{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ECJBADMORGZE6Q3F3K2LQCOSUD","short_pith_number":"pith:ECJBADMO","schema_version":"1.0","canonical_sha256":"2092100d8e89b24f4365dab4b809d2a0d4a49ea195e59018054f69e90f3d47c9","source":{"kind":"arxiv","id":"2508.11414","version":1},"attestation_state":"computed","paper":{"title":"Survey-to-Behavior: Downstream Alignment of Human Values in LLMs via Survey Questions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Charles Welch, David Kacz\\'er, Florian Mai, Lucie Flek, Shangrui Nie, Zhixue Zhao","submitted_at":"2025-08-15T11:36:17Z","abstract_excerpt":"Large language models implicitly encode preferences over human values, yet steering them often requires large training data. In this work, we investigate a simple approach: Can we reliably modify a model's value system in downstream behavior by training it to answer value survey questions accordingly? We first construct value profiles of several open-source LLMs by asking them to rate a series of value-related descriptions spanning 20 distinct human values, which we use as a baseline for subsequent experiments. We then investigate whether the value system of a model can be governed by fine-tun"},"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":"2508.11414","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-15T11:36:17Z","cross_cats_sorted":[],"title_canon_sha256":"822eaf998df5557434690ad43188676ab9d27125ad081528120fa97093288c4f","abstract_canon_sha256":"165bca0bb3b61d11f66d7224bb0ee60315eac6f19ab36dc045fa960bb69de9f3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:54:33.073331Z","signature_b64":"bny8+dhP7Dkcrm8s7MnUc3s3ouhLsdq82E9TZOPBd7haYVWPZ6JWlhqjOBv/ZQItMs7mES4FFW5IYdt30dv5BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2092100d8e89b24f4365dab4b809d2a0d4a49ea195e59018054f69e90f3d47c9","last_reissued_at":"2026-07-05T11:54:33.072922Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:54:33.072922Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Survey-to-Behavior: Downstream Alignment of Human Values in LLMs via Survey Questions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Charles Welch, David Kacz\\'er, Florian Mai, Lucie Flek, Shangrui Nie, Zhixue Zhao","submitted_at":"2025-08-15T11:36:17Z","abstract_excerpt":"Large language models implicitly encode preferences over human values, yet steering them often requires large training data. In this work, we investigate a simple approach: Can we reliably modify a model's value system in downstream behavior by training it to answer value survey questions accordingly? We first construct value profiles of several open-source LLMs by asking them to rate a series of value-related descriptions spanning 20 distinct human values, which we use as a baseline for subsequent experiments. We then investigate whether the value system of a model can be governed by fine-tun"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.11414","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/2508.11414/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":"2508.11414","created_at":"2026-07-05T11:54:33.072974+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.11414v1","created_at":"2026-07-05T11:54:33.072974+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.11414","created_at":"2026-07-05T11:54:33.072974+00:00"},{"alias_kind":"pith_short_12","alias_value":"ECJBADMORGZE","created_at":"2026-07-05T11:54:33.072974+00:00"},{"alias_kind":"pith_short_16","alias_value":"ECJBADMORGZE6Q3F","created_at":"2026-07-05T11:54:33.072974+00:00"},{"alias_kind":"pith_short_8","alias_value":"ECJBADMO","created_at":"2026-07-05T11:54:33.072974+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/ECJBADMORGZE6Q3F3K2LQCOSUD","json":"https://pith.science/pith/ECJBADMORGZE6Q3F3K2LQCOSUD.json","graph_json":"https://pith.science/api/pith-number/ECJBADMORGZE6Q3F3K2LQCOSUD/graph.json","events_json":"https://pith.science/api/pith-number/ECJBADMORGZE6Q3F3K2LQCOSUD/events.json","paper":"https://pith.science/paper/ECJBADMO"},"agent_actions":{"view_html":"https://pith.science/pith/ECJBADMORGZE6Q3F3K2LQCOSUD","download_json":"https://pith.science/pith/ECJBADMORGZE6Q3F3K2LQCOSUD.json","view_paper":"https://pith.science/paper/ECJBADMO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.11414&json=true","fetch_graph":"https://pith.science/api/pith-number/ECJBADMORGZE6Q3F3K2LQCOSUD/graph.json","fetch_events":"https://pith.science/api/pith-number/ECJBADMORGZE6Q3F3K2LQCOSUD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ECJBADMORGZE6Q3F3K2LQCOSUD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ECJBADMORGZE6Q3F3K2LQCOSUD/action/storage_attestation","attest_author":"https://pith.science/pith/ECJBADMORGZE6Q3F3K2LQCOSUD/action/author_attestation","sign_citation":"https://pith.science/pith/ECJBADMORGZE6Q3F3K2LQCOSUD/action/citation_signature","submit_replication":"https://pith.science/pith/ECJBADMORGZE6Q3F3K2LQCOSUD/action/replication_record"}},"created_at":"2026-07-05T11:54:33.072974+00:00","updated_at":"2026-07-05T11:54:33.072974+00:00"}