{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:LND3R6LCFCMAOSI6PZHZSKG3FI","short_pith_number":"pith:LND3R6LC","canonical_record":{"source":{"id":"2502.00089","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-31T18:07:21Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"01e497d21e46c7d4912550024eb865652d4d8258aa457dbba4f50e53898052be","abstract_canon_sha256":"de38e16624e83eb91f7b45b0b8e32f2624e0fb2ff08b89a52e5fa31d1df916ad"},"schema_version":"1.0"},"canonical_sha256":"5b47b8f962289807491e7e4f9928db2a012fefbba1e8d357f99647b3ee6fd042","source":{"kind":"arxiv","id":"2502.00089","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.00089","created_at":"2026-07-05T10:08:29Z"},{"alias_kind":"arxiv_version","alias_value":"2502.00089v1","created_at":"2026-07-05T10:08:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.00089","created_at":"2026-07-05T10:08:29Z"},{"alias_kind":"pith_short_12","alias_value":"LND3R6LCFCMA","created_at":"2026-07-05T10:08:29Z"},{"alias_kind":"pith_short_16","alias_value":"LND3R6LCFCMAOSI6","created_at":"2026-07-05T10:08:29Z"},{"alias_kind":"pith_short_8","alias_value":"LND3R6LC","created_at":"2026-07-05T10:08:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:LND3R6LCFCMAOSI6PZHZSKG3FI","target":"record","payload":{"canonical_record":{"source":{"id":"2502.00089","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-31T18:07:21Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"01e497d21e46c7d4912550024eb865652d4d8258aa457dbba4f50e53898052be","abstract_canon_sha256":"de38e16624e83eb91f7b45b0b8e32f2624e0fb2ff08b89a52e5fa31d1df916ad"},"schema_version":"1.0"},"canonical_sha256":"5b47b8f962289807491e7e4f9928db2a012fefbba1e8d357f99647b3ee6fd042","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:08:29.862016Z","signature_b64":"gq6/4GhfVH9NoEd1RWXYvHPRyN1h2FbmmnVGDp7k4lV1k+i5Y0erkSjFYN8HhXEj8G+Bd4Y5/Qy1iKjTtJHBAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5b47b8f962289807491e7e4f9928db2a012fefbba1e8d357f99647b3ee6fd042","last_reissued_at":"2026-07-05T10:08:29.861571Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:08:29.861571Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.00089","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:08:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nK2BPuCHo6etNjM/NQT29e8rTko5W5wRvRxqBx/4wEsQ1vOatlgkcPXenBTq+Fe/HDYgDsS0/R6NtMWnw4jEDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T02:16:55.264937Z"},"content_sha256":"b9565b8cf7682ca03a1cf092be84508bf048d2359107628e284f872adb68bd88","schema_version":"1.0","event_id":"sha256:b9565b8cf7682ca03a1cf092be84508bf048d2359107628e284f872adb68bd88"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:LND3R6LCFCMAOSI6PZHZSKG3FI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Ensembles of Low-Rank Expert Adapters","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Chao Zhang, MohamadAli Torkamani, Vianne Gao, Yinghao Li","submitted_at":"2025-01-31T18:07:21Z","abstract_excerpt":"The training and fine-tuning of large language models (LLMs) often involve diverse textual data from multiple sources, which poses challenges due to conflicting gradient directions, hindering optimization and specialization. These challenges can undermine model generalization across tasks, resulting in reduced downstream performance. Recent research suggests that fine-tuning LLMs on carefully selected, task-specific subsets of data can match or even surpass the performance of using the entire dataset. Building on these insights, we propose the Ensembles of Low-Rank Expert Adapters (ELREA) fram"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.00089","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/2502.00089/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":"2026-07-05T10:08:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Kw5wypoJaVWXxihYoMPiwkNKVdZBnPV/dWeFihqMVhdxr0DtScPRk3Fzm3zmIn/qfQKnHWlMgip2qFxpyyBhAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T02:16:55.265483Z"},"content_sha256":"21515b589087709ee3f31ac77aa87d2a3484c505afa467706f2ae21fe17a292e","schema_version":"1.0","event_id":"sha256:21515b589087709ee3f31ac77aa87d2a3484c505afa467706f2ae21fe17a292e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LND3R6LCFCMAOSI6PZHZSKG3FI/bundle.json","state_url":"https://pith.science/pith/LND3R6LCFCMAOSI6PZHZSKG3FI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LND3R6LCFCMAOSI6PZHZSKG3FI/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":"2026-08-05T02:16:55Z","links":{"resolver":"https://pith.science/pith/LND3R6LCFCMAOSI6PZHZSKG3FI","bundle":"https://pith.science/pith/LND3R6LCFCMAOSI6PZHZSKG3FI/bundle.json","state":"https://pith.science/pith/LND3R6LCFCMAOSI6PZHZSKG3FI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LND3R6LCFCMAOSI6PZHZSKG3FI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:LND3R6LCFCMAOSI6PZHZSKG3FI","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":"de38e16624e83eb91f7b45b0b8e32f2624e0fb2ff08b89a52e5fa31d1df916ad","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-31T18:07:21Z","title_canon_sha256":"01e497d21e46c7d4912550024eb865652d4d8258aa457dbba4f50e53898052be"},"schema_version":"1.0","source":{"id":"2502.00089","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.00089","created_at":"2026-07-05T10:08:29Z"},{"alias_kind":"arxiv_version","alias_value":"2502.00089v1","created_at":"2026-07-05T10:08:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.00089","created_at":"2026-07-05T10:08:29Z"},{"alias_kind":"pith_short_12","alias_value":"LND3R6LCFCMA","created_at":"2026-07-05T10:08:29Z"},{"alias_kind":"pith_short_16","alias_value":"LND3R6LCFCMAOSI6","created_at":"2026-07-05T10:08:29Z"},{"alias_kind":"pith_short_8","alias_value":"LND3R6LC","created_at":"2026-07-05T10:08:29Z"}],"graph_snapshots":[{"event_id":"sha256:21515b589087709ee3f31ac77aa87d2a3484c505afa467706f2ae21fe17a292e","target":"graph","created_at":"2026-07-05T10:08:29Z","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/2502.00089/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The training and fine-tuning of large language models (LLMs) often involve diverse textual data from multiple sources, which poses challenges due to conflicting gradient directions, hindering optimization and specialization. These challenges can undermine model generalization across tasks, resulting in reduced downstream performance. Recent research suggests that fine-tuning LLMs on carefully selected, task-specific subsets of data can match or even surpass the performance of using the entire dataset. Building on these insights, we propose the Ensembles of Low-Rank Expert Adapters (ELREA) fram","authors_text":"Chao Zhang, MohamadAli Torkamani, Vianne Gao, Yinghao Li","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-31T18:07:21Z","title":"Ensembles of Low-Rank Expert Adapters"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.00089","kind":"arxiv","version":1},"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:b9565b8cf7682ca03a1cf092be84508bf048d2359107628e284f872adb68bd88","target":"record","created_at":"2026-07-05T10:08:29Z","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":"de38e16624e83eb91f7b45b0b8e32f2624e0fb2ff08b89a52e5fa31d1df916ad","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-31T18:07:21Z","title_canon_sha256":"01e497d21e46c7d4912550024eb865652d4d8258aa457dbba4f50e53898052be"},"schema_version":"1.0","source":{"id":"2502.00089","kind":"arxiv","version":1}},"canonical_sha256":"5b47b8f962289807491e7e4f9928db2a012fefbba1e8d357f99647b3ee6fd042","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5b47b8f962289807491e7e4f9928db2a012fefbba1e8d357f99647b3ee6fd042","first_computed_at":"2026-07-05T10:08:29.861571Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:08:29.861571Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"gq6/4GhfVH9NoEd1RWXYvHPRyN1h2FbmmnVGDp7k4lV1k+i5Y0erkSjFYN8HhXEj8G+Bd4Y5/Qy1iKjTtJHBAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:08:29.862016Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.00089","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b9565b8cf7682ca03a1cf092be84508bf048d2359107628e284f872adb68bd88","sha256:21515b589087709ee3f31ac77aa87d2a3484c505afa467706f2ae21fe17a292e"],"state_sha256":"0c046698981ea173e84424d0181c5ee3c7cb1ccee67ef0a1b72e2af4fb975000"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"i4LHPCbROiBNaQlfrm0klwH6RH2Y68Qf1j1OR0q6Q8ppXwbxw/fnp0I3TzsYR62hy2hNUHjaYgODBEMwfmLMCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T02:16:55.270755Z","bundle_sha256":"8281d5ea0815f29bdcedc3a4b9260cf3c661e0ac8e0fcbcaf3997712d780fc30"}}