{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:L7YVM7WSIAPWNTXI24Y742XHBP","short_pith_number":"pith:L7YVM7WS","schema_version":"1.0","canonical_sha256":"5ff1567ed2401f66cee8d731fe6ae70bfe2b9a63c6490eb934856a849f23422c","source":{"kind":"arxiv","id":"2408.14575","version":4},"attestation_state":"computed","paper":{"title":"EVINCE: Optimizing Multi-LLM Dialogues Using Conditional Statistics and Information Theory","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Edward Y. Chang","submitted_at":"2024-08-26T18:48:51Z","abstract_excerpt":"EVINCE (Entropy and Variation IN Conditional Exchanges) is a novel framework for optimizing multi-LLM dialogues using conditional statistics and information theory. It addresses limitations in multi-agent debate (MAS) frameworks, where multiple LLMs ``chat'' without behavior modulation or mutual information quality assessment. Using dual entropy optimization to balance perspective diversity and prior knowledge, $\\EVINCE$ provides quantitative tools to dynamically regulate LLM linguistic behaviors. When mutual information is low and both cross-entropy and Wasserstein distance are high, EVINCE p"},"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":"2408.14575","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-08-26T18:48:51Z","cross_cats_sorted":[],"title_canon_sha256":"96b974ae85e40badae2e90a568c7ee43ff0c56bd8799930c269c0c389e5a64e0","abstract_canon_sha256":"f8388b58f1d6f2641405ab1651b591644a349ee8dc001651ff7bdad36ff022cb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:07:01.572015Z","signature_b64":"MC6JQVjXMC7eIKgNsz+56pVjJXQh0pFUAa/KQM59LZP8YKLJz+YO78Y8XglSbDYSNTC/NeUvJI55PIjSSE6QBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5ff1567ed2401f66cee8d731fe6ae70bfe2b9a63c6490eb934856a849f23422c","last_reissued_at":"2026-07-05T10:07:01.571562Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:07:01.571562Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EVINCE: Optimizing Multi-LLM Dialogues Using Conditional Statistics and Information Theory","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Edward Y. Chang","submitted_at":"2024-08-26T18:48:51Z","abstract_excerpt":"EVINCE (Entropy and Variation IN Conditional Exchanges) is a novel framework for optimizing multi-LLM dialogues using conditional statistics and information theory. It addresses limitations in multi-agent debate (MAS) frameworks, where multiple LLMs ``chat'' without behavior modulation or mutual information quality assessment. Using dual entropy optimization to balance perspective diversity and prior knowledge, $\\EVINCE$ provides quantitative tools to dynamically regulate LLM linguistic behaviors. When mutual information is low and both cross-entropy and Wasserstein distance are high, EVINCE p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.14575","kind":"arxiv","version":4},"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/2408.14575/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":"2408.14575","created_at":"2026-07-05T10:07:01.571621+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.14575v4","created_at":"2026-07-05T10:07:01.571621+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.14575","created_at":"2026-07-05T10:07:01.571621+00:00"},{"alias_kind":"pith_short_12","alias_value":"L7YVM7WSIAPW","created_at":"2026-07-05T10:07:01.571621+00:00"},{"alias_kind":"pith_short_16","alias_value":"L7YVM7WSIAPWNTXI","created_at":"2026-07-05T10:07:01.571621+00:00"},{"alias_kind":"pith_short_8","alias_value":"L7YVM7WS","created_at":"2026-07-05T10:07:01.571621+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.03722","citing_title":"When Outputs Disperse, Does Epistemic Revision Follow? A Black-Box Diagnostic for Machine Collectives","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L7YVM7WSIAPWNTXI24Y742XHBP","json":"https://pith.science/pith/L7YVM7WSIAPWNTXI24Y742XHBP.json","graph_json":"https://pith.science/api/pith-number/L7YVM7WSIAPWNTXI24Y742XHBP/graph.json","events_json":"https://pith.science/api/pith-number/L7YVM7WSIAPWNTXI24Y742XHBP/events.json","paper":"https://pith.science/paper/L7YVM7WS"},"agent_actions":{"view_html":"https://pith.science/pith/L7YVM7WSIAPWNTXI24Y742XHBP","download_json":"https://pith.science/pith/L7YVM7WSIAPWNTXI24Y742XHBP.json","view_paper":"https://pith.science/paper/L7YVM7WS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.14575&json=true","fetch_graph":"https://pith.science/api/pith-number/L7YVM7WSIAPWNTXI24Y742XHBP/graph.json","fetch_events":"https://pith.science/api/pith-number/L7YVM7WSIAPWNTXI24Y742XHBP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L7YVM7WSIAPWNTXI24Y742XHBP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L7YVM7WSIAPWNTXI24Y742XHBP/action/storage_attestation","attest_author":"https://pith.science/pith/L7YVM7WSIAPWNTXI24Y742XHBP/action/author_attestation","sign_citation":"https://pith.science/pith/L7YVM7WSIAPWNTXI24Y742XHBP/action/citation_signature","submit_replication":"https://pith.science/pith/L7YVM7WSIAPWNTXI24Y742XHBP/action/replication_record"}},"created_at":"2026-07-05T10:07:01.571621+00:00","updated_at":"2026-07-05T10:07:01.571621+00:00"}