Wang, Shuang Xie, Ted Chaiwachirasak, Zahra Zanjani Foumani","cross_cats":[],"headline":"SimPersona learns discrete buyer types from clickstreams to let LLM agents simulate diverse real buyer populations in e-commerce.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T00:01:11Z","title":"SimPersona: Learning Discrete Buyer Personas from Raw Clickstreams for Grounded E-Commerce Agents"},"references":{"count":35,"internal_anchors":6,"resolved_work":35,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"k-means++: The advantages of careful seeding","work_id":"ffbfe61b-cb77-4b64-a921-fa4713e96fee","year":2007},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"A dendrite method for cluster analysis.Communications in Statistics – Theory and Methods, 3(1):1–27, 1974","work_id":"bfbe7651-0e14-4401-9191-be17bc96c872","year":1974},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Beyond demographics: Aligning role-playing llm-based agents using human belief networks","work_id":"4dade001-b288-440f-b9cd-3a9c10e96919","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Lawrence Erlbaum Associates, 2 edition","work_id":"d4f908c1-d99f-4777-b721-828c5e0d4b1b","year":1988},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Mind2web: Towards a generalist agent for the web","work_id":"45afb2f7-1446-4115-a5fb-463dbdd1a2e0","year":2023}],"snapshot_sha256":"d81f7cc9d98c22e7140e88f6ddd82c964a319e3991ec09f9b1d6a8e7f1e17087"},"source":{"id":"2605.14205","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T02:52:59.932218Z","id":"5e6cb387-d2e2-478c-888c-176bc54fb474","model_set":{"reader":"grok-4.3"},"one_line_summary":"SimPersona uses VQ-VAE to induce discrete buyer types from clickstreams, maps them to LLM persona tokens, and fine-tunes agents to achieve 78% conversion-rate alignment with real buyers across 42 storefronts.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"SimPersona learns discrete buyer types from clickstreams to let LLM agents simulate diverse real buyer populations in e-commerce.","strongest_claim":"Evaluated on 8.37M buyers across 42 held-out live storefronts, SimPersona achieves 78% conversion-rate alignment with real buyers, exhibits interpretable behavioral variation across buyer types, and outperforms a baseline with 8× more parameters on goal-oriented shopping tasks.","weakest_assumption":"The discrete buyer types learned from historical clickstreams will transfer effectively to LLM agent behavior in new live interactions without major distribution shift or loss of fidelity."}},"verdict_id":"5e6cb387-d2e2-478c-888c-176bc54fb474"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8bca8612131eaca63c649afc1d77ccc1f801dec251269b8e53cc71155a238109","target":"record","created_at":"2026-05-17T23:39:11Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"6c48b9c860951119510e3289ae4339f983bda13e19d0d5cf16be4dad608717bc","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T00:01:11Z","title_canon_sha256":"3d0850380e19e90bf30715c20989772fef818895df33e1f9a1c7527cd3d6f2d0"},"schema_version":"1.0","source":{"id":"2605.14205","kind":"arxiv","version":1}},"canonical_sha256":"6f4d8b4cb483c24fd31cd29378ad14346ad1ebea3b3304867e188455b4c058fb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6f4d8b4cb483c24fd31cd29378ad14346ad1ebea3b3304867e188455b4c058fb","first_computed_at":"2026-05-17T23:39:11.002190Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:11.002190Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZLfMZ6QD5EnetMqklUtz1zANCZV8rMzVj6g3EEECvAhExtC2PYHONVCQAHvQXZgH6x1uZyujgHqG92YmPRPRBg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:11.002885Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14205","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8bca8612131eaca63c649afc1d77ccc1f801dec251269b8e53cc71155a238109","sha256:ea008c4d753eadced32169191ecd6dcc89d0b1e649f95e1fbef53d3a73514dfd"],"state_sha256":"df2fe0ee8945a047b26eb17c572364b53dbf222b347e514e85a3bd2d02ebe956"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uGdtr0Jq/l1NFLsvfFqNiGMqKgSKoo9qaQ/IM8Y5BGYPKFNDfcc+k9L1bpWePyb+WuKvjRqsxFAu2nKZJMYzAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-02T18:44:55.404882Z","bundle_sha256":"01542f148ac5808841b2efc9f15a7d0fa23649a2eabeae39b95b28bf1f1028ca"}}