{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:REHXFOZW6PXSQEBZLYBBMSPWUZ","short_pith_number":"pith:REHXFOZW","canonical_record":{"source":{"id":"2305.10430","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-17T17:59:11Z","cross_cats_sorted":[],"title_canon_sha256":"f6387b9d58db4b8d251016b2a10cacce8ebb26ce09331af8f03cb56a5f177f45","abstract_canon_sha256":"d4ae155f41e9d06c0e9da8f5177f1554cae68f39b690e191977bd2e5ddc01ed2"},"schema_version":"1.0"},"canonical_sha256":"890f72bb36f3ef2810395e021649f6a66e3e9ca152edf2253b4fb890297ce186","source":{"kind":"arxiv","id":"2305.10430","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.10430","created_at":"2026-05-17T23:38:14Z"},{"alias_kind":"arxiv_version","alias_value":"2305.10430v2","created_at":"2026-05-17T23:38:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.10430","created_at":"2026-05-17T23:38:14Z"},{"alias_kind":"pith_short_12","alias_value":"REHXFOZW6PXS","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"REHXFOZW6PXSQEBZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"REHXFOZW","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:REHXFOZW6PXSQEBZLYBBMSPWUZ","target":"record","payload":{"canonical_record":{"source":{"id":"2305.10430","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-17T17:59:11Z","cross_cats_sorted":[],"title_canon_sha256":"f6387b9d58db4b8d251016b2a10cacce8ebb26ce09331af8f03cb56a5f177f45","abstract_canon_sha256":"d4ae155f41e9d06c0e9da8f5177f1554cae68f39b690e191977bd2e5ddc01ed2"},"schema_version":"1.0"},"canonical_sha256":"890f72bb36f3ef2810395e021649f6a66e3e9ca152edf2253b4fb890297ce186","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:14.596546Z","signature_b64":"RBCo2/t8eApr2zPderV4KvLwxeffFUjvQpJ+ZWGCu/z98xrrxLDtaSWScMcyi64PPZGPl4tkGKcfugSVaPbfAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"890f72bb36f3ef2810395e021649f6a66e3e9ca152edf2253b4fb890297ce186","last_reissued_at":"2026-05-17T23:38:14.595898Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:14.595898Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2305.10430","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LXNu02AoQVeatTxySBZf99x3oRCIK72HuPLL6HX2fdVrtRrJdunCA+zQNnMjerHjdcJFv2pbx3axaSF58dcICg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T23:16:06.031779Z"},"content_sha256":"5ca6d424470bb0033dba066edf048fb1e00dea7177ce714a5ae8923b62e08486","schema_version":"1.0","event_id":"sha256:5ca6d424470bb0033dba066edf048fb1e00dea7177ce714a5ae8923b62e08486"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:REHXFOZW6PXSQEBZLYBBMSPWUZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Rethinking the Open-Loop Evaluation of End-to-End Autonomous Driving in nuScenes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A simple MLP using only past trajectories and velocity matches perception-based planners on nuScenes L2 error.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiang-Jiang Liu, Jiang-Tian Zhai, Jingdong Wang, Jinhao Du, Xiaoqing Ye, Yifu Zhang, Yongqiang Mao, Ze Feng, Zichang Tan","submitted_at":"2023-05-17T17:59:11Z","abstract_excerpt":"Modern autonomous driving systems are typically divided into three main tasks: perception, prediction, and planning. The planning task involves predicting the trajectory of the ego vehicle based on inputs from both internal intention and the external environment, and manipulating the vehicle accordingly. Most existing works evaluate their performance on the nuScenes dataset using the L2 error and collision rate between the predicted trajectories and the ground truth. In this paper, we reevaluate these existing evaluation metrics and explore whether they accurately measure the superiority of di"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our simple method achieves similar end-to-end planning performance on the nuScenes dataset with other perception-based methods, reducing the average L2 error by about 20%.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the MLP baseline is trained and evaluated under identical conditions and data splits as the perception-based competitors, with no hidden advantages from trajectory history alone.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A perception-free MLP reduces average L2 trajectory error by ~20% versus perception-based methods on nuScenes, suggesting current open-loop evaluation may reward trajectory mimicry over safe planning.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A simple MLP using only past trajectories and velocity matches perception-based planners on nuScenes L2 error.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"f4031785ae0b6405f0a39a9b4aa4e04a18ea65b04ba39e77d0b9aaa0c0e0d4d4"},"source":{"id":"2305.10430","kind":"arxiv","version":2},"verdict":{"id":"3f09c26b-5774-4036-880c-d43ac8b31c4e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T08:12:15.902335Z","strongest_claim":"Our simple method achieves similar end-to-end planning performance on the nuScenes dataset with other perception-based methods, reducing the average L2 error by about 20%.","one_line_summary":"A perception-free MLP reduces average L2 trajectory error by ~20% versus perception-based methods on nuScenes, suggesting current open-loop evaluation may reward trajectory mimicry over safe planning.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the MLP baseline is trained and evaluated under identical conditions and data splits as the perception-based competitors, with no hidden advantages from trajectory history alone.","pith_extraction_headline":"A simple MLP using only past trajectories and velocity matches perception-based planners on nuScenes L2 error."},"references":{"count":18,"sample":[{"doi":"","year":2020,"title":"nuscenes: A multi- modal dataset for autonomous driving","work_id":"f271e994-7c0e-496e-9efc-13301f68117e","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"What data do we need for train- ing an A V motion planner? InICRA, 2021","work_id":"47a92981-fce2-42dd-8951-13e1d2b0a5b5","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Transfuser: Imitation with transformer-based sensor fusion for autonomous driv- ing","work_id":"092fd10b-fef2-4d00-8168-e2de0f794b4b","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Densetnt: End-to-end trajectory prediction from dense goal sets","work_id":"6358b347-0e19-4c88-843d-97a52f86c69e","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"FIERY: Future instance segmentation in bird’s-eye view from surround monocular cameras","work_id":"f75f13c2-7034-43f7-961a-2152133f9bf2","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":18,"snapshot_sha256":"184cbf060b2c5ad681b523dc1187d9e86aad969c2ef1698a706c9c142d004cf8","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"739b1f39fb41a950e857dd334d84cbcc643509f47e267edec1ce14b6dc3c13d6"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"3f09c26b-5774-4036-880c-d43ac8b31c4e"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+Ckuoe7Eky6P5vgSMOTqUSqBrn36LNkEgLAe8xj6VeE6/G1kjroNaFrfP/ynbolI2aywDKxV7+9qyIBTNTdCDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T23:16:06.032372Z"},"content_sha256":"2bdc5f1dd674ee27458c6a29772900d03d359184fe634ca72c396c1e725335f9","schema_version":"1.0","event_id":"sha256:2bdc5f1dd674ee27458c6a29772900d03d359184fe634ca72c396c1e725335f9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/REHXFOZW6PXSQEBZLYBBMSPWUZ/bundle.json","state_url":"https://pith.science/pith/REHXFOZW6PXSQEBZLYBBMSPWUZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/REHXFOZW6PXSQEBZLYBBMSPWUZ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-03T23:16:06Z","links":{"resolver":"https://pith.science/pith/REHXFOZW6PXSQEBZLYBBMSPWUZ","bundle":"https://pith.science/pith/REHXFOZW6PXSQEBZLYBBMSPWUZ/bundle.json","state":"https://pith.science/pith/REHXFOZW6PXSQEBZLYBBMSPWUZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/REHXFOZW6PXSQEBZLYBBMSPWUZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:REHXFOZW6PXSQEBZLYBBMSPWUZ","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":"d4ae155f41e9d06c0e9da8f5177f1554cae68f39b690e191977bd2e5ddc01ed2","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-17T17:59:11Z","title_canon_sha256":"f6387b9d58db4b8d251016b2a10cacce8ebb26ce09331af8f03cb56a5f177f45"},"schema_version":"1.0","source":{"id":"2305.10430","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.10430","created_at":"2026-05-17T23:38:14Z"},{"alias_kind":"arxiv_version","alias_value":"2305.10430v2","created_at":"2026-05-17T23:38:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.10430","created_at":"2026-05-17T23:38:14Z"},{"alias_kind":"pith_short_12","alias_value":"REHXFOZW6PXS","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"REHXFOZW6PXSQEBZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"REHXFOZW","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:2bdc5f1dd674ee27458c6a29772900d03d359184fe634ca72c396c1e725335f9","target":"graph","created_at":"2026-05-17T23:38:14Z","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":"Our simple method achieves similar end-to-end planning performance on the nuScenes dataset with other perception-based methods, reducing the average L2 error by about 20%."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the MLP baseline is trained and evaluated under identical conditions and data splits as the perception-based competitors, with no hidden advantages from trajectory history alone."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A perception-free MLP reduces average L2 trajectory error by ~20% versus perception-based methods on nuScenes, suggesting current open-loop evaluation may reward trajectory mimicry over safe planning."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A simple MLP using only past trajectories and velocity matches perception-based planners on nuScenes L2 error."}],"snapshot_sha256":"f4031785ae0b6405f0a39a9b4aa4e04a18ea65b04ba39e77d0b9aaa0c0e0d4d4"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"739b1f39fb41a950e857dd334d84cbcc643509f47e267edec1ce14b6dc3c13d6"},"paper":{"abstract_excerpt":"Modern autonomous driving systems are typically divided into three main tasks: perception, prediction, and planning. The planning task involves predicting the trajectory of the ego vehicle based on inputs from both internal intention and the external environment, and manipulating the vehicle accordingly. Most existing works evaluate their performance on the nuScenes dataset using the L2 error and collision rate between the predicted trajectories and the ground truth. In this paper, we reevaluate these existing evaluation metrics and explore whether they accurately measure the superiority of di","authors_text":"Jiang-Jiang Liu, Jiang-Tian Zhai, Jingdong Wang, Jinhao Du, Xiaoqing Ye, Yifu Zhang, Yongqiang Mao, Ze Feng, Zichang Tan","cross_cats":[],"headline":"A simple MLP using only past trajectories and velocity matches perception-based planners on nuScenes L2 error.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-17T17:59:11Z","title":"Rethinking the Open-Loop Evaluation of End-to-End Autonomous Driving in nuScenes"},"references":{"count":18,"internal_anchors":0,"resolved_work":18,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"nuscenes: A multi- modal dataset for autonomous driving","work_id":"f271e994-7c0e-496e-9efc-13301f68117e","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"What data do we need for train- ing an A V motion planner? InICRA, 2021","work_id":"47a92981-fce2-42dd-8951-13e1d2b0a5b5","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Transfuser: Imitation with transformer-based sensor fusion for autonomous driv- ing","work_id":"092fd10b-fef2-4d00-8168-e2de0f794b4b","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Densetnt: End-to-end trajectory prediction from dense goal sets","work_id":"6358b347-0e19-4c88-843d-97a52f86c69e","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"FIERY: Future instance segmentation in bird’s-eye view from surround monocular cameras","work_id":"f75f13c2-7034-43f7-961a-2152133f9bf2","year":null}],"snapshot_sha256":"184cbf060b2c5ad681b523dc1187d9e86aad969c2ef1698a706c9c142d004cf8"},"source":{"id":"2305.10430","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-17T08:12:15.902335Z","id":"3f09c26b-5774-4036-880c-d43ac8b31c4e","model_set":{"reader":"grok-4.3"},"one_line_summary":"A perception-free MLP reduces average L2 trajectory error by ~20% versus perception-based methods on nuScenes, suggesting current open-loop evaluation may reward trajectory mimicry over safe planning.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A simple MLP using only past trajectories and velocity matches perception-based planners on nuScenes L2 error.","strongest_claim":"Our simple method achieves similar end-to-end planning performance on the nuScenes dataset with other perception-based methods, reducing the average L2 error by about 20%.","weakest_assumption":"That the MLP baseline is trained and evaluated under identical conditions and data splits as the perception-based competitors, with no hidden advantages from trajectory history alone."}},"verdict_id":"3f09c26b-5774-4036-880c-d43ac8b31c4e"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:5ca6d424470bb0033dba066edf048fb1e00dea7177ce714a5ae8923b62e08486","target":"record","created_at":"2026-05-17T23:38:14Z","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":"d4ae155f41e9d06c0e9da8f5177f1554cae68f39b690e191977bd2e5ddc01ed2","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-17T17:59:11Z","title_canon_sha256":"f6387b9d58db4b8d251016b2a10cacce8ebb26ce09331af8f03cb56a5f177f45"},"schema_version":"1.0","source":{"id":"2305.10430","kind":"arxiv","version":2}},"canonical_sha256":"890f72bb36f3ef2810395e021649f6a66e3e9ca152edf2253b4fb890297ce186","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"890f72bb36f3ef2810395e021649f6a66e3e9ca152edf2253b4fb890297ce186","first_computed_at":"2026-05-17T23:38:14.595898Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:14.595898Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RBCo2/t8eApr2zPderV4KvLwxeffFUjvQpJ+ZWGCu/z98xrrxLDtaSWScMcyi64PPZGPl4tkGKcfugSVaPbfAQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:14.596546Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.10430","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5ca6d424470bb0033dba066edf048fb1e00dea7177ce714a5ae8923b62e08486","sha256:2bdc5f1dd674ee27458c6a29772900d03d359184fe634ca72c396c1e725335f9"],"state_sha256":"d17f83ccf22b88b961c5c950a1420e17d85085cbe0bc48f44a7876c987771a4b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qfRXRdLAX0r3Mo5BbgouxmgyH0057+Ac+++2kJK2QKdQqlu82hb6bmSlkD0T6bfxgLCaY1Gq4zahIQK5JhVMAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-03T23:16:06.034906Z","bundle_sha256":"20699e7236576efba8e6dc145db82909f44c9c0bacb375bbab6530ab29645144"}}