{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:XS2ZEGFMGWG6N4XVYTAJZN6CNM","short_pith_number":"pith:XS2ZEGFM","canonical_record":{"source":{"id":"2503.01496","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-03T13:08:00Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"027d2af6cf81a83864e3096994b0bd79e9ef305360f5a0acf50485a19b067235","abstract_canon_sha256":"ef60dfd78c8b6ca77c27b55d53b5ec2ea46c14517b8da307646946169311c528"},"schema_version":"1.0"},"canonical_sha256":"bcb59218ac358de6f2f5c4c09cb7c26b3f070bae58dc0df078a6261be41864f0","source":{"kind":"arxiv","id":"2503.01496","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.01496","created_at":"2026-07-05T10:59:27Z"},{"alias_kind":"arxiv_version","alias_value":"2503.01496v2","created_at":"2026-07-05T10:59:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.01496","created_at":"2026-07-05T10:59:27Z"},{"alias_kind":"pith_short_12","alias_value":"XS2ZEGFMGWG6","created_at":"2026-07-05T10:59:27Z"},{"alias_kind":"pith_short_16","alias_value":"XS2ZEGFMGWG6N4XV","created_at":"2026-07-05T10:59:27Z"},{"alias_kind":"pith_short_8","alias_value":"XS2ZEGFM","created_at":"2026-07-05T10:59:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:XS2ZEGFMGWG6N4XVYTAJZN6CNM","target":"record","payload":{"canonical_record":{"source":{"id":"2503.01496","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-03T13:08:00Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"027d2af6cf81a83864e3096994b0bd79e9ef305360f5a0acf50485a19b067235","abstract_canon_sha256":"ef60dfd78c8b6ca77c27b55d53b5ec2ea46c14517b8da307646946169311c528"},"schema_version":"1.0"},"canonical_sha256":"bcb59218ac358de6f2f5c4c09cb7c26b3f070bae58dc0df078a6261be41864f0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:59:27.610385Z","signature_b64":"TJElCex0nAGHdDr3AVhk9tc2uZqhQATRyS2Pg4D7U+zvdnrCepXk5/UnWgfCzaIsfJeTO+hANog67THD8RwLBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bcb59218ac358de6f2f5c4c09cb7c26b3f070bae58dc0df078a6261be41864f0","last_reissued_at":"2026-07-05T10:59:27.609846Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:59:27.609846Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.01496","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":"2026-07-05T10:59:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3Kyuet+WyY6QBEpYm66/TxoglNHww79haqY7Bqpp+Um/A4atTX1gWtap+V7XrmZOSMBBKmofYFxt7J3ATKSXDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T07:22:19.165640Z"},"content_sha256":"5384e6eb8f67ffe512bb20eb70d63fffeab9913c48f589bc0ecfa45119df7805","schema_version":"1.0","event_id":"sha256:5384e6eb8f67ffe512bb20eb70d63fffeab9913c48f589bc0ecfa45119df7805"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:XS2ZEGFMGWG6N4XVYTAJZN6CNM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Liger: Linearizing Large Language Models to Gated Recurrent Structures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Disen Lan, Jiaxi Hu, Jusen Du, Weigao Sun, Yu Cheng","submitted_at":"2025-03-03T13:08:00Z","abstract_excerpt":"Transformers with linear recurrent modeling offer linear-time training and constant-memory inference. Despite their demonstrated efficiency and performance, pretraining such non-standard architectures from scratch remains costly and risky. The linearization of large language models (LLMs) transforms pretrained standard models into linear recurrent structures, enabling more efficient deployment. However, current linearization methods typically introduce additional feature map modules that require extensive fine-tuning and overlook the gating mechanisms used in state-of-the-art linear recurrent "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.01496","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/2503.01496/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":"2026-07-05T10:59:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nVvGlib4qsgtl0lDx4EZ9CFDDoq2K9TpSoZ69CtWG9tUZYMTdUxPqqJUsVpnf4dK2td9C41GHPZXlzeWMgTmAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T07:22:19.166177Z"},"content_sha256":"073dd662a5b1fcfc44f133314ef43b586ff6c606b54eab713b7ea1324152a38f","schema_version":"1.0","event_id":"sha256:073dd662a5b1fcfc44f133314ef43b586ff6c606b54eab713b7ea1324152a38f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XS2ZEGFMGWG6N4XVYTAJZN6CNM/bundle.json","state_url":"https://pith.science/pith/XS2ZEGFMGWG6N4XVYTAJZN6CNM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XS2ZEGFMGWG6N4XVYTAJZN6CNM/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":"2026-08-22T07:22:19Z","links":{"resolver":"https://pith.science/pith/XS2ZEGFMGWG6N4XVYTAJZN6CNM","bundle":"https://pith.science/pith/XS2ZEGFMGWG6N4XVYTAJZN6CNM/bundle.json","state":"https://pith.science/pith/XS2ZEGFMGWG6N4XVYTAJZN6CNM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XS2ZEGFMGWG6N4XVYTAJZN6CNM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XS2ZEGFMGWG6N4XVYTAJZN6CNM","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":"ef60dfd78c8b6ca77c27b55d53b5ec2ea46c14517b8da307646946169311c528","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-03T13:08:00Z","title_canon_sha256":"027d2af6cf81a83864e3096994b0bd79e9ef305360f5a0acf50485a19b067235"},"schema_version":"1.0","source":{"id":"2503.01496","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.01496","created_at":"2026-07-05T10:59:27Z"},{"alias_kind":"arxiv_version","alias_value":"2503.01496v2","created_at":"2026-07-05T10:59:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.01496","created_at":"2026-07-05T10:59:27Z"},{"alias_kind":"pith_short_12","alias_value":"XS2ZEGFMGWG6","created_at":"2026-07-05T10:59:27Z"},{"alias_kind":"pith_short_16","alias_value":"XS2ZEGFMGWG6N4XV","created_at":"2026-07-05T10:59:27Z"},{"alias_kind":"pith_short_8","alias_value":"XS2ZEGFM","created_at":"2026-07-05T10:59:27Z"}],"graph_snapshots":[{"event_id":"sha256:073dd662a5b1fcfc44f133314ef43b586ff6c606b54eab713b7ea1324152a38f","target":"graph","created_at":"2026-07-05T10:59:27Z","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/2503.01496/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Transformers with linear recurrent modeling offer linear-time training and constant-memory inference. Despite their demonstrated efficiency and performance, pretraining such non-standard architectures from scratch remains costly and risky. The linearization of large language models (LLMs) transforms pretrained standard models into linear recurrent structures, enabling more efficient deployment. However, current linearization methods typically introduce additional feature map modules that require extensive fine-tuning and overlook the gating mechanisms used in state-of-the-art linear recurrent ","authors_text":"Disen Lan, Jiaxi Hu, Jusen Du, Weigao Sun, Yu Cheng","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-03T13:08:00Z","title":"Liger: Linearizing Large Language Models to Gated Recurrent Structures"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.01496","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":"sha256:5384e6eb8f67ffe512bb20eb70d63fffeab9913c48f589bc0ecfa45119df7805","target":"record","created_at":"2026-07-05T10:59:27Z","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":"ef60dfd78c8b6ca77c27b55d53b5ec2ea46c14517b8da307646946169311c528","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-03T13:08:00Z","title_canon_sha256":"027d2af6cf81a83864e3096994b0bd79e9ef305360f5a0acf50485a19b067235"},"schema_version":"1.0","source":{"id":"2503.01496","kind":"arxiv","version":2}},"canonical_sha256":"bcb59218ac358de6f2f5c4c09cb7c26b3f070bae58dc0df078a6261be41864f0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bcb59218ac358de6f2f5c4c09cb7c26b3f070bae58dc0df078a6261be41864f0","first_computed_at":"2026-07-05T10:59:27.609846Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:59:27.609846Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"TJElCex0nAGHdDr3AVhk9tc2uZqhQATRyS2Pg4D7U+zvdnrCepXk5/UnWgfCzaIsfJeTO+hANog67THD8RwLBw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:59:27.610385Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.01496","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5384e6eb8f67ffe512bb20eb70d63fffeab9913c48f589bc0ecfa45119df7805","sha256:073dd662a5b1fcfc44f133314ef43b586ff6c606b54eab713b7ea1324152a38f"],"state_sha256":"3e13a2949d27ce12c5aa9ba04c3af29edb0cbd744ae8aafbc736429577f7c9ff"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8N0Mf322kn0E+m6AHzRoD8a6DLKMSsuMpkQjzEcFyY9FL9cG8xZ1SUa62P+/XICvVu/IYPW2wri9R/5ohiI3Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-22T07:22:19.170447Z","bundle_sha256":"8fbe1b015b0361857a190e8d9c39a28c35f7deb1686e100e3bd5508f48996f7a"}}