{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:JNCKSCVRQUIZRDLKLYJ4Q2TAM7","short_pith_number":"pith:JNCKSCVR","canonical_record":{"source":{"id":"2605.12928","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T03:03:30Z","cross_cats_sorted":[],"title_canon_sha256":"8b768de3edb5f072d26865a3fd76fd1ed1a7cfcd9abac425c11eadd03e3c719d","abstract_canon_sha256":"d420a61b1fa40c89431cc362687a93720712c43b755aba16ae7b670fe245945f"},"schema_version":"1.0"},"canonical_sha256":"4b44a90ab18511988d6a5e13c86a6067cb7cb4711e9a45a747f462dfc8dd4538","source":{"kind":"arxiv","id":"2605.12928","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12928","created_at":"2026-05-18T03:09:10Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12928v1","created_at":"2026-05-18T03:09:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12928","created_at":"2026-05-18T03:09:10Z"},{"alias_kind":"pith_short_12","alias_value":"JNCKSCVRQUIZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"JNCKSCVRQUIZRDLK","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"JNCKSCVR","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:JNCKSCVRQUIZRDLKLYJ4Q2TAM7","target":"record","payload":{"canonical_record":{"source":{"id":"2605.12928","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T03:03:30Z","cross_cats_sorted":[],"title_canon_sha256":"8b768de3edb5f072d26865a3fd76fd1ed1a7cfcd9abac425c11eadd03e3c719d","abstract_canon_sha256":"d420a61b1fa40c89431cc362687a93720712c43b755aba16ae7b670fe245945f"},"schema_version":"1.0"},"canonical_sha256":"4b44a90ab18511988d6a5e13c86a6067cb7cb4711e9a45a747f462dfc8dd4538","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:09:10.006664Z","signature_b64":"ntTZnWP/huztNBg3wQ0MmdRjUiMRfrlu2yXHXbnm4Iyu31amjma7r0/I8zrlKu7rZgBPL7OgftmUgPPWo5KHCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4b44a90ab18511988d6a5e13c86a6067cb7cb4711e9a45a747f462dfc8dd4538","last_reissued_at":"2026-05-18T03:09:10.006060Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:09:10.006060Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.12928","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MbxwdsEla2ybS1bMQ1xN/ylHWrP6i5UuND1At47R0LyfZs8KT5nbHGaX8+q3r87pTYVA7X3aUBlb/ONqTzOLAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T11:33:10.687760Z"},"content_sha256":"097c8f7b770e9f6e9445170610d9b2ffb5ea23a6310f2a5dcd77500f94bbcadd","schema_version":"1.0","event_id":"sha256:097c8f7b770e9f6e9445170610d9b2ffb5ea23a6310f2a5dcd77500f94bbcadd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:JNCKSCVRQUIZRDLKLYJ4Q2TAM7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Efficiency Gap in Byte Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Byte modeling incurs a larger scaling penalty under masked diffusion than under autoregressive training because diffusion destroys local byte contiguity.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexander M. Rush, Celine Lee, Chen Liang, Derek Cheng, Ed Chi, Fernando Pereira, Jeremiah Liu, Jiaxin Shi, Jing Nathan Yan, Pengcheng Yin, Ruoxi Wang, Yin Zhang","submitted_at":"2026-05-13T03:03:30Z","abstract_excerpt":"Modern language models have historically relied on two dominant design choices: subword tokenization and autoregressive (AR) ordering. These design decisions bake in priors that dictate a model's learning. Recently, two alternative paradigms have challenged this: byte-level modeling, which bypasses static statistically-derived token vocabularies, and masked diffusion modeling (MDM), which conducts parallel, non-sequential generation. Their intersection represents a fully end-to-end modality-agnostic generative prototype; however, removing these structural priors incurs a significant computatio"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"the performance penalty of byte modeling is not uniform; across scale, the scaling overhead of byte modeling is worse for MDM than for AR. We hypothesize that this disparity stems from context fragility: while AR's stable causal history allows models to naturally rediscover subword patterns, the MDM objective destroys the local contiguity required to efficiently resolve semantics from raw bytes.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"that the observed scaling disparity is caused by context fragility in MDM rather than differences in how compute is allocated or other unmeasured factors in the experimental setup.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Byte modeling incurs greater scaling overhead for masked diffusion than autoregressive models because the diffusion objective destroys local byte contiguity needed to resolve semantics.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Byte modeling incurs a larger scaling penalty under masked diffusion than under autoregressive training because diffusion destroys local byte contiguity.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"91b77bcb96316b90d3fb483bee7bd37220fbc0f4c4b31f257ec78b0cf6f135af"},"source":{"id":"2605.12928","kind":"arxiv","version":1},"verdict":{"id":"270adfc7-ae15-4a01-a746-3aaba57b3726","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T19:40:05.302095Z","strongest_claim":"the performance penalty of byte modeling is not uniform; across scale, the scaling overhead of byte modeling is worse for MDM than for AR. We hypothesize that this disparity stems from context fragility: while AR's stable causal history allows models to naturally rediscover subword patterns, the MDM objective destroys the local contiguity required to efficiently resolve semantics from raw bytes.","one_line_summary":"Byte modeling incurs greater scaling overhead for masked diffusion than autoregressive models because the diffusion objective destroys local byte contiguity needed to resolve semantics.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"that the observed scaling disparity is caused by context fragility in MDM rather than differences in how compute is allocated or other unmeasured factors in the experimental setup.","pith_extraction_headline":"Byte modeling incurs a larger scaling penalty under masked diffusion than under autoregressive training because diffusion destroys local byte contiguity."},"references":{"count":51,"sample":[{"doi":"","year":2025,"title":"Adapters for Altering","work_id":"e8429366-c015-4c81-b8b7-42fc056c1caf","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Arnaud Pannatier and Evann Courdier and François Fleuret , year=. 2404.09562 , archivePrefix=","work_id":"98f61b5e-922e-40ff-b51f-f75fe5c9b44a","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Shkarin, D. A. , title =. Problems of Information Transmission , year =","work_id":"28e15183-5ccc-4733-9f87-281db91d5733","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"2024 , howpublished =","work_id":"f88ce759-208c-4f98-8571-9cdded5ee014","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Kucherawy , title =","work_id":"78597206-9893-4182-abb6-fb475f931ef5","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":51,"snapshot_sha256":"1155c111f33740be3ecb76eeed05967e4c78ad545e13729368ada515c0e372da","internal_anchors":3},"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":"270adfc7-ae15-4a01-a746-3aaba57b3726"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OK/qBmNA9gDkAwptU8UiEDYEjTT5nZ9OoM0G02tjqiVAXaafWZKZNmuzBAbwvpA72erPwDbwh636kr7w2q47CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T11:33:10.688310Z"},"content_sha256":"7dcfb91cf487e1f576d887e749d7e83e42962eeed469069ac1b3163b0cc7039c","schema_version":"1.0","event_id":"sha256:7dcfb91cf487e1f576d887e749d7e83e42962eeed469069ac1b3163b0cc7039c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JNCKSCVRQUIZRDLKLYJ4Q2TAM7/bundle.json","state_url":"https://pith.science/pith/JNCKSCVRQUIZRDLKLYJ4Q2TAM7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JNCKSCVRQUIZRDLKLYJ4Q2TAM7/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":"2026-06-04T11:33:10Z","links":{"resolver":"https://pith.science/pith/JNCKSCVRQUIZRDLKLYJ4Q2TAM7","bundle":"https://pith.science/pith/JNCKSCVRQUIZRDLKLYJ4Q2TAM7/bundle.json","state":"https://pith.science/pith/JNCKSCVRQUIZRDLKLYJ4Q2TAM7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JNCKSCVRQUIZRDLKLYJ4Q2TAM7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:JNCKSCVRQUIZRDLKLYJ4Q2TAM7","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":"d420a61b1fa40c89431cc362687a93720712c43b755aba16ae7b670fe245945f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T03:03:30Z","title_canon_sha256":"8b768de3edb5f072d26865a3fd76fd1ed1a7cfcd9abac425c11eadd03e3c719d"},"schema_version":"1.0","source":{"id":"2605.12928","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12928","created_at":"2026-05-18T03:09:10Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12928v1","created_at":"2026-05-18T03:09:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12928","created_at":"2026-05-18T03:09:10Z"},{"alias_kind":"pith_short_12","alias_value":"JNCKSCVRQUIZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"JNCKSCVRQUIZRDLK","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"JNCKSCVR","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:7dcfb91cf487e1f576d887e749d7e83e42962eeed469069ac1b3163b0cc7039c","target":"graph","created_at":"2026-05-18T03:09:10Z","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":"the performance penalty of byte modeling is not uniform; across scale, the scaling overhead of byte modeling is worse for MDM than for AR. We hypothesize that this disparity stems from context fragility: while AR's stable causal history allows models to naturally rediscover subword patterns, the MDM objective destroys the local contiguity required to efficiently resolve semantics from raw bytes."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"that the observed scaling disparity is caused by context fragility in MDM rather than differences in how compute is allocated or other unmeasured factors in the experimental setup."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Byte modeling incurs greater scaling overhead for masked diffusion than autoregressive models because the diffusion objective destroys local byte contiguity needed to resolve semantics."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Byte modeling incurs a larger scaling penalty under masked diffusion than under autoregressive training because diffusion destroys local byte contiguity."}],"snapshot_sha256":"91b77bcb96316b90d3fb483bee7bd37220fbc0f4c4b31f257ec78b0cf6f135af"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Modern language models have historically relied on two dominant design choices: subword tokenization and autoregressive (AR) ordering. These design decisions bake in priors that dictate a model's learning. Recently, two alternative paradigms have challenged this: byte-level modeling, which bypasses static statistically-derived token vocabularies, and masked diffusion modeling (MDM), which conducts parallel, non-sequential generation. Their intersection represents a fully end-to-end modality-agnostic generative prototype; however, removing these structural priors incurs a significant computatio","authors_text":"Alexander M. Rush, Celine Lee, Chen Liang, Derek Cheng, Ed Chi, Fernando Pereira, Jeremiah Liu, Jiaxin Shi, Jing Nathan Yan, Pengcheng Yin, Ruoxi Wang, Yin Zhang","cross_cats":[],"headline":"Byte modeling incurs a larger scaling penalty under masked diffusion than under autoregressive training because diffusion destroys local byte contiguity.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T03:03:30Z","title":"The Efficiency Gap in Byte Modeling"},"references":{"count":51,"internal_anchors":3,"resolved_work":51,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Adapters for Altering","work_id":"e8429366-c015-4c81-b8b7-42fc056c1caf","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Arnaud Pannatier and Evann Courdier and François Fleuret , year=. 2404.09562 , archivePrefix=","work_id":"98f61b5e-922e-40ff-b51f-f75fe5c9b44a","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Shkarin, D. A. , title =. Problems of Information Transmission , year =","work_id":"28e15183-5ccc-4733-9f87-281db91d5733","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"2024 , howpublished =","work_id":"f88ce759-208c-4f98-8571-9cdded5ee014","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Kucherawy , title =","work_id":"78597206-9893-4182-abb6-fb475f931ef5","year":2021}],"snapshot_sha256":"1155c111f33740be3ecb76eeed05967e4c78ad545e13729368ada515c0e372da"},"source":{"id":"2605.12928","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T19:40:05.302095Z","id":"270adfc7-ae15-4a01-a746-3aaba57b3726","model_set":{"reader":"grok-4.3"},"one_line_summary":"Byte modeling incurs greater scaling overhead for masked diffusion than autoregressive models because the diffusion objective destroys local byte contiguity needed to resolve semantics.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Byte modeling incurs a larger scaling penalty under masked diffusion than under autoregressive training because diffusion destroys local byte contiguity.","strongest_claim":"the performance penalty of byte modeling is not uniform; across scale, the scaling overhead of byte modeling is worse for MDM than for AR. We hypothesize that this disparity stems from context fragility: while AR's stable causal history allows models to naturally rediscover subword patterns, the MDM objective destroys the local contiguity required to efficiently resolve semantics from raw bytes.","weakest_assumption":"that the observed scaling disparity is caused by context fragility in MDM rather than differences in how compute is allocated or other unmeasured factors in the experimental setup."}},"verdict_id":"270adfc7-ae15-4a01-a746-3aaba57b3726"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:097c8f7b770e9f6e9445170610d9b2ffb5ea23a6310f2a5dcd77500f94bbcadd","target":"record","created_at":"2026-05-18T03:09:10Z","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":"d420a61b1fa40c89431cc362687a93720712c43b755aba16ae7b670fe245945f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T03:03:30Z","title_canon_sha256":"8b768de3edb5f072d26865a3fd76fd1ed1a7cfcd9abac425c11eadd03e3c719d"},"schema_version":"1.0","source":{"id":"2605.12928","kind":"arxiv","version":1}},"canonical_sha256":"4b44a90ab18511988d6a5e13c86a6067cb7cb4711e9a45a747f462dfc8dd4538","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4b44a90ab18511988d6a5e13c86a6067cb7cb4711e9a45a747f462dfc8dd4538","first_computed_at":"2026-05-18T03:09:10.006060Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:09:10.006060Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ntTZnWP/huztNBg3wQ0MmdRjUiMRfrlu2yXHXbnm4Iyu31amjma7r0/I8zrlKu7rZgBPL7OgftmUgPPWo5KHCQ==","signature_status":"signed_v1","signed_at":"2026-05-18T03:09:10.006664Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.12928","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:097c8f7b770e9f6e9445170610d9b2ffb5ea23a6310f2a5dcd77500f94bbcadd","sha256:7dcfb91cf487e1f576d887e749d7e83e42962eeed469069ac1b3163b0cc7039c"],"state_sha256":"041b26f28081d2a13156c91797a8edf00154c1c01341b43125cdc3bfec8a4c40"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"r75k8B1vk0866ECdQS3kV4NWx+oDLG3exeb0qWugnPwrJcnt1Mwz/Q620/1kpsk7q/k7oBaJPDDmTg6/V3RJDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-04T11:33:10.690806Z","bundle_sha256":"62811d7392bcbc462b10fc12e9fe68d38aa02abf14f7c93f4790d587473999e1"}}