{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:KK5M6N4EU6PVPORAKIMNTLZJPV","short_pith_number":"pith:KK5M6N4E","canonical_record":{"source":{"id":"2605.20803","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T06:50:10Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"747c347bb89a8a609cc0d865e42319ba6d20ddd1f6e61704bd7bd5d86437457a","abstract_canon_sha256":"70f9ef49d556e733a7aeeab28f88955c6686fbe1ce7ed17ef488b4cc7ec565d2"},"schema_version":"1.0"},"canonical_sha256":"52bacf3784a79f57ba205218d9af297d6cfd0df34ba9383b6860d67fd07931d6","source":{"kind":"arxiv","id":"2605.20803","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.20803","created_at":"2026-05-21T01:04:55Z"},{"alias_kind":"arxiv_version","alias_value":"2605.20803v1","created_at":"2026-05-21T01:04:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.20803","created_at":"2026-05-21T01:04:55Z"},{"alias_kind":"pith_short_12","alias_value":"KK5M6N4EU6PV","created_at":"2026-05-21T01:04:55Z"},{"alias_kind":"pith_short_16","alias_value":"KK5M6N4EU6PVPORA","created_at":"2026-05-21T01:04:55Z"},{"alias_kind":"pith_short_8","alias_value":"KK5M6N4E","created_at":"2026-05-21T01:04:55Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:KK5M6N4EU6PVPORAKIMNTLZJPV","target":"record","payload":{"canonical_record":{"source":{"id":"2605.20803","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T06:50:10Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"747c347bb89a8a609cc0d865e42319ba6d20ddd1f6e61704bd7bd5d86437457a","abstract_canon_sha256":"70f9ef49d556e733a7aeeab28f88955c6686fbe1ce7ed17ef488b4cc7ec565d2"},"schema_version":"1.0"},"canonical_sha256":"52bacf3784a79f57ba205218d9af297d6cfd0df34ba9383b6860d67fd07931d6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:04:55.362721Z","signature_b64":"koe5+a2zHBEoRkcAR6E2KOQoFAihwXME7bRAMV06cQ1JjazOmRfvQ2YcxFvK4IT7kjBAUNP1+EpA3W4KPgl0AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"52bacf3784a79f57ba205218d9af297d6cfd0df34ba9383b6860d67fd07931d6","last_reissued_at":"2026-05-21T01:04:55.361812Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:04:55.361812Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.20803","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":"2026-05-21T01:04:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Pg9B2svgQPEvQze7ugxRq+WieB0zrPz3mmB9D4c4tMMyc1onyZbzteLaHokpr/8+66KQOXuXbAy6Gq3wN791Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-24T14:34:05.056721Z"},"content_sha256":"eb400743e066d9c5757243a22552b5b26d5e0eb6b4fdc594fe4a591b88e1ce55","schema_version":"1.0","event_id":"sha256:eb400743e066d9c5757243a22552b5b26d5e0eb6b4fdc594fe4a591b88e1ce55"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:KK5M6N4EU6PVPORAKIMNTLZJPV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Tunable MAGMAX: Preference-Aware Model Merging for Continual Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Kei Hiroshima, Kento Uchida, Shinichi Shirakawa","submitted_at":"2026-05-20T06:50:10Z","abstract_excerpt":"Continual learning (CL) aims to train models sequentially on multiple tasks while mitigating catastrophic forgetting of previously learned knowledge. Recent advances in large pre-trained models (LPMs) and model merging techniques, such as MAGMAX, have demonstrated effective CL performance by combining task-specific parameters. However, existing methods primarily focus on average performance across all tasks and do not adequately address how to construct models accommodating different deployment environments or varying user preferences. This paper proposes a model merging framework, termed Tuna"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.20803","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/2605.20803/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":"2026-05-21T01:04:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"47++toB6eqNsxqvWee1iUXdrtTyHh/87aebRj7SQYYZp6/YbN7q2TTMEJGUaRnUNeU//PXyzaodYU8DHCDYDCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-24T14:34:05.057112Z"},"content_sha256":"05dd6216a2a5eeb18336efda1c432e6f7f52397e0264c258c7b6578dafce1c8a","schema_version":"1.0","event_id":"sha256:05dd6216a2a5eeb18336efda1c432e6f7f52397e0264c258c7b6578dafce1c8a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KK5M6N4EU6PVPORAKIMNTLZJPV/bundle.json","state_url":"https://pith.science/pith/KK5M6N4EU6PVPORAKIMNTLZJPV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KK5M6N4EU6PVPORAKIMNTLZJPV/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":"2026-05-24T14:34:05Z","links":{"resolver":"https://pith.science/pith/KK5M6N4EU6PVPORAKIMNTLZJPV","bundle":"https://pith.science/pith/KK5M6N4EU6PVPORAKIMNTLZJPV/bundle.json","state":"https://pith.science/pith/KK5M6N4EU6PVPORAKIMNTLZJPV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KK5M6N4EU6PVPORAKIMNTLZJPV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:KK5M6N4EU6PVPORAKIMNTLZJPV","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":"70f9ef49d556e733a7aeeab28f88955c6686fbe1ce7ed17ef488b4cc7ec565d2","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T06:50:10Z","title_canon_sha256":"747c347bb89a8a609cc0d865e42319ba6d20ddd1f6e61704bd7bd5d86437457a"},"schema_version":"1.0","source":{"id":"2605.20803","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.20803","created_at":"2026-05-21T01:04:55Z"},{"alias_kind":"arxiv_version","alias_value":"2605.20803v1","created_at":"2026-05-21T01:04:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.20803","created_at":"2026-05-21T01:04:55Z"},{"alias_kind":"pith_short_12","alias_value":"KK5M6N4EU6PV","created_at":"2026-05-21T01:04:55Z"},{"alias_kind":"pith_short_16","alias_value":"KK5M6N4EU6PVPORA","created_at":"2026-05-21T01:04:55Z"},{"alias_kind":"pith_short_8","alias_value":"KK5M6N4E","created_at":"2026-05-21T01:04:55Z"}],"graph_snapshots":[{"event_id":"sha256:05dd6216a2a5eeb18336efda1c432e6f7f52397e0264c258c7b6578dafce1c8a","target":"graph","created_at":"2026-05-21T01:04:55Z","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/2605.20803/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Continual learning (CL) aims to train models sequentially on multiple tasks while mitigating catastrophic forgetting of previously learned knowledge. Recent advances in large pre-trained models (LPMs) and model merging techniques, such as MAGMAX, have demonstrated effective CL performance by combining task-specific parameters. However, existing methods primarily focus on average performance across all tasks and do not adequately address how to construct models accommodating different deployment environments or varying user preferences. This paper proposes a model merging framework, termed Tuna","authors_text":"Kei Hiroshima, Kento Uchida, Shinichi Shirakawa","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T06:50:10Z","title":"Tunable MAGMAX: Preference-Aware Model Merging for Continual Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.20803","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:eb400743e066d9c5757243a22552b5b26d5e0eb6b4fdc594fe4a591b88e1ce55","target":"record","created_at":"2026-05-21T01:04:55Z","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":"70f9ef49d556e733a7aeeab28f88955c6686fbe1ce7ed17ef488b4cc7ec565d2","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T06:50:10Z","title_canon_sha256":"747c347bb89a8a609cc0d865e42319ba6d20ddd1f6e61704bd7bd5d86437457a"},"schema_version":"1.0","source":{"id":"2605.20803","kind":"arxiv","version":1}},"canonical_sha256":"52bacf3784a79f57ba205218d9af297d6cfd0df34ba9383b6860d67fd07931d6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"52bacf3784a79f57ba205218d9af297d6cfd0df34ba9383b6860d67fd07931d6","first_computed_at":"2026-05-21T01:04:55.361812Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-21T01:04:55.361812Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"koe5+a2zHBEoRkcAR6E2KOQoFAihwXME7bRAMV06cQ1JjazOmRfvQ2YcxFvK4IT7kjBAUNP1+EpA3W4KPgl0AQ==","signature_status":"signed_v1","signed_at":"2026-05-21T01:04:55.362721Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.20803","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:eb400743e066d9c5757243a22552b5b26d5e0eb6b4fdc594fe4a591b88e1ce55","sha256:05dd6216a2a5eeb18336efda1c432e6f7f52397e0264c258c7b6578dafce1c8a"],"state_sha256":"9d43797803d8fdec7bf9a2c402ad200e01796e49d1005cada1b763c6f9290bd1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2kg1mhWpkMF803TvDAZCHGF2yZ8a0xjs6N0xpc03NcO3ma+wPuIOCZhZMUoR/lZ3sRNZP9e01Eixfags5YhyDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-24T14:34:05.059739Z","bundle_sha256":"79b8687d813c4da099a097e41dd67882c6e0ceb87e7e27bf79f762209c2a4678"}}