{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:OC4PEHPWOBCH66J3ME6XSLYZL3","short_pith_number":"pith:OC4PEHPW","canonical_record":{"source":{"id":"2410.05078","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-07T14:32:03Z","cross_cats_sorted":["cs.AI","cs.IT","math.IT"],"title_canon_sha256":"534287d34ac3e2be0a4d424a5cc50574175c6f09bb908340ebfb832e92257aa1","abstract_canon_sha256":"c2f636ee77d77ff6612a6deb5c3f8c244c3f48fd5d267ac86778507e50033e5c"},"schema_version":"1.0"},"canonical_sha256":"70b8f21df670447f793b613d792f195ee32fe0bb0c78099110a7c7f2a0c3b3ce","source":{"kind":"arxiv","id":"2410.05078","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.05078","created_at":"2026-07-05T11:08:01Z"},{"alias_kind":"arxiv_version","alias_value":"2410.05078v2","created_at":"2026-07-05T11:08:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.05078","created_at":"2026-07-05T11:08:01Z"},{"alias_kind":"pith_short_12","alias_value":"OC4PEHPWOBCH","created_at":"2026-07-05T11:08:01Z"},{"alias_kind":"pith_short_16","alias_value":"OC4PEHPWOBCH66J3","created_at":"2026-07-05T11:08:01Z"},{"alias_kind":"pith_short_8","alias_value":"OC4PEHPW","created_at":"2026-07-05T11:08:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:OC4PEHPWOBCH66J3ME6XSLYZL3","target":"record","payload":{"canonical_record":{"source":{"id":"2410.05078","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-07T14:32:03Z","cross_cats_sorted":["cs.AI","cs.IT","math.IT"],"title_canon_sha256":"534287d34ac3e2be0a4d424a5cc50574175c6f09bb908340ebfb832e92257aa1","abstract_canon_sha256":"c2f636ee77d77ff6612a6deb5c3f8c244c3f48fd5d267ac86778507e50033e5c"},"schema_version":"1.0"},"canonical_sha256":"70b8f21df670447f793b613d792f195ee32fe0bb0c78099110a7c7f2a0c3b3ce","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:01.528682Z","signature_b64":"os1bvFEFP3cotYMJ2FpeP52/EYApkCi7sdQpufjtPz7Tt3IikrSOV2QlAmmTJ6lT/Dvl3gh/W/Juki1VkhY1DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"70b8f21df670447f793b613d792f195ee32fe0bb0c78099110a7c7f2a0c3b3ce","last_reissued_at":"2026-07-05T11:08:01.528011Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:01.528011Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.05078","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":"2026-07-05T11:08:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gWY/MppR+3ReVLpyz2qYG+/LGSf+MYCjPw9TXCKNMBloQqo6uyQWpPq++45wfxRXWPCNWA1vSUZYxBiun8jfBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-28T08:14:07.816991Z"},"content_sha256":"a4523e6e4d0d993188c0cd4a121940dbfb0416bd0a1646dc067959f561c3b5c0","schema_version":"1.0","event_id":"sha256:a4523e6e4d0d993188c0cd4a121940dbfb0416bd0a1646dc067959f561c3b5c0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:OC4PEHPWOBCH66J3ME6XSLYZL3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Compression via Pre-trained Transformers: A Study on Byte-Level Multimodal Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Anian Ruoss, David Heurtel-Depeiges, Joel Veness, Tim Genewein","submitted_at":"2024-10-07T14:32:03Z","abstract_excerpt":"Foundation models are strong data compressors, but when accounting for their parameter size, their compression ratios are inferior to standard compression algorithms. Naively reducing the parameter count does not necessarily help as it deteriorates predictions and, accordingly, compression. We conduct a large-scale empirical study to find a sweet spot where pre-trained vanilla transformers can achieve competitive compression ratios. To this end, we train models on 165GB of raw byte sequences of either text, image, or audio data (and all possible combinations of the three) and then compress 1GB"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.05078","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/2410.05078/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":"2026-07-05T11:08:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nYMoQCYalupXJin+5ayI3EiW3kPzyA1kmDk1p3iZSHDhPwimWxZrd6iBYrWgIt0xpPYGEbvWb/SJWKYH5wS3AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-28T08:14:07.817391Z"},"content_sha256":"659462ee16f6df973d190e587fd2aafc3b408c4495715f38d28018541a8d3de0","schema_version":"1.0","event_id":"sha256:659462ee16f6df973d190e587fd2aafc3b408c4495715f38d28018541a8d3de0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OC4PEHPWOBCH66J3ME6XSLYZL3/bundle.json","state_url":"https://pith.science/pith/OC4PEHPWOBCH66J3ME6XSLYZL3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OC4PEHPWOBCH66J3ME6XSLYZL3/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":"2026-07-28T08:14:07Z","links":{"resolver":"https://pith.science/pith/OC4PEHPWOBCH66J3ME6XSLYZL3","bundle":"https://pith.science/pith/OC4PEHPWOBCH66J3ME6XSLYZL3/bundle.json","state":"https://pith.science/pith/OC4PEHPWOBCH66J3ME6XSLYZL3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OC4PEHPWOBCH66J3ME6XSLYZL3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:OC4PEHPWOBCH66J3ME6XSLYZL3","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":"c2f636ee77d77ff6612a6deb5c3f8c244c3f48fd5d267ac86778507e50033e5c","cross_cats_sorted":["cs.AI","cs.IT","math.IT"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-07T14:32:03Z","title_canon_sha256":"534287d34ac3e2be0a4d424a5cc50574175c6f09bb908340ebfb832e92257aa1"},"schema_version":"1.0","source":{"id":"2410.05078","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.05078","created_at":"2026-07-05T11:08:01Z"},{"alias_kind":"arxiv_version","alias_value":"2410.05078v2","created_at":"2026-07-05T11:08:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.05078","created_at":"2026-07-05T11:08:01Z"},{"alias_kind":"pith_short_12","alias_value":"OC4PEHPWOBCH","created_at":"2026-07-05T11:08:01Z"},{"alias_kind":"pith_short_16","alias_value":"OC4PEHPWOBCH66J3","created_at":"2026-07-05T11:08:01Z"},{"alias_kind":"pith_short_8","alias_value":"OC4PEHPW","created_at":"2026-07-05T11:08:01Z"}],"graph_snapshots":[{"event_id":"sha256:659462ee16f6df973d190e587fd2aafc3b408c4495715f38d28018541a8d3de0","target":"graph","created_at":"2026-07-05T11:08:01Z","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/2410.05078/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Foundation models are strong data compressors, but when accounting for their parameter size, their compression ratios are inferior to standard compression algorithms. Naively reducing the parameter count does not necessarily help as it deteriorates predictions and, accordingly, compression. We conduct a large-scale empirical study to find a sweet spot where pre-trained vanilla transformers can achieve competitive compression ratios. To this end, we train models on 165GB of raw byte sequences of either text, image, or audio data (and all possible combinations of the three) and then compress 1GB","authors_text":"Anian Ruoss, David Heurtel-Depeiges, Joel Veness, Tim Genewein","cross_cats":["cs.AI","cs.IT","math.IT"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-07T14:32:03Z","title":"Compression via Pre-trained Transformers: A Study on Byte-Level Multimodal Data"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.05078","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:a4523e6e4d0d993188c0cd4a121940dbfb0416bd0a1646dc067959f561c3b5c0","target":"record","created_at":"2026-07-05T11:08:01Z","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":"c2f636ee77d77ff6612a6deb5c3f8c244c3f48fd5d267ac86778507e50033e5c","cross_cats_sorted":["cs.AI","cs.IT","math.IT"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-07T14:32:03Z","title_canon_sha256":"534287d34ac3e2be0a4d424a5cc50574175c6f09bb908340ebfb832e92257aa1"},"schema_version":"1.0","source":{"id":"2410.05078","kind":"arxiv","version":2}},"canonical_sha256":"70b8f21df670447f793b613d792f195ee32fe0bb0c78099110a7c7f2a0c3b3ce","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"70b8f21df670447f793b613d792f195ee32fe0bb0c78099110a7c7f2a0c3b3ce","first_computed_at":"2026-07-05T11:08:01.528011Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:08:01.528011Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"os1bvFEFP3cotYMJ2FpeP52/EYApkCi7sdQpufjtPz7Tt3IikrSOV2QlAmmTJ6lT/Dvl3gh/W/Juki1VkhY1DQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:08:01.528682Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.05078","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a4523e6e4d0d993188c0cd4a121940dbfb0416bd0a1646dc067959f561c3b5c0","sha256:659462ee16f6df973d190e587fd2aafc3b408c4495715f38d28018541a8d3de0"],"state_sha256":"0d47a6e875250a53f10ce6f1678508e5fc23aee65b586efd1ea7d150d758b094"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Pp1ph2ATPpTx+gEQeAYuIy/Y6ClAyLue8d/AipIUYTGAqo3ZgqbH7afQfOIqCjLbfeD2m43Lap+ZoHiP6DxqCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-28T08:14:07.819792Z","bundle_sha256":"4e41b24287ea8bc549ec292a6fb30404693bf23ba28f71d373e3a44c54704bd6"}}