{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:G7KKMBNGRFQM7X5VLH3IRVQLAY","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":"96d6a7a8343c519df05069304470aab8a6f9c4b32489e818f8e558a24f651279","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-06-11T06:18:26Z","title_canon_sha256":"bfc3f6c68a9e2b362c4e80b0e33a14901356db046042ed1ac0bb4b097b81116e"},"schema_version":"1.0","source":{"id":"2406.06978","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.06978","created_at":"2026-05-18T03:50:37Z"},{"alias_kind":"arxiv_version","alias_value":"2406.06978v4","created_at":"2026-05-18T03:50:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.06978","created_at":"2026-05-18T03:50:37Z"},{"alias_kind":"pith_short_12","alias_value":"G7KKMBNGRFQM","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"G7KKMBNGRFQM7X5V","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"G7KKMBNG","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:a74ca52d058dc5d403a4e9481041bf51169b25ddfa8c5c93727dd7fc8413e413","target":"graph","created_at":"2026-05-18T03:50:37Z","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":"This method achieves the 1st place in the Navsim challenge, demonstrating significant improvements in generalization across diverse driving environments and conditions."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The multi-head decoder can simultaneously absorb conflicting signals from human and rule-based teachers without mode collapse or degraded performance on any single metric."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Hydra-MDP uses multi-teacher distillation and a multi-head decoder to learn diverse, metric-specific trajectories in an end-to-end autonomous-driving planner, winning the Navsim challenge."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Hydra-MDP trains an end-to-end planner by distilling knowledge from both human demonstrations and rule-based experts into a multi-head decoder that outputs diverse trajectories."}],"snapshot_sha256":"de0c17d258b240ad85c5a3c16695a2db5e4cfb1acc8ecbb1ce5a4525f905cc2b"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"c62df76cd2204039c5969cebd20513310a5be001de94ef67328d611a0164f5b7"},"paper":{"abstract_excerpt":"We propose Hydra-MDP, a novel paradigm employing multiple teachers in a teacher-student model. This approach uses knowledge distillation from both human and rule-based teachers to train the student model, which features a multi-head decoder to learn diverse trajectory candidates tailored to various evaluation metrics. With the knowledge of rule-based teachers, Hydra-MDP learns how the environment influences the planning in an end-to-end manner instead of resorting to non-differentiable post-processing. This method achieves the $1^{st}$ place in the Navsim challenge, demonstrating significant i","authors_text":"Jan Kautz, Jose M. Alvarez, Kailin Li, Shihao Wang, Shiyi Lan, Yishen Ji, Yu-Gang Jiang, Zhenxin Li, Zhiding Yu, Zhiqi Li, Ziyue Zhu, Zuxuan Wu","cross_cats":[],"headline":"Hydra-MDP trains an end-to-end planner by distilling knowledge from both human demonstrations and rule-based experts into a multi-head decoder that outputs diverse trajectories.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-06-11T06:18:26Z","title":"Hydra-MDP: End-to-end Multimodal Planning with Multi-target Hydra-Distillation"},"references":{"count":19,"internal_anchors":2,"resolved_work":19,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Quad: Query-based in- terpretable neural motion planning for autonomous driving","work_id":"1a045cde-fbd3-4c5d-a59a-342f16668d6a","year":2024},{"cited_arxiv_id":"2106.11810","doi":"","is_internal_anchor":true,"ref_index":3,"title":"NuPlan: A closed-loop ML-based planning benchmark for autonomous vehicles","work_id":"b16aace6-deff-4546-b333-bcb7c9c07cdb","year":2021},{"cited_arxiv_id":"2402.13243","doi":"","is_internal_anchor":true,"ref_index":4,"title":"VADv2: End-to-End Vectorized Autonomous Driving via Probabilistic Planning","work_id":"e7670f83-e1e1-41e7-86eb-39477a3a10b2","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Transfuser: Imita- tion with transformer-based sensor fusion for autonomous driving","work_id":"8b9d9ecc-fbc9-4be4-88eb-e6170c75b588","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":6,"title":"Navsim: Data-driven non-reactive autonomous vehicle simulation","work_id":"ab2ad404-1d69-4f8c-bf71-ceae9114f626","year":2024}],"snapshot_sha256":"8c42e9a988fece09949b512fd17a08917c7b60c5907f38794a4ae7d315b9a618"},"source":{"id":"2406.06978","kind":"arxiv","version":4},"verdict":{"created_at":"2026-05-13T23:08:25.962349Z","id":"8a072b7e-d671-4586-894f-e8acd65f7dfd","model_set":{"reader":"grok-4.3"},"one_line_summary":"Hydra-MDP uses multi-teacher distillation and a multi-head decoder to learn diverse, metric-specific trajectories in an end-to-end autonomous-driving planner, winning the Navsim challenge.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Hydra-MDP trains an end-to-end planner by distilling knowledge from both human demonstrations and rule-based experts into a multi-head decoder that outputs diverse trajectories.","strongest_claim":"This method achieves the 1st place in the Navsim challenge, demonstrating significant improvements in generalization across diverse driving environments and conditions.","weakest_assumption":"The multi-head decoder can simultaneously absorb conflicting signals from human and rule-based teachers without mode collapse or degraded performance on any single metric."}},"verdict_id":"8a072b7e-d671-4586-894f-e8acd65f7dfd"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:54963ba7a22570ba81edd6c9a90692cdb6c477e52ddbe708117d63a9b5612b14","target":"record","created_at":"2026-05-18T03:50:37Z","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":"96d6a7a8343c519df05069304470aab8a6f9c4b32489e818f8e558a24f651279","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-06-11T06:18:26Z","title_canon_sha256":"bfc3f6c68a9e2b362c4e80b0e33a14901356db046042ed1ac0bb4b097b81116e"},"schema_version":"1.0","source":{"id":"2406.06978","kind":"arxiv","version":4}},"canonical_sha256":"37d4a605a68960cfdfb559f688d60b062c21aa89e90533927c8571bc25b2f50c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"37d4a605a68960cfdfb559f688d60b062c21aa89e90533927c8571bc25b2f50c","first_computed_at":"2026-05-18T03:50:37.320811Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:50:37.320811Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GONj9MhQjiTvl6BrmyMCA0wDSux/YUXhR4bjG+xi3UuQudkbwYos1K2VPELDuUvThu1hlIi4ukfZyQkutQJ6Cw==","signature_status":"signed_v1","signed_at":"2026-05-18T03:50:37.321499Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.06978","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:54963ba7a22570ba81edd6c9a90692cdb6c477e52ddbe708117d63a9b5612b14","sha256:a74ca52d058dc5d403a4e9481041bf51169b25ddfa8c5c93727dd7fc8413e413"],"state_sha256":"794357226838f2ab90cd2d67189a35227126898aea08f76c2f464ec4960283fd"}