{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:KGPC5XBR5UX5BOGWPH5B6Y4PYG","short_pith_number":"pith:KGPC5XBR","canonical_record":{"source":{"id":"2402.17934","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-27T23:12:45Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c54e145ba6dad7a0bc19281566ca1b8cdab4895c96b5da9404f8f1e9bd341afa","abstract_canon_sha256":"978c60aa52d01a05dd1ce6e95a9d2b9cc3eb6b574350346981f5192d328a1685"},"schema_version":"1.0"},"canonical_sha256":"519e2edc31ed2fd0b8d679fa1f638fc1a0e931b12b4f476b6c0d48922d4f4589","source":{"kind":"arxiv","id":"2402.17934","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.17934","created_at":"2026-07-05T08:50:54Z"},{"alias_kind":"arxiv_version","alias_value":"2402.17934v2","created_at":"2026-07-05T08:50:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.17934","created_at":"2026-07-05T08:50:54Z"},{"alias_kind":"pith_short_12","alias_value":"KGPC5XBR5UX5","created_at":"2026-07-05T08:50:54Z"},{"alias_kind":"pith_short_16","alias_value":"KGPC5XBR5UX5BOGW","created_at":"2026-07-05T08:50:54Z"},{"alias_kind":"pith_short_8","alias_value":"KGPC5XBR","created_at":"2026-07-05T08:50:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:KGPC5XBR5UX5BOGWPH5B6Y4PYG","target":"record","payload":{"canonical_record":{"source":{"id":"2402.17934","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-27T23:12:45Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c54e145ba6dad7a0bc19281566ca1b8cdab4895c96b5da9404f8f1e9bd341afa","abstract_canon_sha256":"978c60aa52d01a05dd1ce6e95a9d2b9cc3eb6b574350346981f5192d328a1685"},"schema_version":"1.0"},"canonical_sha256":"519e2edc31ed2fd0b8d679fa1f638fc1a0e931b12b4f476b6c0d48922d4f4589","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:50:54.234613Z","signature_b64":"XsNKOf6/mtSXiAOKr4vwiu+WtTajsGSx3/ukHFePnlkIW1mmk/FCTrqx2iAHoA0JUsIKL4Ydi6Jnc19JM6kQAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"519e2edc31ed2fd0b8d679fa1f638fc1a0e931b12b4f476b6c0d48922d4f4589","last_reissued_at":"2026-07-05T08:50:54.234142Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:50:54.234142Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.17934","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":"2026-07-05T08:50:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A/r5fj86AjjITt5ESggiv2rDPo9ZOOm90gJrwCL9yjvUd7IwnPWBSRJg7UCmMCuWa1pT8uqGfr13xJXcPZynCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T06:14:52.840408Z"},"content_sha256":"ac72cd3d672749f4f14b1a388d26af8599855eb550e21d5a7000f79324f539bb","schema_version":"1.0","event_id":"sha256:ac72cd3d672749f4f14b1a388d26af8599855eb550e21d5a7000f79324f539bb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:KGPC5XBR5UX5BOGWPH5B6Y4PYG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Inducing Generalization across Languages and Tasks using Featurized Low-Rank Mixtures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chenxi Whitehouse, Chu-Cheng Lin, Han Lu, Hongkun Yu, Jonathan H. Clark, Xinyi Wang, Yun Zhu","submitted_at":"2024-02-27T23:12:45Z","abstract_excerpt":"Adapting pretrained large language models (LLMs) to various downstream tasks in tens or hundreds of human languages is computationally expensive. Parameter-efficient fine-tuning (PEFT) significantly reduces the adaptation cost, by tuning only a small amount of parameters. However, common PEFT methods LoRA (Hu et al., 2022) suffer from suboptimal performance on diverse dataset mixtures, due to aggressive parameter tying and negative interference among different datasets. In this work, we propose Featurized Low-rank Mixtures (FLix), a novel PEFT method designed for effective multitask multilingu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.17934","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/2402.17934/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":"2026-07-05T08:50:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0M/YCqtnt8FIY8Ai1xDlVHGq/KUn1Est/1qShMHpFixnalMYF9DXc/fjc0+eSBdSRnaFDdIC7H4fOxjfFly+AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T06:14:52.840925Z"},"content_sha256":"666afc4139152bf4a170e7ab3b411badcfe2baf74c3cb983bff15931196f4f0d","schema_version":"1.0","event_id":"sha256:666afc4139152bf4a170e7ab3b411badcfe2baf74c3cb983bff15931196f4f0d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KGPC5XBR5UX5BOGWPH5B6Y4PYG/bundle.json","state_url":"https://pith.science/pith/KGPC5XBR5UX5BOGWPH5B6Y4PYG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KGPC5XBR5UX5BOGWPH5B6Y4PYG/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":"2026-08-12T06:14:52Z","links":{"resolver":"https://pith.science/pith/KGPC5XBR5UX5BOGWPH5B6Y4PYG","bundle":"https://pith.science/pith/KGPC5XBR5UX5BOGWPH5B6Y4PYG/bundle.json","state":"https://pith.science/pith/KGPC5XBR5UX5BOGWPH5B6Y4PYG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KGPC5XBR5UX5BOGWPH5B6Y4PYG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:KGPC5XBR5UX5BOGWPH5B6Y4PYG","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":"978c60aa52d01a05dd1ce6e95a9d2b9cc3eb6b574350346981f5192d328a1685","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-27T23:12:45Z","title_canon_sha256":"c54e145ba6dad7a0bc19281566ca1b8cdab4895c96b5da9404f8f1e9bd341afa"},"schema_version":"1.0","source":{"id":"2402.17934","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.17934","created_at":"2026-07-05T08:50:54Z"},{"alias_kind":"arxiv_version","alias_value":"2402.17934v2","created_at":"2026-07-05T08:50:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.17934","created_at":"2026-07-05T08:50:54Z"},{"alias_kind":"pith_short_12","alias_value":"KGPC5XBR5UX5","created_at":"2026-07-05T08:50:54Z"},{"alias_kind":"pith_short_16","alias_value":"KGPC5XBR5UX5BOGW","created_at":"2026-07-05T08:50:54Z"},{"alias_kind":"pith_short_8","alias_value":"KGPC5XBR","created_at":"2026-07-05T08:50:54Z"}],"graph_snapshots":[{"event_id":"sha256:666afc4139152bf4a170e7ab3b411badcfe2baf74c3cb983bff15931196f4f0d","target":"graph","created_at":"2026-07-05T08:50:54Z","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/2402.17934/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Adapting pretrained large language models (LLMs) to various downstream tasks in tens or hundreds of human languages is computationally expensive. Parameter-efficient fine-tuning (PEFT) significantly reduces the adaptation cost, by tuning only a small amount of parameters. However, common PEFT methods LoRA (Hu et al., 2022) suffer from suboptimal performance on diverse dataset mixtures, due to aggressive parameter tying and negative interference among different datasets. In this work, we propose Featurized Low-rank Mixtures (FLix), a novel PEFT method designed for effective multitask multilingu","authors_text":"Chenxi Whitehouse, Chu-Cheng Lin, Han Lu, Hongkun Yu, Jonathan H. Clark, Xinyi Wang, Yun Zhu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-27T23:12:45Z","title":"Inducing Generalization across Languages and Tasks using Featurized Low-Rank Mixtures"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.17934","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:ac72cd3d672749f4f14b1a388d26af8599855eb550e21d5a7000f79324f539bb","target":"record","created_at":"2026-07-05T08:50:54Z","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":"978c60aa52d01a05dd1ce6e95a9d2b9cc3eb6b574350346981f5192d328a1685","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-27T23:12:45Z","title_canon_sha256":"c54e145ba6dad7a0bc19281566ca1b8cdab4895c96b5da9404f8f1e9bd341afa"},"schema_version":"1.0","source":{"id":"2402.17934","kind":"arxiv","version":2}},"canonical_sha256":"519e2edc31ed2fd0b8d679fa1f638fc1a0e931b12b4f476b6c0d48922d4f4589","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"519e2edc31ed2fd0b8d679fa1f638fc1a0e931b12b4f476b6c0d48922d4f4589","first_computed_at":"2026-07-05T08:50:54.234142Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:50:54.234142Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XsNKOf6/mtSXiAOKr4vwiu+WtTajsGSx3/ukHFePnlkIW1mmk/FCTrqx2iAHoA0JUsIKL4Ydi6Jnc19JM6kQAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:50:54.234613Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.17934","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ac72cd3d672749f4f14b1a388d26af8599855eb550e21d5a7000f79324f539bb","sha256:666afc4139152bf4a170e7ab3b411badcfe2baf74c3cb983bff15931196f4f0d"],"state_sha256":"e3b6af186b5481a60c28367dfa3f0f74b9d5411d837582bc469a823e8196ca66"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GNJ+mjsJ79xtZB4wYNl6wWbCYMWLHxllq/It6Abo3P09N6M9bxN3Wu6Rk8HceKldpHhOtsnFKPzrPOV9Um0TDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T06:14:52.846187Z","bundle_sha256":"6fd9ffcc368d097240944eca9f9a59ab842c22d800d074604f380b9e6ab8d4b4"}}