{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:QGOCE4Q3DQPPKK5JCYYAIHIA2J","short_pith_number":"pith:QGOCE4Q3","canonical_record":{"source":{"id":"2604.02714","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-03T04:14:13Z","cross_cats_sorted":[],"title_canon_sha256":"5a4e080a757fe678ddd72d06339cbe70306a74802b42d0eb1be850379a315768","abstract_canon_sha256":"66556723c5ac9f806e37ae13c11a8a0060be8111ced4c43e23ac68b1bea9640f"},"schema_version":"1.0"},"canonical_sha256":"819c22721b1c1ef52ba91630041d00d27f7f0d18bac569471e463cd627b3023f","source":{"kind":"arxiv","id":"2604.02714","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.02714","created_at":"2026-06-30T02:17:20Z"},{"alias_kind":"arxiv_version","alias_value":"2604.02714v2","created_at":"2026-06-30T02:17:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.02714","created_at":"2026-06-30T02:17:20Z"},{"alias_kind":"pith_short_12","alias_value":"QGOCE4Q3DQPP","created_at":"2026-06-30T02:17:20Z"},{"alias_kind":"pith_short_16","alias_value":"QGOCE4Q3DQPPKK5J","created_at":"2026-06-30T02:17:20Z"},{"alias_kind":"pith_short_8","alias_value":"QGOCE4Q3","created_at":"2026-06-30T02:17:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:QGOCE4Q3DQPPKK5JCYYAIHIA2J","target":"record","payload":{"canonical_record":{"source":{"id":"2604.02714","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-03T04:14:13Z","cross_cats_sorted":[],"title_canon_sha256":"5a4e080a757fe678ddd72d06339cbe70306a74802b42d0eb1be850379a315768","abstract_canon_sha256":"66556723c5ac9f806e37ae13c11a8a0060be8111ced4c43e23ac68b1bea9640f"},"schema_version":"1.0"},"canonical_sha256":"819c22721b1c1ef52ba91630041d00d27f7f0d18bac569471e463cd627b3023f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T02:17:20.039500Z","signature_b64":"xCp7cm2YmXSvU5asUqecLKZ0NxVu0i8Yw8VQlEyR/zt8aVvXVLa9WzhgvJ638XsDXCIeh1es607mk1HsEmCQDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"819c22721b1c1ef52ba91630041d00d27f7f0d18bac569471e463cd627b3023f","last_reissued_at":"2026-06-30T02:17:20.038914Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T02:17:20.038914Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.02714","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":"2026-06-30T02:17:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LQEr4+BKo6y3cWYMrO1NR1z6rjxKgfRyLdoZzZaE/4JFC4WFr3Dw0fNcCIWmtZCFA9vpfZLMrat+4ITCkFtyDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T00:00:22.743195Z"},"content_sha256":"ad102db81c55c8a504183b08e7ba4fdae62ecae25509174b4888fc8fdf3c0dc1","schema_version":"1.0","event_id":"sha256:ad102db81c55c8a504183b08e7ba4fdae62ecae25509174b4888fc8fdf3c0dc1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:QGOCE4Q3DQPPKK5JCYYAIHIA2J","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ExploreVLA: Dense World Modeling and Exploration for End-to-End Autonomous Driving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Augmenting VLA driving models with future image prediction supplies both dense supervision and an uncertainty signal for safe policy exploration.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jingru Luo, Liu Ren, Sikai Chen, Xin Ye, Zihao Sheng","submitted_at":"2026-04-03T04:14:13Z","abstract_excerpt":"End-to-end autonomous driving models based on Vision-Language-Action (VLA) architectures have shown promising results by learning driving policies through behavior cloning on expert demonstrations. However, imitation learning inherently limits the model to replicating observed behaviors without exploring diverse driving strategies, leaving it brittle in novel or out-of-distribution scenarios. Reinforcement learning (RL) offers a natural remedy by enabling policy exploration beyond the expert distribution. Yet VLA models, typically trained on offline datasets, lack directly observable state tra"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experiments on the NAVSIM and nuScenes benchmarks demonstrate the effectiveness of our approach, achieving a state-of-the-art PDMS score of 93.7 and an EPDMS of 88.8 on NAVSIM.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the world model's image prediction uncertainty reliably indicates both novelty and safety, allowing the safety-gated reward to produce valuable exploration without introducing unsafe behaviors or training instability.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"ExploreVLA augments VLA driving models with future RGB and depth prediction for dense supervision and uses prediction uncertainty as a safety-gated intrinsic reward for RL-based exploration, reaching SOTA PDMS 93.7 on NAVSIM.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Augmenting VLA driving models with future image prediction supplies both dense supervision and an uncertainty signal for safe policy exploration.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"f96b9547195532b5fa3bc83f25fa906b05a8741f6412d8fa3b2b2664d7740e2f"},"source":{"id":"2604.02714","kind":"arxiv","version":2},"verdict":{"id":"9cd83236-73a2-44dc-816b-ceb576fade08","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T20:49:17.071003Z","strongest_claim":"Experiments on the NAVSIM and nuScenes benchmarks demonstrate the effectiveness of our approach, achieving a state-of-the-art PDMS score of 93.7 and an EPDMS of 88.8 on NAVSIM.","one_line_summary":"ExploreVLA augments VLA driving models with future RGB and depth prediction for dense supervision and uses prediction uncertainty as a safety-gated intrinsic reward for RL-based exploration, reaching SOTA PDMS 93.7 on NAVSIM.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the world model's image prediction uncertainty reliably indicates both novelty and safety, allowing the safety-gated reward to produce valuable exploration without introducing unsafe behaviors or training instability.","pith_extraction_headline":"Augmenting VLA driving models with future image prediction supplies both dense supervision and an uncertainty signal for safe policy exploration."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.02714/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"},"verdict_id":"9cd83236-73a2-44dc-816b-ceb576fade08"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-30T02:17:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zMwZw9IkUoi/n/fScFxsnVF/s8zZnB/umj5XBSL8OyL7X+GZzCcXpoJPiPVEmnJptq/CgwA32+Ld+8Y6mmLvCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T00:00:22.743696Z"},"content_sha256":"91bf4025ef25a001dcae1c22fac9a1427ea18899796ea5fd9ab96be702c49845","schema_version":"1.0","event_id":"sha256:91bf4025ef25a001dcae1c22fac9a1427ea18899796ea5fd9ab96be702c49845"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QGOCE4Q3DQPPKK5JCYYAIHIA2J/bundle.json","state_url":"https://pith.science/pith/QGOCE4Q3DQPPKK5JCYYAIHIA2J/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QGOCE4Q3DQPPKK5JCYYAIHIA2J/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":"2026-08-04T00:00:22Z","links":{"resolver":"https://pith.science/pith/QGOCE4Q3DQPPKK5JCYYAIHIA2J","bundle":"https://pith.science/pith/QGOCE4Q3DQPPKK5JCYYAIHIA2J/bundle.json","state":"https://pith.science/pith/QGOCE4Q3DQPPKK5JCYYAIHIA2J/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QGOCE4Q3DQPPKK5JCYYAIHIA2J/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:QGOCE4Q3DQPPKK5JCYYAIHIA2J","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":"66556723c5ac9f806e37ae13c11a8a0060be8111ced4c43e23ac68b1bea9640f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-03T04:14:13Z","title_canon_sha256":"5a4e080a757fe678ddd72d06339cbe70306a74802b42d0eb1be850379a315768"},"schema_version":"1.0","source":{"id":"2604.02714","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.02714","created_at":"2026-06-30T02:17:20Z"},{"alias_kind":"arxiv_version","alias_value":"2604.02714v2","created_at":"2026-06-30T02:17:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.02714","created_at":"2026-06-30T02:17:20Z"},{"alias_kind":"pith_short_12","alias_value":"QGOCE4Q3DQPP","created_at":"2026-06-30T02:17:20Z"},{"alias_kind":"pith_short_16","alias_value":"QGOCE4Q3DQPPKK5J","created_at":"2026-06-30T02:17:20Z"},{"alias_kind":"pith_short_8","alias_value":"QGOCE4Q3","created_at":"2026-06-30T02:17:20Z"}],"graph_snapshots":[{"event_id":"sha256:91bf4025ef25a001dcae1c22fac9a1427ea18899796ea5fd9ab96be702c49845","target":"graph","created_at":"2026-06-30T02:17:20Z","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":"Experiments on the NAVSIM and nuScenes benchmarks demonstrate the effectiveness of our approach, achieving a state-of-the-art PDMS score of 93.7 and an EPDMS of 88.8 on NAVSIM."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the world model's image prediction uncertainty reliably indicates both novelty and safety, allowing the safety-gated reward to produce valuable exploration without introducing unsafe behaviors or training instability."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"ExploreVLA augments VLA driving models with future RGB and depth prediction for dense supervision and uses prediction uncertainty as a safety-gated intrinsic reward for RL-based exploration, reaching SOTA PDMS 93.7 on NAVSIM."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Augmenting VLA driving models with future image prediction supplies both dense supervision and an uncertainty signal for safe policy exploration."}],"snapshot_sha256":"f96b9547195532b5fa3bc83f25fa906b05a8741f6412d8fa3b2b2664d7740e2f"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2604.02714/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"End-to-end autonomous driving models based on Vision-Language-Action (VLA) architectures have shown promising results by learning driving policies through behavior cloning on expert demonstrations. However, imitation learning inherently limits the model to replicating observed behaviors without exploring diverse driving strategies, leaving it brittle in novel or out-of-distribution scenarios. Reinforcement learning (RL) offers a natural remedy by enabling policy exploration beyond the expert distribution. Yet VLA models, typically trained on offline datasets, lack directly observable state tra","authors_text":"Jingru Luo, Liu Ren, Sikai Chen, Xin Ye, Zihao Sheng","cross_cats":[],"headline":"Augmenting VLA driving models with future image prediction supplies both dense supervision and an uncertainty signal for safe policy exploration.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-03T04:14:13Z","title":"ExploreVLA: Dense World Modeling and Exploration for End-to-End Autonomous Driving"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.02714","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-13T20:49:17.071003Z","id":"9cd83236-73a2-44dc-816b-ceb576fade08","model_set":{"reader":"grok-4.3"},"one_line_summary":"ExploreVLA augments VLA driving models with future RGB and depth prediction for dense supervision and uses prediction uncertainty as a safety-gated intrinsic reward for RL-based exploration, reaching SOTA PDMS 93.7 on NAVSIM.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Augmenting VLA driving models with future image prediction supplies both dense supervision and an uncertainty signal for safe policy exploration.","strongest_claim":"Experiments on the NAVSIM and nuScenes benchmarks demonstrate the effectiveness of our approach, achieving a state-of-the-art PDMS score of 93.7 and an EPDMS of 88.8 on NAVSIM.","weakest_assumption":"That the world model's image prediction uncertainty reliably indicates both novelty and safety, allowing the safety-gated reward to produce valuable exploration without introducing unsafe behaviors or training instability."}},"verdict_id":"9cd83236-73a2-44dc-816b-ceb576fade08"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ad102db81c55c8a504183b08e7ba4fdae62ecae25509174b4888fc8fdf3c0dc1","target":"record","created_at":"2026-06-30T02:17:20Z","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":"66556723c5ac9f806e37ae13c11a8a0060be8111ced4c43e23ac68b1bea9640f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-03T04:14:13Z","title_canon_sha256":"5a4e080a757fe678ddd72d06339cbe70306a74802b42d0eb1be850379a315768"},"schema_version":"1.0","source":{"id":"2604.02714","kind":"arxiv","version":2}},"canonical_sha256":"819c22721b1c1ef52ba91630041d00d27f7f0d18bac569471e463cd627b3023f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"819c22721b1c1ef52ba91630041d00d27f7f0d18bac569471e463cd627b3023f","first_computed_at":"2026-06-30T02:17:20.038914Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-30T02:17:20.038914Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xCp7cm2YmXSvU5asUqecLKZ0NxVu0i8Yw8VQlEyR/zt8aVvXVLa9WzhgvJ638XsDXCIeh1es607mk1HsEmCQDw==","signature_status":"signed_v1","signed_at":"2026-06-30T02:17:20.039500Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.02714","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ad102db81c55c8a504183b08e7ba4fdae62ecae25509174b4888fc8fdf3c0dc1","sha256:91bf4025ef25a001dcae1c22fac9a1427ea18899796ea5fd9ab96be702c49845"],"state_sha256":"fe41a1bf7b1dd24de6d117f250a47debc2a83c166fa31070fecd7a30052b1792"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"p9RHfyhNFrQZtVr7We8whKBTeEqKdg4zldsz5Xi/61cs1kkwS+GCfu7KE+t4gTdNSQ0FZYsDysvK4Tfic151Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T00:00:22.746690Z","bundle_sha256":"7efdeea1f20352e00a5d6e89bba9d791f6cdc0972675123c5875552a264d8239"}}