{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:TJO6ZOTASS3GHLSCGV674U26V6","short_pith_number":"pith:TJO6ZOTA","canonical_record":{"source":{"id":"2404.08707","kind":"arxiv","version":7},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-11T17:44:56Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"4e0d4fd7922241c9cc713c4b9b70c74bf3e7454160e0cef1db2086e7f483d882","abstract_canon_sha256":"6d613bc6ba9615b941777c06bb9b239f0d0993d7861600f763206c51867a7c0d"},"schema_version":"1.0"},"canonical_sha256":"9a5decba6094b663ae42357dfe535eafa10f0dc709c3e11b2ff83beaffcc31c0","source":{"kind":"arxiv","id":"2404.08707","version":7},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.08707","created_at":"2026-07-05T09:49:42Z"},{"alias_kind":"arxiv_version","alias_value":"2404.08707v7","created_at":"2026-07-05T09:49:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.08707","created_at":"2026-07-05T09:49:42Z"},{"alias_kind":"pith_short_12","alias_value":"TJO6ZOTASS3G","created_at":"2026-07-05T09:49:42Z"},{"alias_kind":"pith_short_16","alias_value":"TJO6ZOTASS3GHLSC","created_at":"2026-07-05T09:49:42Z"},{"alias_kind":"pith_short_8","alias_value":"TJO6ZOTA","created_at":"2026-07-05T09:49:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:TJO6ZOTASS3GHLSCGV674U26V6","target":"record","payload":{"canonical_record":{"source":{"id":"2404.08707","kind":"arxiv","version":7},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-11T17:44:56Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"4e0d4fd7922241c9cc713c4b9b70c74bf3e7454160e0cef1db2086e7f483d882","abstract_canon_sha256":"6d613bc6ba9615b941777c06bb9b239f0d0993d7861600f763206c51867a7c0d"},"schema_version":"1.0"},"canonical_sha256":"9a5decba6094b663ae42357dfe535eafa10f0dc709c3e11b2ff83beaffcc31c0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:49:42.103057Z","signature_b64":"VB1wqsZcqQmlLQXPt3DBUa8WW1Lo4XQjPPUbtZMPccr1ABg8m3YXi/AoVerQ5ATwzei8iDzPs/SdFJHWO9G7BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9a5decba6094b663ae42357dfe535eafa10f0dc709c3e11b2ff83beaffcc31c0","last_reissued_at":"2026-07-05T09:49:42.102530Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:49:42.102530Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.08707","source_version":7,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:49:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"n/eNYBf6Mx68X+2bakPd/XOik/rDkV8oVuTq+yb0ydXKslceXjqJ9v2ey+kn4Y+kjcHr8J29jG/0L7+coSFfBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T05:14:16.460507Z"},"content_sha256":"9caa2e36e7de26be9b2afa7ed4414b7b3dead6f12048c6c4f427ad46c5c137cc","schema_version":"1.0","event_id":"sha256:9caa2e36e7de26be9b2afa7ed4414b7b3dead6f12048c6c4f427ad46c5c137cc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:TJO6ZOTASS3GHLSCGV674U26V6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CEM: A Data-Efficient Method for Large Language Models to Continue Evolving From Mistakes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Chengyu Du, Haixia Han, Haokun Zhao, Jiaqing Liang, Jie Shi, Yanghua Xiao","submitted_at":"2024-04-11T17:44:56Z","abstract_excerpt":"As world knowledge advances and new task schemas emerge, Continual Learning (CL) becomes essential for keeping Large Language Models (LLMs) current and addressing their shortcomings. This process typically involves continual instruction tuning (CIT) and continual pre-training (CPT) to enable these models to adapt to novel tasks and acquire critical knowledge. However, collecting sufficient CPT data and efficiently bridging knowledge gaps remain significant challenges. Inspired by the 'summarizing mistakes' strategy, we propose the Continue Evolving from Mistakes (CEM) method, a data-efficient "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.08707","kind":"arxiv","version":7},"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/2404.08707/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":"2026-07-05T09:49:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZqikMwZmKQt/OMv76LdEtI2AC4k2/eOz+cbEErnpxUZDUgwMtmuqMO/cQKE+WW05hsj/pSaKHHa9OOTqNPcbAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T05:14:16.461512Z"},"content_sha256":"9be05f455e0b30a8872ec41e3eccb9e9035cda01fc05918c254cb52e441e4d71","schema_version":"1.0","event_id":"sha256:9be05f455e0b30a8872ec41e3eccb9e9035cda01fc05918c254cb52e441e4d71"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TJO6ZOTASS3GHLSCGV674U26V6/bundle.json","state_url":"https://pith.science/pith/TJO6ZOTASS3GHLSCGV674U26V6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TJO6ZOTASS3GHLSCGV674U26V6/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":"2026-08-09T05:14:16Z","links":{"resolver":"https://pith.science/pith/TJO6ZOTASS3GHLSCGV674U26V6","bundle":"https://pith.science/pith/TJO6ZOTASS3GHLSCGV674U26V6/bundle.json","state":"https://pith.science/pith/TJO6ZOTASS3GHLSCGV674U26V6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TJO6ZOTASS3GHLSCGV674U26V6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:TJO6ZOTASS3GHLSCGV674U26V6","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":"6d613bc6ba9615b941777c06bb9b239f0d0993d7861600f763206c51867a7c0d","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-11T17:44:56Z","title_canon_sha256":"4e0d4fd7922241c9cc713c4b9b70c74bf3e7454160e0cef1db2086e7f483d882"},"schema_version":"1.0","source":{"id":"2404.08707","kind":"arxiv","version":7}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.08707","created_at":"2026-07-05T09:49:42Z"},{"alias_kind":"arxiv_version","alias_value":"2404.08707v7","created_at":"2026-07-05T09:49:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.08707","created_at":"2026-07-05T09:49:42Z"},{"alias_kind":"pith_short_12","alias_value":"TJO6ZOTASS3G","created_at":"2026-07-05T09:49:42Z"},{"alias_kind":"pith_short_16","alias_value":"TJO6ZOTASS3GHLSC","created_at":"2026-07-05T09:49:42Z"},{"alias_kind":"pith_short_8","alias_value":"TJO6ZOTA","created_at":"2026-07-05T09:49:42Z"}],"graph_snapshots":[{"event_id":"sha256:9be05f455e0b30a8872ec41e3eccb9e9035cda01fc05918c254cb52e441e4d71","target":"graph","created_at":"2026-07-05T09:49:42Z","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/2404.08707/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As world knowledge advances and new task schemas emerge, Continual Learning (CL) becomes essential for keeping Large Language Models (LLMs) current and addressing their shortcomings. This process typically involves continual instruction tuning (CIT) and continual pre-training (CPT) to enable these models to adapt to novel tasks and acquire critical knowledge. However, collecting sufficient CPT data and efficiently bridging knowledge gaps remain significant challenges. Inspired by the 'summarizing mistakes' strategy, we propose the Continue Evolving from Mistakes (CEM) method, a data-efficient ","authors_text":"Chengyu Du, Haixia Han, Haokun Zhao, Jiaqing Liang, Jie Shi, Yanghua Xiao","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-11T17:44:56Z","title":"CEM: A Data-Efficient Method for Large Language Models to Continue Evolving From Mistakes"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.08707","kind":"arxiv","version":7},"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":"sha256:9caa2e36e7de26be9b2afa7ed4414b7b3dead6f12048c6c4f427ad46c5c137cc","target":"record","created_at":"2026-07-05T09:49:42Z","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":"6d613bc6ba9615b941777c06bb9b239f0d0993d7861600f763206c51867a7c0d","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-11T17:44:56Z","title_canon_sha256":"4e0d4fd7922241c9cc713c4b9b70c74bf3e7454160e0cef1db2086e7f483d882"},"schema_version":"1.0","source":{"id":"2404.08707","kind":"arxiv","version":7}},"canonical_sha256":"9a5decba6094b663ae42357dfe535eafa10f0dc709c3e11b2ff83beaffcc31c0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9a5decba6094b663ae42357dfe535eafa10f0dc709c3e11b2ff83beaffcc31c0","first_computed_at":"2026-07-05T09:49:42.102530Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:49:42.102530Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VB1wqsZcqQmlLQXPt3DBUa8WW1Lo4XQjPPUbtZMPccr1ABg8m3YXi/AoVerQ5ATwzei8iDzPs/SdFJHWO9G7BA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:49:42.103057Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.08707","source_kind":"arxiv","source_version":7}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9caa2e36e7de26be9b2afa7ed4414b7b3dead6f12048c6c4f427ad46c5c137cc","sha256:9be05f455e0b30a8872ec41e3eccb9e9035cda01fc05918c254cb52e441e4d71"],"state_sha256":"d812c628e748febd88c436e17e58b7518818bfe07902b679a8645e36c91781bc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aR74OhvrfLgg6hrH5NvF9kNobj2cPluFPpWxmYp9UdU5MXpg9SdYow3b6pcAhkLbivWgM42MqRWpIxEPq7KLCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T05:14:16.467088Z","bundle_sha256":"322408429e7a4a52f02cf4eba519d8138127c29c9531bf9c60b57e3a5e07f2dd"}}