{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:P5I3RYXPJ7EZMWSYUQRJPJK3PD","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":"c9dbb85e25b7b0ef1ab2de1d92b80270a1d209915da9cabe3eb99441278a959a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-05T17:21:16Z","title_canon_sha256":"2360c5dd49177ea61a9a32ab73a99250482a8a32761f6e4c66437befd8f0e86e"},"schema_version":"1.0","source":{"id":"2506.05260","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.05260","created_at":"2026-07-05T11:16:44Z"},{"alias_kind":"arxiv_version","alias_value":"2506.05260v1","created_at":"2026-07-05T11:16:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.05260","created_at":"2026-07-05T11:16:44Z"},{"alias_kind":"pith_short_12","alias_value":"P5I3RYXPJ7EZ","created_at":"2026-07-05T11:16:44Z"},{"alias_kind":"pith_short_16","alias_value":"P5I3RYXPJ7EZMWSY","created_at":"2026-07-05T11:16:44Z"},{"alias_kind":"pith_short_8","alias_value":"P5I3RYXP","created_at":"2026-07-05T11:16:44Z"}],"graph_snapshots":[{"event_id":"sha256:80523f4885eb8aee553027bf9312ac6d7a31d2f7d21f941d679455ace83adf9c","target":"graph","created_at":"2026-07-05T11:16:44Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.05260/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Most Video Large Language Models (Video-LLMs) adopt preference alignment techniques, e.g., DPO~\\citep{rafailov2024dpo}, to optimize the reward margin between a winning response ($y_w$) and a losing response ($y_l$). However, the likelihood displacement observed in DPO indicates that both $\\log \\pi_\\theta (y_w\\mid x)$ and $\\log \\pi_\\theta (y_l\\mid x) $ often decrease during training, inadvertently boosting the probabilities of non-target responses. In this paper, we systematically revisit this phenomenon from LLMs to Video-LLMs, showing that it intensifies when dealing with the redundant comple","authors_text":"Jinfa Huang, Li Yuan, Peixi Peng, Xiaodong Wang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-05T17:21:16Z","title":"LeanPO: Lean Preference Optimization for Likelihood Alignment in Video-LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.05260","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:624d50b3a17e918315c64f4e4ff963e1e209dd033536d54b115eb80967cd3b84","target":"record","created_at":"2026-07-05T11:16:44Z","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":"c9dbb85e25b7b0ef1ab2de1d92b80270a1d209915da9cabe3eb99441278a959a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-05T17:21:16Z","title_canon_sha256":"2360c5dd49177ea61a9a32ab73a99250482a8a32761f6e4c66437befd8f0e86e"},"schema_version":"1.0","source":{"id":"2506.05260","kind":"arxiv","version":1}},"canonical_sha256":"7f51b8e2ef4fc9965a58a42297a55b78eea66056f5e390b90e56a9aedbf44467","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7f51b8e2ef4fc9965a58a42297a55b78eea66056f5e390b90e56a9aedbf44467","first_computed_at":"2026-07-05T11:16:44.921532Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:16:44.921532Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YKvs/MA4d04hoP6R52zZW+auCnc99S54x9f82GIPq/vI1zA9ttiPOON8o4xcI4kWO13aBn8aLCkbSrdhGgy6BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:16:44.922018Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.05260","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:624d50b3a17e918315c64f4e4ff963e1e209dd033536d54b115eb80967cd3b84","sha256:80523f4885eb8aee553027bf9312ac6d7a31d2f7d21f941d679455ace83adf9c"],"state_sha256":"d96e1c3794edc3f8c872c372ec61faa24af3edfd5b6fa8849fe6d0930c7d7671"}