{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:DMEKZWWHXG2DO2OKA2UO62W4QY","short_pith_number":"pith:DMEKZWWH","canonical_record":{"source":{"id":"2607.24465","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-27T14:04:48Z","cross_cats_sorted":[],"title_canon_sha256":"fa5ca46ad160f0a748bdd158e9d3f248418e12c2ad46da4cc62fea2ca2920d16","abstract_canon_sha256":"adca7473bc3e8e5cebe30cb216f5c2fdfe6ad562946835fdb82790054565fab6"},"schema_version":"1.0"},"canonical_sha256":"1b08acdac7b9b43769ca06a8ef6adc8606c1618340e0bd767c5abebe7f1b2ee0","source":{"kind":"arxiv","id":"2607.24465","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.24465","created_at":"2026-07-28T02:24:05Z"},{"alias_kind":"arxiv_version","alias_value":"2607.24465v1","created_at":"2026-07-28T02:24:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.24465","created_at":"2026-07-28T02:24:05Z"},{"alias_kind":"pith_short_12","alias_value":"DMEKZWWHXG2D","created_at":"2026-07-28T02:24:05Z"},{"alias_kind":"pith_short_16","alias_value":"DMEKZWWHXG2DO2OK","created_at":"2026-07-28T02:24:05Z"},{"alias_kind":"pith_short_8","alias_value":"DMEKZWWH","created_at":"2026-07-28T02:24:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:DMEKZWWHXG2DO2OKA2UO62W4QY","target":"record","payload":{"canonical_record":{"source":{"id":"2607.24465","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-27T14:04:48Z","cross_cats_sorted":[],"title_canon_sha256":"fa5ca46ad160f0a748bdd158e9d3f248418e12c2ad46da4cc62fea2ca2920d16","abstract_canon_sha256":"adca7473bc3e8e5cebe30cb216f5c2fdfe6ad562946835fdb82790054565fab6"},"schema_version":"1.0"},"canonical_sha256":"1b08acdac7b9b43769ca06a8ef6adc8606c1618340e0bd767c5abebe7f1b2ee0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T02:24:05.206100Z","signature_b64":"ANXEp+tJkUbWiPM0ABJ5nmd66Z7pU+n8DE6IYPNIah3wOIEpaNtkInzLpFfJ2NPXST3yNMVDQAzmmAn1fkJyDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1b08acdac7b9b43769ca06a8ef6adc8606c1618340e0bd767c5abebe7f1b2ee0","last_reissued_at":"2026-07-28T02:24:05.205219Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T02:24:05.205219Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.24465","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":"2026-07-28T02:24:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rGNwvueScma0Jt8P/pCRhjbkfBWl+aVqoI8afx21fRhzgpy0UhQsMxMxxlWTzJSHHRW+oxevuAPS66NiVRcjBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T13:57:25.049569Z"},"content_sha256":"9a0a7b15dc3b0f695fc7abb2b6aa5fd86bc2b5722a155cb109f8b91d89972b75","schema_version":"1.0","event_id":"sha256:9a0a7b15dc3b0f695fc7abb2b6aa5fd86bc2b5722a155cb109f8b91d89972b75"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:DMEKZWWHXG2DO2OKA2UO62W4QY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Rethinking Expert Training for Model Merging with Prompt Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aniello Panariello, Christos Georgakilas, Dimosthenis Karatzas, Joost Van De Weijer, Samir El Karrat Moreno, Simone Calderara","submitted_at":"2026-07-27T14:04:48Z","abstract_excerpt":"Model merging aims to combine multiple domain-specialized experts trained from a shared foundation model into a single multi-task model. Existing approaches largely focus on improving the merging procedure itself and typically assume experts obtained through full-parameter fine-tuning. In this work, we revisit expert training for model merging. We first show that prompt-based adaptation provides a strong baseline: independently learned prompts can be exploited across tasks while keeping the backbone fixed, avoiding the interference introduced by weight merging. Building on this observation, we"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.24465","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/2607.24465/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":"2026-07-28T02:24:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kJnL84ITNJfwdSjNO1LpKvHLo1oMrG+QSEA6fatJwHMcUJ4F+Ix5Ck/SJHcV+kcd9ks+7MK7V0nwxkuus4ZUCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T13:57:25.050120Z"},"content_sha256":"f7c2b0ba81f99d63913f1ed9ebf35a471d6bbc1302b3525b1c59bd0c5291c5d5","schema_version":"1.0","event_id":"sha256:f7c2b0ba81f99d63913f1ed9ebf35a471d6bbc1302b3525b1c59bd0c5291c5d5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DMEKZWWHXG2DO2OKA2UO62W4QY/bundle.json","state_url":"https://pith.science/pith/DMEKZWWHXG2DO2OKA2UO62W4QY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DMEKZWWHXG2DO2OKA2UO62W4QY/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":"2026-08-08T13:57:25Z","links":{"resolver":"https://pith.science/pith/DMEKZWWHXG2DO2OKA2UO62W4QY","bundle":"https://pith.science/pith/DMEKZWWHXG2DO2OKA2UO62W4QY/bundle.json","state":"https://pith.science/pith/DMEKZWWHXG2DO2OKA2UO62W4QY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DMEKZWWHXG2DO2OKA2UO62W4QY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DMEKZWWHXG2DO2OKA2UO62W4QY","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":"adca7473bc3e8e5cebe30cb216f5c2fdfe6ad562946835fdb82790054565fab6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-27T14:04:48Z","title_canon_sha256":"fa5ca46ad160f0a748bdd158e9d3f248418e12c2ad46da4cc62fea2ca2920d16"},"schema_version":"1.0","source":{"id":"2607.24465","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.24465","created_at":"2026-07-28T02:24:05Z"},{"alias_kind":"arxiv_version","alias_value":"2607.24465v1","created_at":"2026-07-28T02:24:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.24465","created_at":"2026-07-28T02:24:05Z"},{"alias_kind":"pith_short_12","alias_value":"DMEKZWWHXG2D","created_at":"2026-07-28T02:24:05Z"},{"alias_kind":"pith_short_16","alias_value":"DMEKZWWHXG2DO2OK","created_at":"2026-07-28T02:24:05Z"},{"alias_kind":"pith_short_8","alias_value":"DMEKZWWH","created_at":"2026-07-28T02:24:05Z"}],"graph_snapshots":[{"event_id":"sha256:f7c2b0ba81f99d63913f1ed9ebf35a471d6bbc1302b3525b1c59bd0c5291c5d5","target":"graph","created_at":"2026-07-28T02:24:05Z","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/2607.24465/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Model merging aims to combine multiple domain-specialized experts trained from a shared foundation model into a single multi-task model. Existing approaches largely focus on improving the merging procedure itself and typically assume experts obtained through full-parameter fine-tuning. In this work, we revisit expert training for model merging. We first show that prompt-based adaptation provides a strong baseline: independently learned prompts can be exploited across tasks while keeping the backbone fixed, avoiding the interference introduced by weight merging. Building on this observation, we","authors_text":"Aniello Panariello, Christos Georgakilas, Dimosthenis Karatzas, Joost Van De Weijer, Samir El Karrat Moreno, Simone Calderara","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.24465","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":"sha256:9a0a7b15dc3b0f695fc7abb2b6aa5fd86bc2b5722a155cb109f8b91d89972b75","target":"record","created_at":"2026-07-28T02:24:05Z","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":"adca7473bc3e8e5cebe30cb216f5c2fdfe6ad562946835fdb82790054565fab6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-27T14:04:48Z","title_canon_sha256":"fa5ca46ad160f0a748bdd158e9d3f248418e12c2ad46da4cc62fea2ca2920d16"},"schema_version":"1.0","source":{"id":"2607.24465","kind":"arxiv","version":1}},"canonical_sha256":"1b08acdac7b9b43769ca06a8ef6adc8606c1618340e0bd767c5abebe7f1b2ee0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1b08acdac7b9b43769ca06a8ef6adc8606c1618340e0bd767c5abebe7f1b2ee0","first_computed_at":"2026-07-28T02:24:05.205219Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-28T02:24:05.205219Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ANXEp+tJkUbWiPM0ABJ5nmd66Z7pU+n8DE6IYPNIah3wOIEpaNtkInzLpFfJ2NPXST3yNMVDQAzmmAn1fkJyDA==","signature_status":"signed_v1","signed_at":"2026-07-28T02:24:05.206100Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.24465","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9a0a7b15dc3b0f695fc7abb2b6aa5fd86bc2b5722a155cb109f8b91d89972b75","sha256:f7c2b0ba81f99d63913f1ed9ebf35a471d6bbc1302b3525b1c59bd0c5291c5d5"],"state_sha256":"71538f93e13720294da940016ebdf7b64f5f97f20e6b04fd50c10102912b7994"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2yRcPZruAJZMSDa7DXPSv9AcEO86NbGenESxtiXRA0mO1qwllMMk1dtFl15GjAYxEvmRjEY5ch4iq7w3qdNfDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T13:57:25.053592Z","bundle_sha256":"ac1665e1299c64d4008e2a5a648160fe2edd5a34f1439c0fae8c95ab6f1c80bd"}}