{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:WOPPOUGGU5W3X7KEWSDRV7YCQF","short_pith_number":"pith:WOPPOUGG","canonical_record":{"source":{"id":"2512.12307","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-12-13T12:26:57Z","cross_cats_sorted":["cs.GR"],"title_canon_sha256":"be59b4e9e9375c8734ece525d438c93a13313ad3de53703227cb9468999e5dbd","abstract_canon_sha256":"4b825ec5bc64202ca4177257427cdf787946bb39895984d79d6b83dbc5f0dc30"},"schema_version":"1.0"},"canonical_sha256":"b39ef750c6a76dbbfd44b4871aff02815a5fe038f3d2c36f8f10e0d9ec007ce7","source":{"kind":"arxiv","id":"2512.12307","version":5},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.12307","created_at":"2026-07-31T01:33:01Z"},{"alias_kind":"arxiv_version","alias_value":"2512.12307v5","created_at":"2026-07-31T01:33:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.12307","created_at":"2026-07-31T01:33:01Z"},{"alias_kind":"pith_short_12","alias_value":"WOPPOUGGU5W3","created_at":"2026-07-31T01:33:01Z"},{"alias_kind":"pith_short_16","alias_value":"WOPPOUGGU5W3X7KE","created_at":"2026-07-31T01:33:01Z"},{"alias_kind":"pith_short_8","alias_value":"WOPPOUGG","created_at":"2026-07-31T01:33:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:WOPPOUGGU5W3X7KEWSDRV7YCQF","target":"record","payload":{"canonical_record":{"source":{"id":"2512.12307","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-12-13T12:26:57Z","cross_cats_sorted":["cs.GR"],"title_canon_sha256":"be59b4e9e9375c8734ece525d438c93a13313ad3de53703227cb9468999e5dbd","abstract_canon_sha256":"4b825ec5bc64202ca4177257427cdf787946bb39895984d79d6b83dbc5f0dc30"},"schema_version":"1.0"},"canonical_sha256":"b39ef750c6a76dbbfd44b4871aff02815a5fe038f3d2c36f8f10e0d9ec007ce7","receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b39ef750c6a76dbbfd44b4871aff02815a5fe038f3d2c36f8f10e0d9ec007ce7","last_reissued_at":"2026-07-31T01:33:01.473396Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-31T01:33:01.473396Z"},"source_kind":"arxiv","source_id":"2512.12307","source_version":5,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-31T01:33:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hc7KCikkaWiD44K/In7FWlTVJcazo8NEuBgmzFchibgFYz3zUg1IElobis38Ldo1fcHrMJtM0L8gwRLAMXdACA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T21:29:54.845028Z"},"content_sha256":"e79ac87e964ac8a6f723be967e89844d442baf1b980ef13166940206a9bcd9f2","schema_version":"1.0","event_id":"sha256:e79ac87e964ac8a6f723be967e89844d442baf1b980ef13166940206a9bcd9f2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:WOPPOUGGU5W3X7KEWSDRV7YCQF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MRD: Using Physically Based Differentiable Rendering to Probe Vision Models for 3D Scene Understanding","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.GR"],"primary_cat":"cs.CV","authors_text":"Benjamin Beilharz, Thomas S. A. Wallis","submitted_at":"2025-12-13T12:26:57Z","abstract_excerpt":"While deep learning methods have achieved impressive success in many vision benchmarks, it remains difficult to understand and explain the representations and decisions of these models. Though vision models are typically trained on 2D inputs, they are often assumed to develop an implicit representation of the underlying 3D scene (for example, showing tolerance to partial occlusion, or the ability to reason about relative depth). Here, we introduce MRD (metamers rendered differentiably), an approach that uses physically based differentiable rendering to probe vision models' implicit understandi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.12307","kind":"arxiv","version":5},"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/2512.12307/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":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-31T01:33:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"N2wdkrCpCP1OhKGRZkBibjJrlcXfJGUqMG30Ep2LFY89BTyCGKw1NHnMVDR0Y4c+oMb26l9FccMJznvfsIhhBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T21:29:54.845585Z"},"content_sha256":"9e1f54917d17ada64e98dbac8433c90f35c9fb61ed2726f241511226ae00d3dd","schema_version":"1.0","event_id":"sha256:9e1f54917d17ada64e98dbac8433c90f35c9fb61ed2726f241511226ae00d3dd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WOPPOUGGU5W3X7KEWSDRV7YCQF/bundle.json","state_url":"https://pith.science/pith/WOPPOUGGU5W3X7KEWSDRV7YCQF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WOPPOUGGU5W3X7KEWSDRV7YCQF/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":"2026-08-04T21:29:54Z","links":{"resolver":"https://pith.science/pith/WOPPOUGGU5W3X7KEWSDRV7YCQF","bundle":"https://pith.science/pith/WOPPOUGGU5W3X7KEWSDRV7YCQF/bundle.json","state":"https://pith.science/pith/WOPPOUGGU5W3X7KEWSDRV7YCQF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WOPPOUGGU5W3X7KEWSDRV7YCQF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:WOPPOUGGU5W3X7KEWSDRV7YCQF","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":"4b825ec5bc64202ca4177257427cdf787946bb39895984d79d6b83dbc5f0dc30","cross_cats_sorted":["cs.GR"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-12-13T12:26:57Z","title_canon_sha256":"be59b4e9e9375c8734ece525d438c93a13313ad3de53703227cb9468999e5dbd"},"schema_version":"1.0","source":{"id":"2512.12307","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.12307","created_at":"2026-07-31T01:33:01Z"},{"alias_kind":"arxiv_version","alias_value":"2512.12307v5","created_at":"2026-07-31T01:33:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.12307","created_at":"2026-07-31T01:33:01Z"},{"alias_kind":"pith_short_12","alias_value":"WOPPOUGGU5W3","created_at":"2026-07-31T01:33:01Z"},{"alias_kind":"pith_short_16","alias_value":"WOPPOUGGU5W3X7KE","created_at":"2026-07-31T01:33:01Z"},{"alias_kind":"pith_short_8","alias_value":"WOPPOUGG","created_at":"2026-07-31T01:33:01Z"}],"graph_snapshots":[{"event_id":"sha256:9e1f54917d17ada64e98dbac8433c90f35c9fb61ed2726f241511226ae00d3dd","target":"graph","created_at":"2026-07-31T01:33:01Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2512.12307/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While deep learning methods have achieved impressive success in many vision benchmarks, it remains difficult to understand and explain the representations and decisions of these models. Though vision models are typically trained on 2D inputs, they are often assumed to develop an implicit representation of the underlying 3D scene (for example, showing tolerance to partial occlusion, or the ability to reason about relative depth). Here, we introduce MRD (metamers rendered differentiably), an approach that uses physically based differentiable rendering to probe vision models' implicit understandi","authors_text":"Benjamin Beilharz, Thomas S. A. Wallis","cross_cats":["cs.GR"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-12-13T12:26:57Z","title":"MRD: Using Physically Based Differentiable Rendering to Probe Vision Models for 3D Scene Understanding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.12307","kind":"arxiv","version":5},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e79ac87e964ac8a6f723be967e89844d442baf1b980ef13166940206a9bcd9f2","target":"record","created_at":"2026-07-31T01:33:01Z","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":"4b825ec5bc64202ca4177257427cdf787946bb39895984d79d6b83dbc5f0dc30","cross_cats_sorted":["cs.GR"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-12-13T12:26:57Z","title_canon_sha256":"be59b4e9e9375c8734ece525d438c93a13313ad3de53703227cb9468999e5dbd"},"schema_version":"1.0","source":{"id":"2512.12307","kind":"arxiv","version":5}},"canonical_sha256":"b39ef750c6a76dbbfd44b4871aff02815a5fe038f3d2c36f8f10e0d9ec007ce7","receipt":{"builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b39ef750c6a76dbbfd44b4871aff02815a5fe038f3d2c36f8f10e0d9ec007ce7","first_computed_at":"2026-07-31T01:33:01.473396Z","kind":"pith_receipt","last_reissued_at":"2026-07-31T01:33:01.473396Z","receipt_version":"0.3","signature_status":"unsigned_v0"},"source_id":"2512.12307","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e79ac87e964ac8a6f723be967e89844d442baf1b980ef13166940206a9bcd9f2","sha256:9e1f54917d17ada64e98dbac8433c90f35c9fb61ed2726f241511226ae00d3dd"],"state_sha256":"bf6b81dc9a4782103a0210be39c1b1927c54446b01ed727ed691a56f5377a8c6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cP+McoIoT8YyH9u10n2y1h58lDapC4cgjwRHhtP0s1GhKKEdDkXEKU6B/YOpzwhSR5/hpUJ+W5mSPKEhBIpXDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T21:29:54.851358Z","bundle_sha256":"d07584e9fc5f8467b7820d5c28190896c66d9626e866468788ade5099cd64819"}}