{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:C6A7VSIWLOH3JVSH5QZXB4IMMI","short_pith_number":"pith:C6A7VSIW","canonical_record":{"source":{"id":"2605.18993","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-18T18:11:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"54f8681727d76106d5378b2f81306f574451c1d29b2ae79783c0bca66c9a8db9","abstract_canon_sha256":"4d97c009f2dd9865c40a159bf6a52dd7b692ad3eebae8e7cedc5ae26b953c672"},"schema_version":"1.0"},"canonical_sha256":"1781fac9165b8fb4d647ec3370f10c620e3c04a2afa0274837a52e52ddc07291","source":{"kind":"arxiv","id":"2605.18993","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.18993","created_at":"2026-05-20T00:06:41Z"},{"alias_kind":"arxiv_version","alias_value":"2605.18993v1","created_at":"2026-05-20T00:06:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18993","created_at":"2026-05-20T00:06:41Z"},{"alias_kind":"pith_short_12","alias_value":"C6A7VSIWLOH3","created_at":"2026-05-20T00:06:41Z"},{"alias_kind":"pith_short_16","alias_value":"C6A7VSIWLOH3JVSH","created_at":"2026-05-20T00:06:41Z"},{"alias_kind":"pith_short_8","alias_value":"C6A7VSIW","created_at":"2026-05-20T00:06:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:C6A7VSIWLOH3JVSH5QZXB4IMMI","target":"record","payload":{"canonical_record":{"source":{"id":"2605.18993","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-18T18:11:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"54f8681727d76106d5378b2f81306f574451c1d29b2ae79783c0bca66c9a8db9","abstract_canon_sha256":"4d97c009f2dd9865c40a159bf6a52dd7b692ad3eebae8e7cedc5ae26b953c672"},"schema_version":"1.0"},"canonical_sha256":"1781fac9165b8fb4d647ec3370f10c620e3c04a2afa0274837a52e52ddc07291","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:06:41.985033Z","signature_b64":"Smb14mqQxNUkQRdtfnWKmreGvD0Vf5PpISyk7g+yKbAZVU1WuOtTZymVS6LgPOn+QGXTfyr3yleREk4wMmnWBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1781fac9165b8fb4d647ec3370f10c620e3c04a2afa0274837a52e52ddc07291","last_reissued_at":"2026-05-20T00:06:41.984092Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:06:41.984092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.18993","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":"2026-05-20T00:06:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A/XHjwUdbFNrVKnHf+elRCwoo20FJnJjacxqM0l2HDCCs2strwM3UWrZdP2qLmvof97qEeXfyqjz37eItl72AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T08:54:25.228216Z"},"content_sha256":"db49ecc3013c00dcd6d6d8964e31e3a70fbea54ef1ef3b2d54a2325719120a2a","schema_version":"1.0","event_id":"sha256:db49ecc3013c00dcd6d6d8964e31e3a70fbea54ef1ef3b2d54a2325719120a2a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:C6A7VSIWLOH3JVSH5QZXB4IMMI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Distilling Linearized Behavior for Effective Task Arithmetic","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Angelo Porrello, Francesca Morandi, Simone Calderara, Thomas Sommariva","submitted_at":"2026-05-18T18:11:34Z","abstract_excerpt":"Task vector composition has emerged as a promising paradigm for editing pre-trained models, enabling model merging through addition and unlearning through subtraction. Fine-tuning in the tangent space of a pre-trained model (linear fine-tuning) has proven effective, as it produces task vectors that are naturally disentangled and resistant to interference. However, linearized models suffer from limited expressivity during training and incur higher computational costs at inference time, which restrict their practical applicability. In this work, we bridge the gap between linear and standard non-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18993","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/2605.18993/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":"2026-05-20T00:06:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0JqZ0+oQyJOhhoqoee2ihanOhJb9WL8QpwyMF610JRRIhK4l2KTrg6CDsGq6XfUBWgo3XGg9MOYGHIkE4aDxCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T08:54:25.228958Z"},"content_sha256":"2dadf6f6f35a1f5785deaf29f70aa36c156eaf0ee38be7ac5d34067d0ea51294","schema_version":"1.0","event_id":"sha256:2dadf6f6f35a1f5785deaf29f70aa36c156eaf0ee38be7ac5d34067d0ea51294"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/C6A7VSIWLOH3JVSH5QZXB4IMMI/bundle.json","state_url":"https://pith.science/pith/C6A7VSIWLOH3JVSH5QZXB4IMMI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/C6A7VSIWLOH3JVSH5QZXB4IMMI/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":"2026-05-27T08:54:25Z","links":{"resolver":"https://pith.science/pith/C6A7VSIWLOH3JVSH5QZXB4IMMI","bundle":"https://pith.science/pith/C6A7VSIWLOH3JVSH5QZXB4IMMI/bundle.json","state":"https://pith.science/pith/C6A7VSIWLOH3JVSH5QZXB4IMMI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/C6A7VSIWLOH3JVSH5QZXB4IMMI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:C6A7VSIWLOH3JVSH5QZXB4IMMI","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":"4d97c009f2dd9865c40a159bf6a52dd7b692ad3eebae8e7cedc5ae26b953c672","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-18T18:11:34Z","title_canon_sha256":"54f8681727d76106d5378b2f81306f574451c1d29b2ae79783c0bca66c9a8db9"},"schema_version":"1.0","source":{"id":"2605.18993","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.18993","created_at":"2026-05-20T00:06:41Z"},{"alias_kind":"arxiv_version","alias_value":"2605.18993v1","created_at":"2026-05-20T00:06:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18993","created_at":"2026-05-20T00:06:41Z"},{"alias_kind":"pith_short_12","alias_value":"C6A7VSIWLOH3","created_at":"2026-05-20T00:06:41Z"},{"alias_kind":"pith_short_16","alias_value":"C6A7VSIWLOH3JVSH","created_at":"2026-05-20T00:06:41Z"},{"alias_kind":"pith_short_8","alias_value":"C6A7VSIW","created_at":"2026-05-20T00:06:41Z"}],"graph_snapshots":[{"event_id":"sha256:2dadf6f6f35a1f5785deaf29f70aa36c156eaf0ee38be7ac5d34067d0ea51294","target":"graph","created_at":"2026-05-20T00:06:41Z","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/2605.18993/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Task vector composition has emerged as a promising paradigm for editing pre-trained models, enabling model merging through addition and unlearning through subtraction. Fine-tuning in the tangent space of a pre-trained model (linear fine-tuning) has proven effective, as it produces task vectors that are naturally disentangled and resistant to interference. However, linearized models suffer from limited expressivity during training and incur higher computational costs at inference time, which restrict their practical applicability. In this work, we bridge the gap between linear and standard non-","authors_text":"Angelo Porrello, Francesca Morandi, Simone Calderara, Thomas Sommariva","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-18T18:11:34Z","title":"Distilling Linearized Behavior for Effective Task Arithmetic"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18993","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:db49ecc3013c00dcd6d6d8964e31e3a70fbea54ef1ef3b2d54a2325719120a2a","target":"record","created_at":"2026-05-20T00:06:41Z","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":"4d97c009f2dd9865c40a159bf6a52dd7b692ad3eebae8e7cedc5ae26b953c672","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-18T18:11:34Z","title_canon_sha256":"54f8681727d76106d5378b2f81306f574451c1d29b2ae79783c0bca66c9a8db9"},"schema_version":"1.0","source":{"id":"2605.18993","kind":"arxiv","version":1}},"canonical_sha256":"1781fac9165b8fb4d647ec3370f10c620e3c04a2afa0274837a52e52ddc07291","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1781fac9165b8fb4d647ec3370f10c620e3c04a2afa0274837a52e52ddc07291","first_computed_at":"2026-05-20T00:06:41.984092Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:06:41.984092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Smb14mqQxNUkQRdtfnWKmreGvD0Vf5PpISyk7g+yKbAZVU1WuOtTZymVS6LgPOn+QGXTfyr3yleREk4wMmnWBQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:06:41.985033Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.18993","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:db49ecc3013c00dcd6d6d8964e31e3a70fbea54ef1ef3b2d54a2325719120a2a","sha256:2dadf6f6f35a1f5785deaf29f70aa36c156eaf0ee38be7ac5d34067d0ea51294"],"state_sha256":"9828c01214ef480974e4e2b0d3471c2fb6ac0f71ba0fea900f58450ee6798026"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UNIJWGY6kmAsfqJn30/jBHQxwAXKCjNv+LOdyjGxzEiJI0jsKW6YSpT9pKqDpJweLkFAlcdzMOga2g8b65HwDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T08:54:25.232636Z","bundle_sha256":"1588899663d6b392192ac005439c9e9e09aea2e838f904a7dbe56f4f1a904408"}}