{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:IAZSHC2CEOVJS5LFPG2ACLALBB","short_pith_number":"pith:IAZSHC2C","canonical_record":{"source":{"id":"2401.03321","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-06T22:49:38Z","cross_cats_sorted":[],"title_canon_sha256":"116acf5c8e86ca342bfc6e9b68d7e0a213ef1043dd459235cf3e67ded5134f23","abstract_canon_sha256":"15fed9cdec205952b193885672ba07c81d187e36d77f407888593a2b86873241"},"schema_version":"1.0"},"canonical_sha256":"4033238b4223aa99756579b4012c0b087ef2121d56f6c9c345a18709683a0609","source":{"kind":"arxiv","id":"2401.03321","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.03321","created_at":"2026-07-05T07:48:51Z"},{"alias_kind":"arxiv_version","alias_value":"2401.03321v2","created_at":"2026-07-05T07:48:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.03321","created_at":"2026-07-05T07:48:51Z"},{"alias_kind":"pith_short_12","alias_value":"IAZSHC2CEOVJ","created_at":"2026-07-05T07:48:51Z"},{"alias_kind":"pith_short_16","alias_value":"IAZSHC2CEOVJS5LF","created_at":"2026-07-05T07:48:51Z"},{"alias_kind":"pith_short_8","alias_value":"IAZSHC2C","created_at":"2026-07-05T07:48:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:IAZSHC2CEOVJS5LFPG2ACLALBB","target":"record","payload":{"canonical_record":{"source":{"id":"2401.03321","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-06T22:49:38Z","cross_cats_sorted":[],"title_canon_sha256":"116acf5c8e86ca342bfc6e9b68d7e0a213ef1043dd459235cf3e67ded5134f23","abstract_canon_sha256":"15fed9cdec205952b193885672ba07c81d187e36d77f407888593a2b86873241"},"schema_version":"1.0"},"canonical_sha256":"4033238b4223aa99756579b4012c0b087ef2121d56f6c9c345a18709683a0609","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:48:51.619127Z","signature_b64":"5Uv20MPVeAG7ULAvdzHyIkpom/JLUrkVqkZbulGEx7f3AgtIXk7VVEnvITcU/JWmh+aZ8YbtzhjSA/dGuJ15Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4033238b4223aa99756579b4012c0b087ef2121d56f6c9c345a18709683a0609","last_reissued_at":"2026-07-05T07:48:51.618690Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:48:51.618690Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2401.03321","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":"2026-07-05T07:48:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"G95+nYxX10GF3iMG0zmmrBTW0SIcybhM7/XasqqChNCevLKlVZjVA6EKwJYtOins/4enZ4E7AnA+/26F89jMBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T11:17:40.213114Z"},"content_sha256":"8e2625ccb338cef70993a1427b31190cab2499ea82d69b75635068d34b1b4605","schema_version":"1.0","event_id":"sha256:8e2625ccb338cef70993a1427b31190cab2499ea82d69b75635068d34b1b4605"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:IAZSHC2CEOVJS5LFPG2ACLALBB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PIXAR: Auto-Regressive Language Modeling in Pixel Space","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alessandro Suglia, Antonio Vergari, Xiyang Liao, Yintao Tai","submitted_at":"2024-01-06T22:49:38Z","abstract_excerpt":"Recent work showed the possibility of building open-vocabulary large language models (LLMs) that directly operate on pixel representations. These models are implemented as autoencoders that reconstruct masked patches of rendered text. However, these pixel-based LLMs are limited to discriminative tasks (e.g., classification) and, similar to BERT, cannot be used to generate text. Therefore, they cannot be used for generative tasks such as free-form question answering. In this work, we introduce PIXAR, the first pixel-based autoregressive LLM that performs text generation. Consisting of only a de"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.03321","kind":"arxiv","version":2},"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/2401.03321/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":"2026-07-05T07:48:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hmgoAl8km/zMMllrNPvbTtV1PxVuyKqv4qyJN0jGJ8cpwKF03pbUNOfRXN700KqmP6WPjSeQ9vuiQ9Aq85HbBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T11:17:40.213625Z"},"content_sha256":"770923cf9644fd7580899d1c14390733dc70ef6b46ba434343f3aa44f526103f","schema_version":"1.0","event_id":"sha256:770923cf9644fd7580899d1c14390733dc70ef6b46ba434343f3aa44f526103f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IAZSHC2CEOVJS5LFPG2ACLALBB/bundle.json","state_url":"https://pith.science/pith/IAZSHC2CEOVJS5LFPG2ACLALBB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IAZSHC2CEOVJS5LFPG2ACLALBB/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":"2026-08-14T11:17:40Z","links":{"resolver":"https://pith.science/pith/IAZSHC2CEOVJS5LFPG2ACLALBB","bundle":"https://pith.science/pith/IAZSHC2CEOVJS5LFPG2ACLALBB/bundle.json","state":"https://pith.science/pith/IAZSHC2CEOVJS5LFPG2ACLALBB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IAZSHC2CEOVJS5LFPG2ACLALBB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:IAZSHC2CEOVJS5LFPG2ACLALBB","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":"15fed9cdec205952b193885672ba07c81d187e36d77f407888593a2b86873241","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-06T22:49:38Z","title_canon_sha256":"116acf5c8e86ca342bfc6e9b68d7e0a213ef1043dd459235cf3e67ded5134f23"},"schema_version":"1.0","source":{"id":"2401.03321","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.03321","created_at":"2026-07-05T07:48:51Z"},{"alias_kind":"arxiv_version","alias_value":"2401.03321v2","created_at":"2026-07-05T07:48:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.03321","created_at":"2026-07-05T07:48:51Z"},{"alias_kind":"pith_short_12","alias_value":"IAZSHC2CEOVJ","created_at":"2026-07-05T07:48:51Z"},{"alias_kind":"pith_short_16","alias_value":"IAZSHC2CEOVJS5LF","created_at":"2026-07-05T07:48:51Z"},{"alias_kind":"pith_short_8","alias_value":"IAZSHC2C","created_at":"2026-07-05T07:48:51Z"}],"graph_snapshots":[{"event_id":"sha256:770923cf9644fd7580899d1c14390733dc70ef6b46ba434343f3aa44f526103f","target":"graph","created_at":"2026-07-05T07:48:51Z","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/2401.03321/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent work showed the possibility of building open-vocabulary large language models (LLMs) that directly operate on pixel representations. These models are implemented as autoencoders that reconstruct masked patches of rendered text. However, these pixel-based LLMs are limited to discriminative tasks (e.g., classification) and, similar to BERT, cannot be used to generate text. Therefore, they cannot be used for generative tasks such as free-form question answering. In this work, we introduce PIXAR, the first pixel-based autoregressive LLM that performs text generation. Consisting of only a de","authors_text":"Alessandro Suglia, Antonio Vergari, Xiyang Liao, Yintao Tai","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-06T22:49:38Z","title":"PIXAR: Auto-Regressive Language Modeling in Pixel Space"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.03321","kind":"arxiv","version":2},"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":"sha256:8e2625ccb338cef70993a1427b31190cab2499ea82d69b75635068d34b1b4605","target":"record","created_at":"2026-07-05T07:48:51Z","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":"15fed9cdec205952b193885672ba07c81d187e36d77f407888593a2b86873241","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-06T22:49:38Z","title_canon_sha256":"116acf5c8e86ca342bfc6e9b68d7e0a213ef1043dd459235cf3e67ded5134f23"},"schema_version":"1.0","source":{"id":"2401.03321","kind":"arxiv","version":2}},"canonical_sha256":"4033238b4223aa99756579b4012c0b087ef2121d56f6c9c345a18709683a0609","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4033238b4223aa99756579b4012c0b087ef2121d56f6c9c345a18709683a0609","first_computed_at":"2026-07-05T07:48:51.618690Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:48:51.618690Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"5Uv20MPVeAG7ULAvdzHyIkpom/JLUrkVqkZbulGEx7f3AgtIXk7VVEnvITcU/JWmh+aZ8YbtzhjSA/dGuJ15Cw==","signature_status":"signed_v1","signed_at":"2026-07-05T07:48:51.619127Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.03321","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8e2625ccb338cef70993a1427b31190cab2499ea82d69b75635068d34b1b4605","sha256:770923cf9644fd7580899d1c14390733dc70ef6b46ba434343f3aa44f526103f"],"state_sha256":"d0c017158611504abe56529a556313d6069154478a2bdf40561861f777e5af9c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qPyziHct1PFCgFS2yDGlPwUgIhhYy0RgYPmn90Sh93HiMT4SkQE384wy2mLkQn6kJuXZ4krDq+sRJUlEKqJeDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T11:17:40.219695Z","bundle_sha256":"9b4913f7caddabeb08c1bde3a8ed057fb8eaad91b34a3f7bffed53bb78dc0c4a"}}