{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:NEQVXOG2XQKNKQNLC45PPHLLV6","short_pith_number":"pith:NEQVXOG2","canonical_record":{"source":{"id":"2407.04528","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-05T14:16:47Z","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"title_canon_sha256":"786df197bcecb40b0b6867e7a84f5d5d4dae732cc28041a779746ff39a275174","abstract_canon_sha256":"a799e4c6a912d38ec1132fe8d9db6b01e23adde46671659aefa00aea9b80c6cd"},"schema_version":"1.0"},"canonical_sha256":"69215bb8dabc14d541ab173af79d6baf96c5bc00dafac7601b9e8dd2cfc5a47b","source":{"kind":"arxiv","id":"2407.04528","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.04528","created_at":"2026-07-05T09:25:45Z"},{"alias_kind":"arxiv_version","alias_value":"2407.04528v4","created_at":"2026-07-05T09:25:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.04528","created_at":"2026-07-05T09:25:45Z"},{"alias_kind":"pith_short_12","alias_value":"NEQVXOG2XQKN","created_at":"2026-07-05T09:25:45Z"},{"alias_kind":"pith_short_16","alias_value":"NEQVXOG2XQKNKQNL","created_at":"2026-07-05T09:25:45Z"},{"alias_kind":"pith_short_8","alias_value":"NEQVXOG2","created_at":"2026-07-05T09:25:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:NEQVXOG2XQKNKQNLC45PPHLLV6","target":"record","payload":{"canonical_record":{"source":{"id":"2407.04528","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-05T14:16:47Z","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"title_canon_sha256":"786df197bcecb40b0b6867e7a84f5d5d4dae732cc28041a779746ff39a275174","abstract_canon_sha256":"a799e4c6a912d38ec1132fe8d9db6b01e23adde46671659aefa00aea9b80c6cd"},"schema_version":"1.0"},"canonical_sha256":"69215bb8dabc14d541ab173af79d6baf96c5bc00dafac7601b9e8dd2cfc5a47b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:25:45.062131Z","signature_b64":"9YSyJaPgi11PCTyqyI/lGqvtJxujwbo/vtDRJG45Dbp33ZBuBmC9PUAkBrstlHgHqict3+MtTIF0OY0iiQpFDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"69215bb8dabc14d541ab173af79d6baf96c5bc00dafac7601b9e8dd2cfc5a47b","last_reissued_at":"2026-07-05T09:25:45.061640Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:25:45.061640Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2407.04528","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:25:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"r5FXvngD0JdcLXbSEf61hHBeHWI7tV+JH6pnxKfPH8SVNK+cf7UqiRTOmAqukFi2UIgmk8AovNRcrmJzvGR2CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T17:17:15.235922Z"},"content_sha256":"f46fd40c2168442f70d8476a7a065bed33fc735576f1c4a09a0803585d3520e6","schema_version":"1.0","event_id":"sha256:f46fd40c2168442f70d8476a7a065bed33fc735576f1c4a09a0803585d3520e6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:NEQVXOG2XQKNKQNLC45PPHLLV6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"GPT vs RETRO: Exploring the Intersection of Retrieval and Parameter-Efficient Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.IR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Aleksander Ficek, Jiaqi Zeng, Oleksii Kuchaiev","submitted_at":"2024-07-05T14:16:47Z","abstract_excerpt":"Parameter-Efficient Fine-Tuning (PEFT) and Retrieval-Augmented Generation (RAG) have become popular methods for adapting large language models while minimizing compute requirements. In this paper, we apply PEFT methods (P-tuning, Adapters, and LoRA) to a modified Retrieval-Enhanced Transformer (RETRO) and a baseline GPT model across several sizes, ranging from 823 million to 48 billion parameters. We show that RETRO models outperform GPT models in zero-shot settings due to their unique pre-training process but GPT models have higher performance potential with PEFT. Additionally, our study indi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.04528","kind":"arxiv","version":4},"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/2407.04528/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":"2026-07-05T09:25:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FsRek4eouzaMMbNi79csXCa9uT+BbT5ZH0Qs1ir+UGqTsfCZpSQxPKEQirj5lhfLMJedk2EP7Tlvz21T75gCCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T17:17:15.236418Z"},"content_sha256":"228563629cf5575b13c7c4dc4af1cd9a73fb2d0650452d93f1642c593dcc64f7","schema_version":"1.0","event_id":"sha256:228563629cf5575b13c7c4dc4af1cd9a73fb2d0650452d93f1642c593dcc64f7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NEQVXOG2XQKNKQNLC45PPHLLV6/bundle.json","state_url":"https://pith.science/pith/NEQVXOG2XQKNKQNLC45PPHLLV6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NEQVXOG2XQKNKQNLC45PPHLLV6/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":"2026-08-08T17:17:15Z","links":{"resolver":"https://pith.science/pith/NEQVXOG2XQKNKQNLC45PPHLLV6","bundle":"https://pith.science/pith/NEQVXOG2XQKNKQNLC45PPHLLV6/bundle.json","state":"https://pith.science/pith/NEQVXOG2XQKNKQNLC45PPHLLV6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NEQVXOG2XQKNKQNLC45PPHLLV6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:NEQVXOG2XQKNKQNLC45PPHLLV6","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":"a799e4c6a912d38ec1132fe8d9db6b01e23adde46671659aefa00aea9b80c6cd","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-05T14:16:47Z","title_canon_sha256":"786df197bcecb40b0b6867e7a84f5d5d4dae732cc28041a779746ff39a275174"},"schema_version":"1.0","source":{"id":"2407.04528","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.04528","created_at":"2026-07-05T09:25:45Z"},{"alias_kind":"arxiv_version","alias_value":"2407.04528v4","created_at":"2026-07-05T09:25:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.04528","created_at":"2026-07-05T09:25:45Z"},{"alias_kind":"pith_short_12","alias_value":"NEQVXOG2XQKN","created_at":"2026-07-05T09:25:45Z"},{"alias_kind":"pith_short_16","alias_value":"NEQVXOG2XQKNKQNL","created_at":"2026-07-05T09:25:45Z"},{"alias_kind":"pith_short_8","alias_value":"NEQVXOG2","created_at":"2026-07-05T09:25:45Z"}],"graph_snapshots":[{"event_id":"sha256:228563629cf5575b13c7c4dc4af1cd9a73fb2d0650452d93f1642c593dcc64f7","target":"graph","created_at":"2026-07-05T09:25:45Z","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/2407.04528/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Parameter-Efficient Fine-Tuning (PEFT) and Retrieval-Augmented Generation (RAG) have become popular methods for adapting large language models while minimizing compute requirements. In this paper, we apply PEFT methods (P-tuning, Adapters, and LoRA) to a modified Retrieval-Enhanced Transformer (RETRO) and a baseline GPT model across several sizes, ranging from 823 million to 48 billion parameters. We show that RETRO models outperform GPT models in zero-shot settings due to their unique pre-training process but GPT models have higher performance potential with PEFT. Additionally, our study indi","authors_text":"Aleksander Ficek, Jiaqi Zeng, Oleksii Kuchaiev","cross_cats":["cs.AI","cs.IR","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-05T14:16:47Z","title":"GPT vs RETRO: Exploring the Intersection of Retrieval and Parameter-Efficient Fine-Tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.04528","kind":"arxiv","version":4},"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:f46fd40c2168442f70d8476a7a065bed33fc735576f1c4a09a0803585d3520e6","target":"record","created_at":"2026-07-05T09:25:45Z","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":"a799e4c6a912d38ec1132fe8d9db6b01e23adde46671659aefa00aea9b80c6cd","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-05T14:16:47Z","title_canon_sha256":"786df197bcecb40b0b6867e7a84f5d5d4dae732cc28041a779746ff39a275174"},"schema_version":"1.0","source":{"id":"2407.04528","kind":"arxiv","version":4}},"canonical_sha256":"69215bb8dabc14d541ab173af79d6baf96c5bc00dafac7601b9e8dd2cfc5a47b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"69215bb8dabc14d541ab173af79d6baf96c5bc00dafac7601b9e8dd2cfc5a47b","first_computed_at":"2026-07-05T09:25:45.061640Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:25:45.061640Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9YSyJaPgi11PCTyqyI/lGqvtJxujwbo/vtDRJG45Dbp33ZBuBmC9PUAkBrstlHgHqict3+MtTIF0OY0iiQpFDw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:25:45.062131Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.04528","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f46fd40c2168442f70d8476a7a065bed33fc735576f1c4a09a0803585d3520e6","sha256:228563629cf5575b13c7c4dc4af1cd9a73fb2d0650452d93f1642c593dcc64f7"],"state_sha256":"26c1fcf1acaa356c928524356e9b5133ebc26be8cb931caf2c4fb53ba3841766"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7N3jO+8X6cT8jjZubx6VyWWE7A7jcsubV3VPTPgHMkEu7RTDipVVfhpMz4MPKK6m/qRGe6TBGXziryrW28XOCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T17:17:15.240187Z","bundle_sha256":"7085eb87a3326f76c60ddad0183047b70c164ae2496286d3a04ec24d7838a6f8"}}