{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LH3UPC5MQAOWWMYIIJ4X5ZRY52","short_pith_number":"pith:LH3UPC5M","schema_version":"1.0","canonical_sha256":"59f7478bac801d6b330842797ee638eea86c8722a8c34435631f3b14c89be7db","source":{"kind":"arxiv","id":"2605.20082","version":1},"attestation_state":"computed","paper":{"title":"VL-DPO: Vision-Language-Guided Finetuning for Preference-Aligned Autonomous Driving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Ghassen Jerfel, Jeonhyung Kang, Khaled S. Refaat, Marina Haliem, Qi Zhao, Zhefan Xu","submitted_at":"2026-05-19T16:36:34Z","abstract_excerpt":"The rapid growth of autonomous driving datasets has enabled the scaling of powerful motion forecasting models. While large-scale pretraining provides strong performance, the standard imitation objective may not fully capture the complex nuances of human driving preferences. Meanwhile, recent advances in vision-language models (VLMs) have demonstrated impressive reasoning and commonsense understanding. Building on these capabilities, this paper presents VL-DPO, a vision-language-guided framework that aligns ego-vehicle motion forecasting models with human preferences. Our approach leverages a V"},"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":"2605.20082","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-19T16:36:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9a2cea8cf5ac5aab0772d3e168f0c6cd689394ea060ce041a4070424e7d97bd5","abstract_canon_sha256":"562fcf8d5db313547d08265b7cfab101ff2d83bab65961f1813c7b20e663b40b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T02:06:01.427582Z","signature_b64":"XLU7+mZ9bRQ2iRN8aYsz7gHtPxxsRgoqBoeS8vsrcob2BcGBwu9LoTn8mXqj5A+ovj/b/XtHkpNTmpg4dQ+qCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59f7478bac801d6b330842797ee638eea86c8722a8c34435631f3b14c89be7db","last_reissued_at":"2026-05-20T02:06:01.426870Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T02:06:01.426870Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VL-DPO: Vision-Language-Guided Finetuning for Preference-Aligned Autonomous Driving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Ghassen Jerfel, Jeonhyung Kang, Khaled S. Refaat, Marina Haliem, Qi Zhao, Zhefan Xu","submitted_at":"2026-05-19T16:36:34Z","abstract_excerpt":"The rapid growth of autonomous driving datasets has enabled the scaling of powerful motion forecasting models. While large-scale pretraining provides strong performance, the standard imitation objective may not fully capture the complex nuances of human driving preferences. Meanwhile, recent advances in vision-language models (VLMs) have demonstrated impressive reasoning and commonsense understanding. Building on these capabilities, this paper presents VL-DPO, a vision-language-guided framework that aligns ego-vehicle motion forecasting models with human preferences. Our approach leverages a V"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.20082","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/2605.20082/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":"2605.20082","created_at":"2026-05-20T02:06:01.426971+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.20082v1","created_at":"2026-05-20T02:06:01.426971+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.20082","created_at":"2026-05-20T02:06:01.426971+00:00"},{"alias_kind":"pith_short_12","alias_value":"LH3UPC5MQAOW","created_at":"2026-05-20T02:06:01.426971+00:00"},{"alias_kind":"pith_short_16","alias_value":"LH3UPC5MQAOWWMYI","created_at":"2026-05-20T02:06:01.426971+00:00"},{"alias_kind":"pith_short_8","alias_value":"LH3UPC5M","created_at":"2026-05-20T02:06:01.426971+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/LH3UPC5MQAOWWMYIIJ4X5ZRY52","json":"https://pith.science/pith/LH3UPC5MQAOWWMYIIJ4X5ZRY52.json","graph_json":"https://pith.science/api/pith-number/LH3UPC5MQAOWWMYIIJ4X5ZRY52/graph.json","events_json":"https://pith.science/api/pith-number/LH3UPC5MQAOWWMYIIJ4X5ZRY52/events.json","paper":"https://pith.science/paper/LH3UPC5M"},"agent_actions":{"view_html":"https://pith.science/pith/LH3UPC5MQAOWWMYIIJ4X5ZRY52","download_json":"https://pith.science/pith/LH3UPC5MQAOWWMYIIJ4X5ZRY52.json","view_paper":"https://pith.science/paper/LH3UPC5M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.20082&json=true","fetch_graph":"https://pith.science/api/pith-number/LH3UPC5MQAOWWMYIIJ4X5ZRY52/graph.json","fetch_events":"https://pith.science/api/pith-number/LH3UPC5MQAOWWMYIIJ4X5ZRY52/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LH3UPC5MQAOWWMYIIJ4X5ZRY52/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LH3UPC5MQAOWWMYIIJ4X5ZRY52/action/storage_attestation","attest_author":"https://pith.science/pith/LH3UPC5MQAOWWMYIIJ4X5ZRY52/action/author_attestation","sign_citation":"https://pith.science/pith/LH3UPC5MQAOWWMYIIJ4X5ZRY52/action/citation_signature","submit_replication":"https://pith.science/pith/LH3UPC5MQAOWWMYIIJ4X5ZRY52/action/replication_record"}},"created_at":"2026-05-20T02:06:01.426971+00:00","updated_at":"2026-05-20T02:06:01.426971+00:00"}