{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:KOTMXGBXSLJ7VVGI27KQYYBWSL","short_pith_number":"pith:KOTMXGBX","canonical_record":{"source":{"id":"2407.08699","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-11T17:32:40Z","cross_cats_sorted":[],"title_canon_sha256":"c34f1e4b12a527e51bb1ee974b80e3fe1cb9b47398f974bde452522799198de9","abstract_canon_sha256":"3fc25e1eea712c2ea4db33d7ed6d16662e6cbe4ec149979b8a9954334309a617"},"schema_version":"1.0"},"canonical_sha256":"53a6cb983792d3fad4c8d7d50c603692cf89b2d4b0e6469f5ffa60be004d7eec","source":{"kind":"arxiv","id":"2407.08699","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.08699","created_at":"2026-07-05T08:44:15Z"},{"alias_kind":"arxiv_version","alias_value":"2407.08699v2","created_at":"2026-07-05T08:44:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.08699","created_at":"2026-07-05T08:44:15Z"},{"alias_kind":"pith_short_12","alias_value":"KOTMXGBXSLJ7","created_at":"2026-07-05T08:44:15Z"},{"alias_kind":"pith_short_16","alias_value":"KOTMXGBXSLJ7VVGI","created_at":"2026-07-05T08:44:15Z"},{"alias_kind":"pith_short_8","alias_value":"KOTMXGBX","created_at":"2026-07-05T08:44:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:KOTMXGBXSLJ7VVGI27KQYYBWSL","target":"record","payload":{"canonical_record":{"source":{"id":"2407.08699","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-11T17:32:40Z","cross_cats_sorted":[],"title_canon_sha256":"c34f1e4b12a527e51bb1ee974b80e3fe1cb9b47398f974bde452522799198de9","abstract_canon_sha256":"3fc25e1eea712c2ea4db33d7ed6d16662e6cbe4ec149979b8a9954334309a617"},"schema_version":"1.0"},"canonical_sha256":"53a6cb983792d3fad4c8d7d50c603692cf89b2d4b0e6469f5ffa60be004d7eec","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:44:15.798284Z","signature_b64":"L6DY6NuDO/jgPx8/SmF3+N1U475X3CioJzJntPdXWPe/OyB/UJ0hX0nhlyvbWwZdv60HxGCGMqoZX3E36OUqBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53a6cb983792d3fad4c8d7d50c603692cf89b2d4b0e6469f5ffa60be004d7eec","last_reissued_at":"2026-07-05T08:44:15.797860Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:44:15.797860Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2407.08699","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":"2026-07-05T08:44:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xdIJIVTvn4XA66MN6pvOfr/Fq3CVjyl/LVRUhhEyB67jFW9zDJV5CJEbGdjFHl1A2Xr+0SfnSbkm0THOG9fKAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T08:42:28.351913Z"},"content_sha256":"01bd41f412ea95b7f3e05a78c8043d6f46c91e78743d785b9fd16330d15586ca","schema_version":"1.0","event_id":"sha256:01bd41f412ea95b7f3e05a78c8043d6f46c91e78743d785b9fd16330d15586ca"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:KOTMXGBXSLJ7VVGI27KQYYBWSL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Mitigating Catastrophic Forgetting in Language Transfer via Model Merging","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anton Alexandrov, Ce Zhang, Kristina Toutanova, Mark Niklas M\\\"uller, Martin Vechev, Veselin Raychev","submitted_at":"2024-07-11T17:32:40Z","abstract_excerpt":"As open-weight large language models (LLMs) achieve ever more impressive performances across a wide range of tasks in English, practitioners aim to adapt these models to different languages. However, such language adaptation is often accompanied by catastrophic forgetting of the base model's capabilities, severely limiting the usefulness of the resulting model. We address this issue by proposing Branch-and-Merge (BaM), a new adaptation method based on iteratively merging multiple models, fine-tuned on a subset of the available training data. BaM is based on the insight that this yields lower m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.08699","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/2407.08699/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":"2026-07-05T08:44:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nxoWaTTmfNT+GGv9YwWOM/UPaYQs6IGHxEku2lGXA3r/bU/PVsEZ3iKLc65uok5p04FMs8B5fc/E6QvTz3swCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T08:42:28.352273Z"},"content_sha256":"4ada3dab4b25af981b717640593caaf39833d870dfec82e075fa6b2c7411b77f","schema_version":"1.0","event_id":"sha256:4ada3dab4b25af981b717640593caaf39833d870dfec82e075fa6b2c7411b77f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KOTMXGBXSLJ7VVGI27KQYYBWSL/bundle.json","state_url":"https://pith.science/pith/KOTMXGBXSLJ7VVGI27KQYYBWSL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KOTMXGBXSLJ7VVGI27KQYYBWSL/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":"2026-07-07T08:42:28Z","links":{"resolver":"https://pith.science/pith/KOTMXGBXSLJ7VVGI27KQYYBWSL","bundle":"https://pith.science/pith/KOTMXGBXSLJ7VVGI27KQYYBWSL/bundle.json","state":"https://pith.science/pith/KOTMXGBXSLJ7VVGI27KQYYBWSL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KOTMXGBXSLJ7VVGI27KQYYBWSL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:KOTMXGBXSLJ7VVGI27KQYYBWSL","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":"3fc25e1eea712c2ea4db33d7ed6d16662e6cbe4ec149979b8a9954334309a617","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-11T17:32:40Z","title_canon_sha256":"c34f1e4b12a527e51bb1ee974b80e3fe1cb9b47398f974bde452522799198de9"},"schema_version":"1.0","source":{"id":"2407.08699","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.08699","created_at":"2026-07-05T08:44:15Z"},{"alias_kind":"arxiv_version","alias_value":"2407.08699v2","created_at":"2026-07-05T08:44:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.08699","created_at":"2026-07-05T08:44:15Z"},{"alias_kind":"pith_short_12","alias_value":"KOTMXGBXSLJ7","created_at":"2026-07-05T08:44:15Z"},{"alias_kind":"pith_short_16","alias_value":"KOTMXGBXSLJ7VVGI","created_at":"2026-07-05T08:44:15Z"},{"alias_kind":"pith_short_8","alias_value":"KOTMXGBX","created_at":"2026-07-05T08:44:15Z"}],"graph_snapshots":[{"event_id":"sha256:4ada3dab4b25af981b717640593caaf39833d870dfec82e075fa6b2c7411b77f","target":"graph","created_at":"2026-07-05T08:44:15Z","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/2407.08699/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As open-weight large language models (LLMs) achieve ever more impressive performances across a wide range of tasks in English, practitioners aim to adapt these models to different languages. However, such language adaptation is often accompanied by catastrophic forgetting of the base model's capabilities, severely limiting the usefulness of the resulting model. We address this issue by proposing Branch-and-Merge (BaM), a new adaptation method based on iteratively merging multiple models, fine-tuned on a subset of the available training data. BaM is based on the insight that this yields lower m","authors_text":"Anton Alexandrov, Ce Zhang, Kristina Toutanova, Mark Niklas M\\\"uller, Martin Vechev, Veselin Raychev","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-11T17:32:40Z","title":"Mitigating Catastrophic Forgetting in Language Transfer via Model Merging"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.08699","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":"sha256:01bd41f412ea95b7f3e05a78c8043d6f46c91e78743d785b9fd16330d15586ca","target":"record","created_at":"2026-07-05T08:44:15Z","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":"3fc25e1eea712c2ea4db33d7ed6d16662e6cbe4ec149979b8a9954334309a617","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-11T17:32:40Z","title_canon_sha256":"c34f1e4b12a527e51bb1ee974b80e3fe1cb9b47398f974bde452522799198de9"},"schema_version":"1.0","source":{"id":"2407.08699","kind":"arxiv","version":2}},"canonical_sha256":"53a6cb983792d3fad4c8d7d50c603692cf89b2d4b0e6469f5ffa60be004d7eec","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"53a6cb983792d3fad4c8d7d50c603692cf89b2d4b0e6469f5ffa60be004d7eec","first_computed_at":"2026-07-05T08:44:15.797860Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:44:15.797860Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"L6DY6NuDO/jgPx8/SmF3+N1U475X3CioJzJntPdXWPe/OyB/UJ0hX0nhlyvbWwZdv60HxGCGMqoZX3E36OUqBA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:44:15.798284Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.08699","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:01bd41f412ea95b7f3e05a78c8043d6f46c91e78743d785b9fd16330d15586ca","sha256:4ada3dab4b25af981b717640593caaf39833d870dfec82e075fa6b2c7411b77f"],"state_sha256":"ab1926df38033e6a7fafbc3600898b50b3a6a15484e4f994b2b9dfe96e401aa0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Iz7Edni3W+fG+jQTzGibpOmHQNhwKFSRAl1RJNhfLHhWN3LCtLZW2d86VHH+vhVMsDwJixwK/LU7wv8HNXYuCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T08:42:28.354252Z","bundle_sha256":"6fd5d67107549a99d102b09188bee6af12c9611278755abc38bac2fe29e0b47c"}}