{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:66VP6E6SMQO2XFIVRGH3CADKHP","short_pith_number":"pith:66VP6E6S","schema_version":"1.0","canonical_sha256":"f7aaff13d2641dab9515898fb1006a3bd255c062e3bc2667bdd4cd398315377f","source":{"kind":"arxiv","id":"2407.21571","version":1},"attestation_state":"computed","paper":{"title":"PMoE: Progressive Mixture of Experts with Asymmetric Transformer for Continual Learning","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Joohee Kim, Min Jae Jung","submitted_at":"2024-07-31T12:56:14Z","abstract_excerpt":"Large Language Models (LLMs) encounter significant challenges in continual learning due to catastrophic forgetting, where new information overwrites previously acquired knowledge. This limitation leads to substantial environmental and economic waste. In this study, we introduce the PMoE, Progressive Mixture of Experts with Asymmetric Transformer, which aims to minimize forgetting by utilizing an asymmetric design with shallow layers dedicated to general knowledge and deep layers for new knowledge. PMoE incorporates progressively added experts in deep layers and a router that allocates new know"},"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":"2407.21571","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-31T12:56:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"51c5bfb9f9a0a01243739c16d143f13b7a9783b89ac34f5f5f597703c8dd5d7a","abstract_canon_sha256":"7efd8a84ec3c95b9f21cef3979c9508380d9165e08c0710d110731e30cef02e1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:50:40.961164Z","signature_b64":"rcuMdoU/f+HmPJn+upfCQdkwKMr4qno3bm/ijc9qBnbOC7UdtmgTMOaOEmR5EMvedaNMw4poK+YRHq9DXfVFDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f7aaff13d2641dab9515898fb1006a3bd255c062e3bc2667bdd4cd398315377f","last_reissued_at":"2026-07-05T08:50:40.960689Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:50:40.960689Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PMoE: Progressive Mixture of Experts with Asymmetric Transformer for Continual Learning","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Joohee Kim, Min Jae Jung","submitted_at":"2024-07-31T12:56:14Z","abstract_excerpt":"Large Language Models (LLMs) encounter significant challenges in continual learning due to catastrophic forgetting, where new information overwrites previously acquired knowledge. This limitation leads to substantial environmental and economic waste. In this study, we introduce the PMoE, Progressive Mixture of Experts with Asymmetric Transformer, which aims to minimize forgetting by utilizing an asymmetric design with shallow layers dedicated to general knowledge and deep layers for new knowledge. PMoE incorporates progressively added experts in deep layers and a router that allocates new know"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.21571","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/2407.21571/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":"2407.21571","created_at":"2026-07-05T08:50:40.960746+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.21571v1","created_at":"2026-07-05T08:50:40.960746+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.21571","created_at":"2026-07-05T08:50:40.960746+00:00"},{"alias_kind":"pith_short_12","alias_value":"66VP6E6SMQO2","created_at":"2026-07-05T08:50:40.960746+00:00"},{"alias_kind":"pith_short_16","alias_value":"66VP6E6SMQO2XFIV","created_at":"2026-07-05T08:50:40.960746+00:00"},{"alias_kind":"pith_short_8","alias_value":"66VP6E6S","created_at":"2026-07-05T08:50:40.960746+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06109","citing_title":"RoME: Robust Mixture of Low-Rank Experts against Multiple Adversarial Perturbations","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2509.12958","citing_title":"Forget What's Sensitive, Remember What Matters: Token-Level Differential Privacy in Memory Sculpting for Continual Learning","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/66VP6E6SMQO2XFIVRGH3CADKHP","json":"https://pith.science/pith/66VP6E6SMQO2XFIVRGH3CADKHP.json","graph_json":"https://pith.science/api/pith-number/66VP6E6SMQO2XFIVRGH3CADKHP/graph.json","events_json":"https://pith.science/api/pith-number/66VP6E6SMQO2XFIVRGH3CADKHP/events.json","paper":"https://pith.science/paper/66VP6E6S"},"agent_actions":{"view_html":"https://pith.science/pith/66VP6E6SMQO2XFIVRGH3CADKHP","download_json":"https://pith.science/pith/66VP6E6SMQO2XFIVRGH3CADKHP.json","view_paper":"https://pith.science/paper/66VP6E6S","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.21571&json=true","fetch_graph":"https://pith.science/api/pith-number/66VP6E6SMQO2XFIVRGH3CADKHP/graph.json","fetch_events":"https://pith.science/api/pith-number/66VP6E6SMQO2XFIVRGH3CADKHP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/66VP6E6SMQO2XFIVRGH3CADKHP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/66VP6E6SMQO2XFIVRGH3CADKHP/action/storage_attestation","attest_author":"https://pith.science/pith/66VP6E6SMQO2XFIVRGH3CADKHP/action/author_attestation","sign_citation":"https://pith.science/pith/66VP6E6SMQO2XFIVRGH3CADKHP/action/citation_signature","submit_replication":"https://pith.science/pith/66VP6E6SMQO2XFIVRGH3CADKHP/action/replication_record"}},"created_at":"2026-07-05T08:50:40.960746+00:00","updated_at":"2026-07-05T08:50:40.960746+00:00"}