{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:BCBBS4EUFEU67ZFSNWUPTV666P","short_pith_number":"pith:BCBBS4EU","schema_version":"1.0","canonical_sha256":"08821970942929efe4b26da8f9d7def3c95e34b97876f1cc347394ae3f66413b","source":{"kind":"arxiv","id":"2506.23271","version":1},"attestation_state":"computed","paper":{"title":"Mettle: Meta-Token Learning for Memory-Efficient Audio-Visual Adaptation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guangyao Li, Jinxing Zhou, Meng Wang, Mingfei Han, Ruohao Guo, Xiaojun Chang, Yanghao Zhou, Yongqiang Yu, Yuxin Mao, Zhihui Li","submitted_at":"2025-06-29T14:52:01Z","abstract_excerpt":"We present \\textbf{Met}a-\\textbf{T}oken \\textbf{Le}arning (Mettle), a simple and memory-efficient method for adapting large-scale pretrained transformer models to downstream audio-visual tasks. Instead of sequentially modifying the output feature distribution of the transformer backbone, Mettle utilizes a lightweight \\textit{Layer-Centric Distillation (LCD)} module to distill in parallel the intact audio or visual features embedded by each transformer layer into compact meta-tokens. This distillation process considers both pretrained knowledge preservation and task-specific adaptation. The obt"},"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":"2506.23271","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-29T14:52:01Z","cross_cats_sorted":[],"title_canon_sha256":"d6f4645bb5b945820314bf1c83ff6f846e94682f8b333094c7c8399e74f05bba","abstract_canon_sha256":"aea43068cc70c90b7cde511ef6b999af825a2742cd702728d13add4357087b00"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:29:19.045861Z","signature_b64":"YXFtWJ3CFeLc9A4oVC7r7In0wMUjcTMr9JYmG6KhatFKdPvY/cMIg+Ka80u41f5T27Tvom0vJj5L6ndsTjf7AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"08821970942929efe4b26da8f9d7def3c95e34b97876f1cc347394ae3f66413b","last_reissued_at":"2026-07-05T11:29:19.045292Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:29:19.045292Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mettle: Meta-Token Learning for Memory-Efficient Audio-Visual Adaptation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guangyao Li, Jinxing Zhou, Meng Wang, Mingfei Han, Ruohao Guo, Xiaojun Chang, Yanghao Zhou, Yongqiang Yu, Yuxin Mao, Zhihui Li","submitted_at":"2025-06-29T14:52:01Z","abstract_excerpt":"We present \\textbf{Met}a-\\textbf{T}oken \\textbf{Le}arning (Mettle), a simple and memory-efficient method for adapting large-scale pretrained transformer models to downstream audio-visual tasks. Instead of sequentially modifying the output feature distribution of the transformer backbone, Mettle utilizes a lightweight \\textit{Layer-Centric Distillation (LCD)} module to distill in parallel the intact audio or visual features embedded by each transformer layer into compact meta-tokens. This distillation process considers both pretrained knowledge preservation and task-specific adaptation. The obt"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.23271","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/2506.23271/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":"2506.23271","created_at":"2026-07-05T11:29:19.045364+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.23271v1","created_at":"2026-07-05T11:29:19.045364+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.23271","created_at":"2026-07-05T11:29:19.045364+00:00"},{"alias_kind":"pith_short_12","alias_value":"BCBBS4EUFEU6","created_at":"2026-07-05T11:29:19.045364+00:00"},{"alias_kind":"pith_short_16","alias_value":"BCBBS4EUFEU67ZFS","created_at":"2026-07-05T11:29:19.045364+00:00"},{"alias_kind":"pith_short_8","alias_value":"BCBBS4EU","created_at":"2026-07-05T11:29:19.045364+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BCBBS4EUFEU67ZFSNWUPTV666P","json":"https://pith.science/pith/BCBBS4EUFEU67ZFSNWUPTV666P.json","graph_json":"https://pith.science/api/pith-number/BCBBS4EUFEU67ZFSNWUPTV666P/graph.json","events_json":"https://pith.science/api/pith-number/BCBBS4EUFEU67ZFSNWUPTV666P/events.json","paper":"https://pith.science/paper/BCBBS4EU"},"agent_actions":{"view_html":"https://pith.science/pith/BCBBS4EUFEU67ZFSNWUPTV666P","download_json":"https://pith.science/pith/BCBBS4EUFEU67ZFSNWUPTV666P.json","view_paper":"https://pith.science/paper/BCBBS4EU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.23271&json=true","fetch_graph":"https://pith.science/api/pith-number/BCBBS4EUFEU67ZFSNWUPTV666P/graph.json","fetch_events":"https://pith.science/api/pith-number/BCBBS4EUFEU67ZFSNWUPTV666P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BCBBS4EUFEU67ZFSNWUPTV666P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BCBBS4EUFEU67ZFSNWUPTV666P/action/storage_attestation","attest_author":"https://pith.science/pith/BCBBS4EUFEU67ZFSNWUPTV666P/action/author_attestation","sign_citation":"https://pith.science/pith/BCBBS4EUFEU67ZFSNWUPTV666P/action/citation_signature","submit_replication":"https://pith.science/pith/BCBBS4EUFEU67ZFSNWUPTV666P/action/replication_record"}},"created_at":"2026-07-05T11:29:19.045364+00:00","updated_at":"2026-07-05T11:29:19.045364+00:00"}