{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:4WZIP6CZZC54QALZPKZIDDMZOZ","short_pith_number":"pith:4WZIP6CZ","schema_version":"1.0","canonical_sha256":"e5b287f859c8bbc801797ab2818d997671706021fbf0c0c6ccec72869efbbc1a","source":{"kind":"arxiv","id":"2409.19611","version":1},"attestation_state":"computed","paper":{"title":"Learning Attentional Mixture of LoRAs for Language Model Continual Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jialin Liu, Jianhua Wu, Jie Liu, Yutai Duan","submitted_at":"2024-09-29T08:34:54Z","abstract_excerpt":"Fine-tuning large language models (LLMs) with Low-Rank adaption (LoRA) is widely acknowledged as an effective approach for continual learning for new tasks. However, it often suffers from catastrophic forgetting when dealing with multiple tasks sequentially. To this end, we propose Attentional Mixture of LoRAs (AM-LoRA), a continual learning approach tailored for LLMs. Specifically, AM-LoRA learns a sequence of LoRAs for a series of tasks to continually learn knowledge from different tasks. The key of our approach is that we devise an attention mechanism as a knowledge mixture module to adapti"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2409.19611","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-09-29T08:34:54Z","cross_cats_sorted":[],"title_canon_sha256":"c483b8a01473fc4c154c4b45d359fb4c12a88d162bc45df4090c6a854c0da2d9","abstract_canon_sha256":"98dbb88c6de13563b0958d294f2016e3abb116baf0e6ca9571589edf8941a3c5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:13:18.269809Z","signature_b64":"0aTppEHWxpPwT4swN20FHokIvzbioYcvwvOZir3yVOx8pdJPjpThn/rmCatNfq4d/JC7hXWRyETHjMfq2iI2DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e5b287f859c8bbc801797ab2818d997671706021fbf0c0c6ccec72869efbbc1a","last_reissued_at":"2026-07-05T09:13:18.269435Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:13:18.269435Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Attentional Mixture of LoRAs for Language Model Continual Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jialin Liu, Jianhua Wu, Jie Liu, Yutai Duan","submitted_at":"2024-09-29T08:34:54Z","abstract_excerpt":"Fine-tuning large language models (LLMs) with Low-Rank adaption (LoRA) is widely acknowledged as an effective approach for continual learning for new tasks. However, it often suffers from catastrophic forgetting when dealing with multiple tasks sequentially. To this end, we propose Attentional Mixture of LoRAs (AM-LoRA), a continual learning approach tailored for LLMs. Specifically, AM-LoRA learns a sequence of LoRAs for a series of tasks to continually learn knowledge from different tasks. The key of our approach is that we devise an attention mechanism as a knowledge mixture module to adapti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.19611","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/2409.19611/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2409.19611","created_at":"2026-07-05T09:13:18.269495+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.19611v1","created_at":"2026-07-05T09:13:18.269495+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.19611","created_at":"2026-07-05T09:13:18.269495+00:00"},{"alias_kind":"pith_short_12","alias_value":"4WZIP6CZZC54","created_at":"2026-07-05T09:13:18.269495+00:00"},{"alias_kind":"pith_short_16","alias_value":"4WZIP6CZZC54QALZ","created_at":"2026-07-05T09:13:18.269495+00:00"},{"alias_kind":"pith_short_8","alias_value":"4WZIP6CZ","created_at":"2026-07-05T09:13:18.269495+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.02503","citing_title":"Continual Gradient Low-Rank Projection Fine-Tuning for LLMs","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4WZIP6CZZC54QALZPKZIDDMZOZ","json":"https://pith.science/pith/4WZIP6CZZC54QALZPKZIDDMZOZ.json","graph_json":"https://pith.science/api/pith-number/4WZIP6CZZC54QALZPKZIDDMZOZ/graph.json","events_json":"https://pith.science/api/pith-number/4WZIP6CZZC54QALZPKZIDDMZOZ/events.json","paper":"https://pith.science/paper/4WZIP6CZ"},"agent_actions":{"view_html":"https://pith.science/pith/4WZIP6CZZC54QALZPKZIDDMZOZ","download_json":"https://pith.science/pith/4WZIP6CZZC54QALZPKZIDDMZOZ.json","view_paper":"https://pith.science/paper/4WZIP6CZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.19611&json=true","fetch_graph":"https://pith.science/api/pith-number/4WZIP6CZZC54QALZPKZIDDMZOZ/graph.json","fetch_events":"https://pith.science/api/pith-number/4WZIP6CZZC54QALZPKZIDDMZOZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4WZIP6CZZC54QALZPKZIDDMZOZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4WZIP6CZZC54QALZPKZIDDMZOZ/action/storage_attestation","attest_author":"https://pith.science/pith/4WZIP6CZZC54QALZPKZIDDMZOZ/action/author_attestation","sign_citation":"https://pith.science/pith/4WZIP6CZZC54QALZPKZIDDMZOZ/action/citation_signature","submit_replication":"https://pith.science/pith/4WZIP6CZZC54QALZPKZIDDMZOZ/action/replication_record"}},"created_at":"2026-07-05T09:13:18.269495+00:00","updated_at":"2026-07-05T09:13:18.269495+00:00"}