{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:HKV5YEH6PZGTBI47WWSRQCBRTN","short_pith_number":"pith:HKV5YEH6","canonical_record":{"source":{"id":"2406.11385","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-17T10:12:45Z","cross_cats_sorted":[],"title_canon_sha256":"fade798f044fe65bec3dc82edfd5e709025efcc212bbe005e87e0d7833fce80a","abstract_canon_sha256":"dd05a7e05c07cdd1d349807962a541093ec6446b0ee888c434baed5268dac626"},"schema_version":"1.0"},"canonical_sha256":"3aabdc10fe7e4d30a39fb5a51808319b5b351bbf3633727c78992aa0f190878c","source":{"kind":"arxiv","id":"2406.11385","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.11385","created_at":"2026-07-05T08:37:20Z"},{"alias_kind":"arxiv_version","alias_value":"2406.11385v2","created_at":"2026-07-05T08:37:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.11385","created_at":"2026-07-05T08:37:20Z"},{"alias_kind":"pith_short_12","alias_value":"HKV5YEH6PZGT","created_at":"2026-07-05T08:37:20Z"},{"alias_kind":"pith_short_16","alias_value":"HKV5YEH6PZGTBI47","created_at":"2026-07-05T08:37:20Z"},{"alias_kind":"pith_short_8","alias_value":"HKV5YEH6","created_at":"2026-07-05T08:37:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:HKV5YEH6PZGTBI47WWSRQCBRTN","target":"record","payload":{"canonical_record":{"source":{"id":"2406.11385","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-17T10:12:45Z","cross_cats_sorted":[],"title_canon_sha256":"fade798f044fe65bec3dc82edfd5e709025efcc212bbe005e87e0d7833fce80a","abstract_canon_sha256":"dd05a7e05c07cdd1d349807962a541093ec6446b0ee888c434baed5268dac626"},"schema_version":"1.0"},"canonical_sha256":"3aabdc10fe7e4d30a39fb5a51808319b5b351bbf3633727c78992aa0f190878c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:37:20.551829Z","signature_b64":"bxB3NQgbk8C+7yQ00VKssV/Wv8eQ+2qWiAoNsokzWvrlj2iM9JIzJeq5by6aREQFvz6DR14PcV0JYNWSLAc4Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3aabdc10fe7e4d30a39fb5a51808319b5b351bbf3633727c78992aa0f190878c","last_reissued_at":"2026-07-05T08:37:20.551416Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:37:20.551416Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.11385","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":"2026-07-05T08:37:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PsmKf23jDOWlak0whxL3qKTTCm+95Qo8ewMJiodXDmm23aqDIIExoyJClcC9XbaVx0eAFaa8QwvpSsMqHoMDCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-24T01:40:01.981312Z"},"content_sha256":"950162d60eea4df48451767bf0fea9e39dcc80da81cb72795a3e853955792855","schema_version":"1.0","event_id":"sha256:950162d60eea4df48451767bf0fea9e39dcc80da81cb72795a3e853955792855"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:HKV5YEH6PZGTBI47WWSRQCBRTN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MetaGPT: Merging Large Language Models Using Model Exclusive Task Arithmetic","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bingning Wang, Liang Song, Weipeng Chen, Yuyan Zhou","submitted_at":"2024-06-17T10:12:45Z","abstract_excerpt":"The advent of large language models (LLMs) like GPT-4 has catalyzed the exploration of multi-task learning (MTL), in which a single model demonstrates proficiency across diverse tasks. Task arithmetic has emerged as a cost-effective approach for MTL. It enables performance enhancement across multiple tasks by adding their corresponding task vectors to a pre-trained model. However, the current lack of a method that can simultaneously achieve optimal performance, computational efficiency, and data privacy limits their application to LLMs. In this paper, we propose \\textbf{M}odel \\textbf{E}xclusi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.11385","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/2406.11385/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":"2026-07-05T08:37:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MrGkCLDyJGhPfbh+6AvaNFr2+Em6aiCb/qKkPuj16IbfKi81ViifR9RvNekpPUGJCCH+3ZfKfo7owUte+ZCdDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-24T01:40:01.981841Z"},"content_sha256":"cc9e65ab4b7f8406394b7bab5a207083a6fe5c4d5159d190b6220943c5ea3e88","schema_version":"1.0","event_id":"sha256:cc9e65ab4b7f8406394b7bab5a207083a6fe5c4d5159d190b6220943c5ea3e88"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HKV5YEH6PZGTBI47WWSRQCBRTN/bundle.json","state_url":"https://pith.science/pith/HKV5YEH6PZGTBI47WWSRQCBRTN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HKV5YEH6PZGTBI47WWSRQCBRTN/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":"2026-08-24T01:40:01Z","links":{"resolver":"https://pith.science/pith/HKV5YEH6PZGTBI47WWSRQCBRTN","bundle":"https://pith.science/pith/HKV5YEH6PZGTBI47WWSRQCBRTN/bundle.json","state":"https://pith.science/pith/HKV5YEH6PZGTBI47WWSRQCBRTN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HKV5YEH6PZGTBI47WWSRQCBRTN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:HKV5YEH6PZGTBI47WWSRQCBRTN","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":"dd05a7e05c07cdd1d349807962a541093ec6446b0ee888c434baed5268dac626","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-17T10:12:45Z","title_canon_sha256":"fade798f044fe65bec3dc82edfd5e709025efcc212bbe005e87e0d7833fce80a"},"schema_version":"1.0","source":{"id":"2406.11385","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.11385","created_at":"2026-07-05T08:37:20Z"},{"alias_kind":"arxiv_version","alias_value":"2406.11385v2","created_at":"2026-07-05T08:37:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.11385","created_at":"2026-07-05T08:37:20Z"},{"alias_kind":"pith_short_12","alias_value":"HKV5YEH6PZGT","created_at":"2026-07-05T08:37:20Z"},{"alias_kind":"pith_short_16","alias_value":"HKV5YEH6PZGTBI47","created_at":"2026-07-05T08:37:20Z"},{"alias_kind":"pith_short_8","alias_value":"HKV5YEH6","created_at":"2026-07-05T08:37:20Z"}],"graph_snapshots":[{"event_id":"sha256:cc9e65ab4b7f8406394b7bab5a207083a6fe5c4d5159d190b6220943c5ea3e88","target":"graph","created_at":"2026-07-05T08:37:20Z","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/2406.11385/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The advent of large language models (LLMs) like GPT-4 has catalyzed the exploration of multi-task learning (MTL), in which a single model demonstrates proficiency across diverse tasks. Task arithmetic has emerged as a cost-effective approach for MTL. It enables performance enhancement across multiple tasks by adding their corresponding task vectors to a pre-trained model. However, the current lack of a method that can simultaneously achieve optimal performance, computational efficiency, and data privacy limits their application to LLMs. In this paper, we propose \\textbf{M}odel \\textbf{E}xclusi","authors_text":"Bingning Wang, Liang Song, Weipeng Chen, Yuyan Zhou","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-17T10:12:45Z","title":"MetaGPT: Merging Large Language Models Using Model Exclusive Task Arithmetic"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.11385","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":"sha256:950162d60eea4df48451767bf0fea9e39dcc80da81cb72795a3e853955792855","target":"record","created_at":"2026-07-05T08:37:20Z","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":"dd05a7e05c07cdd1d349807962a541093ec6446b0ee888c434baed5268dac626","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-17T10:12:45Z","title_canon_sha256":"fade798f044fe65bec3dc82edfd5e709025efcc212bbe005e87e0d7833fce80a"},"schema_version":"1.0","source":{"id":"2406.11385","kind":"arxiv","version":2}},"canonical_sha256":"3aabdc10fe7e4d30a39fb5a51808319b5b351bbf3633727c78992aa0f190878c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3aabdc10fe7e4d30a39fb5a51808319b5b351bbf3633727c78992aa0f190878c","first_computed_at":"2026-07-05T08:37:20.551416Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:37:20.551416Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bxB3NQgbk8C+7yQ00VKssV/Wv8eQ+2qWiAoNsokzWvrlj2iM9JIzJeq5by6aREQFvz6DR14PcV0JYNWSLAc4Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:37:20.551829Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.11385","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:950162d60eea4df48451767bf0fea9e39dcc80da81cb72795a3e853955792855","sha256:cc9e65ab4b7f8406394b7bab5a207083a6fe5c4d5159d190b6220943c5ea3e88"],"state_sha256":"9c5f9b73bf8b50b02e3ec755419fe345a921cbea6c2a72f02ef675fcc7bce382"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+ZnCWmRjTr3cvpG6zoprH8wUNYK0aVhA8Ngup4bZfChZuL6drMJsfnlXqV86JMB9wHhlYxvY/n85c9ObBex0CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-24T01:40:01.985635Z","bundle_sha256":"4482b4218bffa03849e40041bf8a4fea88a10513c569aef9343cfa3bfbea20f8"}}