{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GLKGAV5X7GZ6UH2YYO37LRJV2K","short_pith_number":"pith:GLKGAV5X","schema_version":"1.0","canonical_sha256":"32d46057b7f9b3ea1f58c3b7f5c535d29bd03f025b0062d741117e1c65a4537d","source":{"kind":"arxiv","id":"2604.22851","version":2},"attestation_state":"computed","paper":{"title":"EgoDyn-Bench: Evaluating Ego-Motion Understanding in Vision-Centric Foundation Models for Autonomous Driving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Vision-centric foundation models fail to align physical ego-motion concepts with visual observations in autonomous driving.","cross_cats":["cs.CL","cs.RO"],"primary_cat":"cs.CV","authors_text":"Dingrui Wang, Finn Rasmus Sch\\\"afer, Johannes Betz, Mattia Piccinini, Sebastian Schmidt, Stephan G\\\"unnemann, Thomas Stauner, Yuan Gao","submitted_at":"2026-04-22T07:49:02Z","abstract_excerpt":"While Vision-Language Models (VLMs) have advanced high-level reasoning in autonomous driving, their ability to ground this reasoning in the underlying physics of ego-motion remains poorly understood. We introduce EgoDyn-Bench [Project page: (https://tum-avs.github.io/EgoDyn-Bench-Website/), Code: (https://github.com/TUM-AVS/EgoDyn-Bench), Dataset: (https://huggingface.co/datasets/fnc1901/EgoDyn-Bench)], a diagnostic benchmark for evaluating the semantic ego-motion understanding of vision-centric foundation models. By mapping continuous vehicle kinematics to discrete motion concepts via a deter"},"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":"2604.22851","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-22T07:49:02Z","cross_cats_sorted":["cs.CL","cs.RO"],"title_canon_sha256":"a128506afe8c73951c8e331fd6c453891df8ef56af1a64c089f3bee35fb2b36f","abstract_canon_sha256":"3a48833869f746eaac78a46ed321385724be23d5f93032fea15757fc92230924"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-08T01:19:13.427749Z","signature_b64":"ncI6RqE3PJE2LIgac1ke+Hv1cJFxObbkDtweDdLuk3cIp3IjNSQifmwxt4n36XxV5IcfMp5+5kL8uiUikeTWBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"32d46057b7f9b3ea1f58c3b7f5c535d29bd03f025b0062d741117e1c65a4537d","last_reissued_at":"2026-07-08T01:19:13.427070Z","signature_status":"signed_v1","first_computed_at":"2026-07-08T01:19:13.427070Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EgoDyn-Bench: Evaluating Ego-Motion Understanding in Vision-Centric Foundation Models for Autonomous Driving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Vision-centric foundation models fail to align physical ego-motion concepts with visual observations in autonomous driving.","cross_cats":["cs.CL","cs.RO"],"primary_cat":"cs.CV","authors_text":"Dingrui Wang, Finn Rasmus Sch\\\"afer, Johannes Betz, Mattia Piccinini, Sebastian Schmidt, Stephan G\\\"unnemann, Thomas Stauner, Yuan Gao","submitted_at":"2026-04-22T07:49:02Z","abstract_excerpt":"While Vision-Language Models (VLMs) have advanced high-level reasoning in autonomous driving, their ability to ground this reasoning in the underlying physics of ego-motion remains poorly understood. We introduce EgoDyn-Bench [Project page: (https://tum-avs.github.io/EgoDyn-Bench-Website/), Code: (https://github.com/TUM-AVS/EgoDyn-Bench), Dataset: (https://huggingface.co/datasets/fnc1901/EgoDyn-Bench)], a diagnostic benchmark for evaluating the semantic ego-motion understanding of vision-centric foundation models. By mapping continuous vehicle kinematics to discrete motion concepts via a deter"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"This failure persists across model scales and domain-specific training, indicating a structural deficit in how current architectures couple visual perception with physical reasoning. We demonstrate that providing explicit trajectory encodings substantially restores physical consistency across all evaluated models, revealing a functional disentanglement between vision and language: egomotion logic is derived almost exclusively from the language modality, while visual observations contribute negligible additional signal.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The deterministic oracle accurately and unbiasedly maps continuous vehicle kinematics to discrete motion concepts, and the benchmark tasks isolate perception from other model capabilities without introducing confounding factors.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"EgoDyn-Bench reveals a perception bottleneck in vision-centric foundation models: ego-motion logic derives from language while visual input adds negligible signal, with explicit trajectories restoring consistency.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Vision-centric foundation models fail to align physical ego-motion concepts with visual observations in autonomous driving.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"57a800bd2a72b8ac5441a1571d74d9d318f7dd39ec2d5620f2f1e5fbaae6cd0e"},"source":{"id":"2604.22851","kind":"arxiv","version":2},"verdict":{"id":"96daf71d-32f3-4fc8-b5c4-37c76086a788","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T00:06:08.396622Z","strongest_claim":"This failure persists across model scales and domain-specific training, indicating a structural deficit in how current architectures couple visual perception with physical reasoning. We demonstrate that providing explicit trajectory encodings substantially restores physical consistency across all evaluated models, revealing a functional disentanglement between vision and language: egomotion logic is derived almost exclusively from the language modality, while visual observations contribute negligible additional signal.","one_line_summary":"EgoDyn-Bench reveals a perception bottleneck in vision-centric foundation models: ego-motion logic derives from language while visual input adds negligible signal, with explicit trajectories restoring consistency.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The deterministic oracle accurately and unbiasedly maps continuous vehicle kinematics to discrete motion concepts, and the benchmark tasks isolate perception from other model capabilities without introducing confounding factors.","pith_extraction_headline":"Vision-centric foundation models fail to align physical ego-motion concepts with visual observations in autonomous driving."},"integrity":{"clean":false,"summary":{"advisory":1,"critical":0,"by_detector":{"doi_compliance":{"total":1,"advisory":1,"critical":0,"informational":0}},"informational":0},"endpoint":"/pith/2604.22851/integrity.json","findings":[{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.1016/0004-3702(81) was visible in the surrounding text but could not be confirmed against doi.org as printed.","detector":"doi_compliance","severity":"advisory","ref_index":12,"audited_at":"2026-05-20T02:06:04.415398Z","detected_doi":"10.1016/0004-3702(81","finding_type":"recoverable_identifier","verdict_class":"incontrovertible","detected_arxiv_id":null}],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-21T14:43:42.791562Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-20T02:06:04.415398Z","status":"completed","version":"1.0.0","findings_count":1}],"snapshot_sha256":"c1905117ebb7fa685ab0136bde82614a3ebc595b9a699c054246fb4f5caf059d"},"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":"2604.22851","created_at":"2026-07-08T01:19:13.427125+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.22851v2","created_at":"2026-07-08T01:19:13.427125+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.22851","created_at":"2026-07-08T01:19:13.427125+00:00"},{"alias_kind":"pith_short_12","alias_value":"GLKGAV5X7GZ6","created_at":"2026-07-08T01:19:13.427125+00:00"},{"alias_kind":"pith_short_16","alias_value":"GLKGAV5X7GZ6UH2Y","created_at":"2026-07-08T01:19:13.427125+00:00"},{"alias_kind":"pith_short_8","alias_value":"GLKGAV5X","created_at":"2026-07-08T01:19:13.427125+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05966","citing_title":"Imagined Rollouts are Kinematic, Not Dynamic: A Diagnosis of Long-Horizon World-Model Failure","ref_index":11,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GLKGAV5X7GZ6UH2YYO37LRJV2K","json":"https://pith.science/pith/GLKGAV5X7GZ6UH2YYO37LRJV2K.json","graph_json":"https://pith.science/api/pith-number/GLKGAV5X7GZ6UH2YYO37LRJV2K/graph.json","events_json":"https://pith.science/api/pith-number/GLKGAV5X7GZ6UH2YYO37LRJV2K/events.json","paper":"https://pith.science/paper/GLKGAV5X"},"agent_actions":{"view_html":"https://pith.science/pith/GLKGAV5X7GZ6UH2YYO37LRJV2K","download_json":"https://pith.science/pith/GLKGAV5X7GZ6UH2YYO37LRJV2K.json","view_paper":"https://pith.science/paper/GLKGAV5X","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.22851&json=true","fetch_graph":"https://pith.science/api/pith-number/GLKGAV5X7GZ6UH2YYO37LRJV2K/graph.json","fetch_events":"https://pith.science/api/pith-number/GLKGAV5X7GZ6UH2YYO37LRJV2K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GLKGAV5X7GZ6UH2YYO37LRJV2K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GLKGAV5X7GZ6UH2YYO37LRJV2K/action/storage_attestation","attest_author":"https://pith.science/pith/GLKGAV5X7GZ6UH2YYO37LRJV2K/action/author_attestation","sign_citation":"https://pith.science/pith/GLKGAV5X7GZ6UH2YYO37LRJV2K/action/citation_signature","submit_replication":"https://pith.science/pith/GLKGAV5X7GZ6UH2YYO37LRJV2K/action/replication_record"}},"created_at":"2026-07-08T01:19:13.427125+00:00","updated_at":"2026-07-08T01:19:13.427125+00:00"}