{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:4W6QVH2CFHZDB27IHV7WBAR3RS","short_pith_number":"pith:4W6QVH2C","canonical_record":{"source":{"id":"2607.25157","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-28T00:05:43Z","cross_cats_sorted":[],"title_canon_sha256":"d1b863ad9031651236f3f547bacf6098923f796750eda612b84244d6d296d37a","abstract_canon_sha256":"199a22f23101e1e6f8be1e2fcf616bc857c42f5ceadddf1967da89ed15ecc307"},"schema_version":"1.0"},"canonical_sha256":"e5bd0a9f4229f230ebe83d7f60823b8c8ec8e4c64aeae632ea0aef07177ce287","source":{"kind":"arxiv","id":"2607.25157","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.25157","created_at":"2026-07-29T00:25:05Z"},{"alias_kind":"arxiv_version","alias_value":"2607.25157v1","created_at":"2026-07-29T00:25:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.25157","created_at":"2026-07-29T00:25:05Z"},{"alias_kind":"pith_short_12","alias_value":"4W6QVH2CFHZD","created_at":"2026-07-29T00:25:05Z"},{"alias_kind":"pith_short_16","alias_value":"4W6QVH2CFHZDB27I","created_at":"2026-07-29T00:25:05Z"},{"alias_kind":"pith_short_8","alias_value":"4W6QVH2C","created_at":"2026-07-29T00:25:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:4W6QVH2CFHZDB27IHV7WBAR3RS","target":"record","payload":{"canonical_record":{"source":{"id":"2607.25157","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-28T00:05:43Z","cross_cats_sorted":[],"title_canon_sha256":"d1b863ad9031651236f3f547bacf6098923f796750eda612b84244d6d296d37a","abstract_canon_sha256":"199a22f23101e1e6f8be1e2fcf616bc857c42f5ceadddf1967da89ed15ecc307"},"schema_version":"1.0"},"canonical_sha256":"e5bd0a9f4229f230ebe83d7f60823b8c8ec8e4c64aeae632ea0aef07177ce287","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-29T00:25:05.929985Z","signature_b64":"xnyNG2hPZXklUgNxgclm2SDwTbjVK6JTCEryoHVPbrIBB2z/C5b+UrrYIuqk5BhdmTc5YIX25IW/WjXzLOkbAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e5bd0a9f4229f230ebe83d7f60823b8c8ec8e4c64aeae632ea0aef07177ce287","last_reissued_at":"2026-07-29T00:25:05.929093Z","signature_status":"signed_v1","first_computed_at":"2026-07-29T00:25:05.929093Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.25157","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":"2026-07-29T00:25:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3mwXNrGulQxQN/yKGX69W3Xkh3QxMedGGRsw4exn9KmDjtJA1lBpbIWrpLEcyVubvUK8wc3nDLLx5EeqYieHBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T12:57:47.643126Z"},"content_sha256":"805675cfe79b2713ba9cd97d7c556e915e379ffc05a38cc041b525bfb0f8fe8a","schema_version":"1.0","event_id":"sha256:805675cfe79b2713ba9cd97d7c556e915e379ffc05a38cc041b525bfb0f8fe8a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:4W6QVH2CFHZDB27IHV7WBAR3RS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PreDiff-LM: Pretrained Discrete Masked Diffusion Language Modeling with Hybrid Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chenhao Wei, Chenqian Le, Guang Yang, Haoyan Xu, Jesson Wang, Jiayi Cheng, Junhao Dong, Michael Yue, Runhao Li, Siheng Wang, Xupeng Chen, Yanfang Ye, Youran Shen, Zhengqing Yuan, Zhengtao Yao","submitted_at":"2026-07-28T00:05:43Z","abstract_excerpt":"Discrete masked diffusion language models support bidirectional generation and infilling, but adapting pretrained autoregressive (AR) transformers requires reconciling causal pretraining with bidirectional denoising. We study this problem at the level of attention rather than claiming AR-weight reuse itself as novel. PreDiff-LM preserves causal attention within the observed prompt while allowing full bidirectional attention within the masked target. Under a matched GPT-2 Medium, WikiText-103, 90K-step setup, this hybrid mask improves unconditional perplexity from 34.1 to 28.7 and MAUVE from 0."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.25157","kind":"arxiv","version":1},"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/2607.25157/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":"2026-07-29T00:25:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UaERMJCYlA/vSGjM7IwUuwW/5zmPtmlunAWU94OxhlCP/KtznAjk4wq57qzHgI/O5hM2RmgD4tNS7KCjNGuVBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T12:57:47.643690Z"},"content_sha256":"5fa19048c944e6b95f93902fa5c473624a96d98c03eda08d870cf944ec56dcb8","schema_version":"1.0","event_id":"sha256:5fa19048c944e6b95f93902fa5c473624a96d98c03eda08d870cf944ec56dcb8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4W6QVH2CFHZDB27IHV7WBAR3RS/bundle.json","state_url":"https://pith.science/pith/4W6QVH2CFHZDB27IHV7WBAR3RS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4W6QVH2CFHZDB27IHV7WBAR3RS/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":"2026-08-07T12:57:47Z","links":{"resolver":"https://pith.science/pith/4W6QVH2CFHZDB27IHV7WBAR3RS","bundle":"https://pith.science/pith/4W6QVH2CFHZDB27IHV7WBAR3RS/bundle.json","state":"https://pith.science/pith/4W6QVH2CFHZDB27IHV7WBAR3RS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4W6QVH2CFHZDB27IHV7WBAR3RS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4W6QVH2CFHZDB27IHV7WBAR3RS","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":"199a22f23101e1e6f8be1e2fcf616bc857c42f5ceadddf1967da89ed15ecc307","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-28T00:05:43Z","title_canon_sha256":"d1b863ad9031651236f3f547bacf6098923f796750eda612b84244d6d296d37a"},"schema_version":"1.0","source":{"id":"2607.25157","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.25157","created_at":"2026-07-29T00:25:05Z"},{"alias_kind":"arxiv_version","alias_value":"2607.25157v1","created_at":"2026-07-29T00:25:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.25157","created_at":"2026-07-29T00:25:05Z"},{"alias_kind":"pith_short_12","alias_value":"4W6QVH2CFHZD","created_at":"2026-07-29T00:25:05Z"},{"alias_kind":"pith_short_16","alias_value":"4W6QVH2CFHZDB27I","created_at":"2026-07-29T00:25:05Z"},{"alias_kind":"pith_short_8","alias_value":"4W6QVH2C","created_at":"2026-07-29T00:25:05Z"}],"graph_snapshots":[{"event_id":"sha256:5fa19048c944e6b95f93902fa5c473624a96d98c03eda08d870cf944ec56dcb8","target":"graph","created_at":"2026-07-29T00:25:05Z","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/2607.25157/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Discrete masked diffusion language models support bidirectional generation and infilling, but adapting pretrained autoregressive (AR) transformers requires reconciling causal pretraining with bidirectional denoising. We study this problem at the level of attention rather than claiming AR-weight reuse itself as novel. PreDiff-LM preserves causal attention within the observed prompt while allowing full bidirectional attention within the masked target. Under a matched GPT-2 Medium, WikiText-103, 90K-step setup, this hybrid mask improves unconditional perplexity from 34.1 to 28.7 and MAUVE from 0.","authors_text":"Chenhao Wei, Chenqian Le, Guang Yang, Haoyan Xu, Jesson Wang, Jiayi Cheng, Junhao Dong, Michael Yue, Runhao Li, Siheng Wang, Xupeng Chen, Yanfang Ye, Youran Shen, Zhengqing Yuan, Zhengtao Yao","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-28T00:05:43Z","title":"PreDiff-LM: Pretrained Discrete Masked Diffusion Language Modeling with Hybrid Attention"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.25157","kind":"arxiv","version":1},"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":"sha256:805675cfe79b2713ba9cd97d7c556e915e379ffc05a38cc041b525bfb0f8fe8a","target":"record","created_at":"2026-07-29T00:25:05Z","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":"199a22f23101e1e6f8be1e2fcf616bc857c42f5ceadddf1967da89ed15ecc307","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-28T00:05:43Z","title_canon_sha256":"d1b863ad9031651236f3f547bacf6098923f796750eda612b84244d6d296d37a"},"schema_version":"1.0","source":{"id":"2607.25157","kind":"arxiv","version":1}},"canonical_sha256":"e5bd0a9f4229f230ebe83d7f60823b8c8ec8e4c64aeae632ea0aef07177ce287","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e5bd0a9f4229f230ebe83d7f60823b8c8ec8e4c64aeae632ea0aef07177ce287","first_computed_at":"2026-07-29T00:25:05.929093Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-29T00:25:05.929093Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xnyNG2hPZXklUgNxgclm2SDwTbjVK6JTCEryoHVPbrIBB2z/C5b+UrrYIuqk5BhdmTc5YIX25IW/WjXzLOkbAg==","signature_status":"signed_v1","signed_at":"2026-07-29T00:25:05.929985Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.25157","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:805675cfe79b2713ba9cd97d7c556e915e379ffc05a38cc041b525bfb0f8fe8a","sha256:5fa19048c944e6b95f93902fa5c473624a96d98c03eda08d870cf944ec56dcb8"],"state_sha256":"9c3d45f0db9d96ea326d0379b38f299a3e86dde92b8794102200a7c7bf5819fc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jDO9peTI0PuVZ32qyTlqXzP+JSdyyJCVDqWT6OqQ6mkKAPqO6IdSNCaU0lbP0Y0RwlV6NcVi6ECxklucWUDEDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T12:57:47.647728Z","bundle_sha256":"d0df3995f376816b7d0f1d0d009a0b0d6c9b7bb71e96981585fdf446afd03a5e"}}