{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:V5LIG6OZF5PER336NLLXVNRM3S","short_pith_number":"pith:V5LIG6OZ","canonical_record":{"source":{"id":"2605.13880","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T04:34:43Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"f33fac511cef95d8d740b902d062a69cd90aca34d371ab437bb74f51cd762dea","abstract_canon_sha256":"cff4abde4570b51cd1946a9a209b315530b2fe54f925b569f88b961b7b457649"},"schema_version":"1.0"},"canonical_sha256":"af568379d92f5e48ef7e6ad77ab62cdca2dbc4dc0147d965637b81954cb16714","source":{"kind":"arxiv","id":"2605.13880","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13880","created_at":"2026-05-17T23:39:19Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13880v1","created_at":"2026-05-17T23:39:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13880","created_at":"2026-05-17T23:39:19Z"},{"alias_kind":"pith_short_12","alias_value":"V5LIG6OZF5PE","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"V5LIG6OZF5PER336","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"V5LIG6OZ","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:V5LIG6OZF5PER336NLLXVNRM3S","target":"record","payload":{"canonical_record":{"source":{"id":"2605.13880","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T04:34:43Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"f33fac511cef95d8d740b902d062a69cd90aca34d371ab437bb74f51cd762dea","abstract_canon_sha256":"cff4abde4570b51cd1946a9a209b315530b2fe54f925b569f88b961b7b457649"},"schema_version":"1.0"},"canonical_sha256":"af568379d92f5e48ef7e6ad77ab62cdca2dbc4dc0147d965637b81954cb16714","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:19.198215Z","signature_b64":"oHgD6sYlvFwe/H32cLb1UMfbazvp7HScGMnXnfVG53ciFECcJalsIhQm7jX0teSNuxU1AKTWFIjgulFhfHtUDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"af568379d92f5e48ef7e6ad77ab62cdca2dbc4dc0147d965637b81954cb16714","last_reissued_at":"2026-05-17T23:39:19.197500Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:19.197500Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.13880","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":"2026-05-17T23:39:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZK4DlJEVepqvOpdw3WbPOGsDcjARj05m3X/uUZTQCPuyFB/JlPUQH1zWj94Lb7NX8srVcFqx94xddv74NQGJCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T16:50:30.335747Z"},"content_sha256":"659e51e1abf561f3987e50a681d58675377df78d0614e7b942fbbc4433ba8eff","schema_version":"1.0","event_id":"sha256:659e51e1abf561f3987e50a681d58675377df78d0614e7b942fbbc4433ba8eff"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:V5LIG6OZF5PER336NLLXVNRM3S","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PREPING: Building Agent Memory without Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Agents can construct competitive procedural memory for new environments using only self-generated synthetic tasks before any real experience.","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Jinheon Baek, Minki Kang, Sangwoo Park, Sung Ju Hwang, Yumin Choi","submitted_at":"2026-05-11T04:34:43Z","abstract_excerpt":"Agent memory is typically constructed either offline from curated demonstrations or online from post-deployment interactions. However, regardless of how it is built, an agent faces a cold-start gap when first introduced to a new environment without any task-specific experience available. In this paper, we study pre-task memory construction: whether an agent can build procedural memory before observing any target-environment tasks, using only self-generated synthetic practice. Yet, synthetic interaction alone is insufficient, as without controlling what to practice and what to store, synthetic "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Preping substantially improves over a no-memory baseline and achieves performance competitive with strong playbook-based methods built from offline or online experience, with deployment cost $2.99× lower on AppWorld and $2.23× lower on BFCL v3 than online memory construction.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That synthetic tasks generated and filtered by the proposer-validator loop will transfer to real target-environment tasks without any direct experience of those tasks.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Preping builds agent memory via proposer-guided synthetic practice and selective validation, matching offline/online methods at 2-3x lower deployment cost.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Agents can construct competitive procedural memory for new environments using only self-generated synthetic tasks before any real experience.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"6d51bf3eb1a40f87addfa0a6437e69e7717fa2c8358465e2b5b4b19fce4c94d6"},"source":{"id":"2605.13880","kind":"arxiv","version":1},"verdict":{"id":"11a37770-dc45-4a86-9c31-d818ca02939b","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T06:10:55.020382Z","strongest_claim":"Preping substantially improves over a no-memory baseline and achieves performance competitive with strong playbook-based methods built from offline or online experience, with deployment cost $2.99× lower on AppWorld and $2.23× lower on BFCL v3 than online memory construction.","one_line_summary":"Preping builds agent memory via proposer-guided synthetic practice and selective validation, matching offline/online methods at 2-3x lower deployment cost.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That synthetic tasks generated and filtered by the proposer-validator loop will transfer to real target-environment tasks without any direct experience of those tasks.","pith_extraction_headline":"Agents can construct competitive procedural memory for new environments using only self-generated synthetic tasks before any real experience."},"references":{"count":53,"sample":[{"doi":"10.48550/arxiv.2602.21320","year":2026,"title":"Tool-r0: Self-evolving llm agents for tool-learning from zero data","work_id":"682365bd-b781-4d4c-a78a-a0351adbdc31","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Introducing the model context protocol, November 2024","work_id":"f0cdfef8-384f-4258-b107-3db84142d5ca","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"DeepSeek-V3.2: Pushing the frontier of open large language models, 2025","work_id":"812ba0f2-05ea-49f6-a758-c2ae6d0fc49f","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.48550/arxiv.2510.04851","year":2025,"title":"LEGOMem: Modular Procedural Memory for Multi-agent LLM Systems for Workflow Automation","work_id":"81b5a1e3-7f36-4854-a305-10e386c1b07f","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.48550/arxiv.2508.05004","year":2025,"title":"R-Zero: Self-Evolving Reasoning LLM from Zero Data","work_id":"45b39aae-1151-4087-9b48-dd16313e6306","ref_index":5,"cited_arxiv_id":"2508.05004","is_internal_anchor":true}],"resolved_work":53,"snapshot_sha256":"320ef9a362b9f83b5cba69ff837dea30c9ab8274d36be88aa5764a1597a07c53","internal_anchors":7},"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":"11a37770-dc45-4a86-9c31-d818ca02939b"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sZvCg4QcExb9RfeX1uUXHS6zm2XDRbo7j42QLO/YDx3kN//wT7xssrmfjsgGBeK2BgBZomv83q9HSljOZMIACQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T16:50:30.336353Z"},"content_sha256":"372f8115648ac64b9af005a538f7804c04574b81393349352c41e135c8082df5","schema_version":"1.0","event_id":"sha256:372f8115648ac64b9af005a538f7804c04574b81393349352c41e135c8082df5"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:V5LIG6OZF5PER336NLLXVNRM3S","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.18653/v1/2024.acl-long.850.URLhttps://aclanthology.org/2024.acl-long.850/) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Harsh Trivedi, Tushar Khot, Mareike Hartmann, Ruskin Manku, Vinty Dong, Edward Li, Shashank Gupta, Ashish Sabharwal, and Niranjan Balasubramanian. Appworld: A controllable world of apps and people for benchmarking interactive coding agents.","arxiv_id":"2605.13880","detector":"doi_compliance","evidence":{"ref_index":25,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"Harsh Trivedi, Tushar Khot, Mareike Hartmann, Ruskin Manku, Vinty Dong, Edward Li, Shashank Gupta, Ashish Sabharwal, and Niranjan Balasubramanian. Appworld: A controllable world of apps and people for benchmarking interactive coding agents.","reconstructed_doi":"10.18653/v1/2024.acl-long.850.URLhttps://aclanthology.org/2024.acl-long.850/"},"severity":"advisory","ref_index":25,"audited_at":"2026-05-19T09:47:59.513531Z","event_type":"pith.integrity.v1","detected_doi":"10.18653/v1/2024.acl-long.850.URLhttps://aclanthology.org/2024.acl-long.850/","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"1ea342ddbb10dc5c2bc8e083d9290df119cd114e6a36256d3113437fdff387c3","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.0.0","detected_arxiv_id":null,"integrity_event_id":679,"payload_sha256":"baa653412683ca00683ca970a904cdae7a8d2d3b4e7f0e7c963e2343f03e6cca","signature_b64":"wngN/q8YK/z33x19aRQt0VUPwkH7Dg9u7rdfV6OjwD6DPvCw0uTFxLcHH1GAHdldTauHq/H9qaPoUvODV+9DDA==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-19T09:51:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wVIp4i+8irAwToyHD12JK9YBgTY06nveIRujycUv2xp4twhQtU7MD9gGXsJBchnxgZlaNi761/NbDu9a2qnWAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T16:50:30.337526Z"},"content_sha256":"548b973e1845378a0e83f10ab2392f5527ad148f272591dd7d33d4a832e914e0","schema_version":"1.0","event_id":"sha256:548b973e1845378a0e83f10ab2392f5527ad148f272591dd7d33d4a832e914e0"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:V5LIG6OZF5PER336NLLXVNRM3S","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.48550/arXiv.2504.07952.URLhttps://arxiv.org/abs/2504.07952) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Mirac Suzgun, Mert Yüksekgönul, Federico Bianchi, Dan Jurafsky, and James Zou. Dynamic cheatsheet: Test-time learning with adaptive memory.arXiv, 2025. doi: 10.48550/arXiv.2504. 07952. URLhttps://arxiv.org/abs/2504.07952","arxiv_id":"2605.13880","detector":"doi_compliance","evidence":{"ref_index":23,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"Mirac Suzgun, Mert Yüksekgönul, Federico Bianchi, Dan Jurafsky, and James Zou. Dynamic cheatsheet: Test-time learning with adaptive memory.arXiv, 2025. doi: 10.48550/arXiv.2504. 07952. URLhttps://arxiv.org/abs/2504.07952","reconstructed_doi":"10.48550/arXiv.2504.07952.URLhttps://arxiv.org/abs/2504.07952"},"severity":"advisory","ref_index":23,"audited_at":"2026-05-19T09:47:59.513531Z","event_type":"pith.integrity.v1","detected_doi":"10.48550/arXiv.2504.07952.URLhttps://arxiv.org/abs/2504.07952","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"db375d654d18ba32df13e7af2826be31bd6c80af8a314612b1f128e44ff2a87b","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.0.0","detected_arxiv_id":null,"integrity_event_id":678,"payload_sha256":"2b27a820f6e5d23d9f523c8d55df02473e38a67cce682d95c46a84cbca9bff45","signature_b64":"yGl/EGJgtPYUtuPm1g23RAsPyr62SHxa9AFU7gKLCXhYGzZ60VCJUXVJqm87T58fko0ABiHOonIaLHDmAz3HCg==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-19T09:51:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"71HYGDkR+bOlaaDedNXF9LpgR1cjl52HSjz+tWybanxgivQ167r1nIHV3QL4dhsXXARufdQJZfwYmS5iohkFAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T16:50:30.337839Z"},"content_sha256":"be93ddb88081722dad1ad1dd82caaa6c1f6ddaa4ea5aa9df40954847c222befc","schema_version":"1.0","event_id":"sha256:be93ddb88081722dad1ad1dd82caaa6c1f6ddaa4ea5aa9df40954847c222befc"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:V5LIG6OZF5PER336NLLXVNRM3S","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.48550/arXiv.2310.10134.URLhttps://arxiv.org/abs/2310.10134) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Bodhisattwa Prasad Majumder, Bhavana Dalvi Mishra, Peter Jansen, Oyvind Tafjord, Niket Tandon, Li Zhang, Chris Callison-Burch, and Peter Clark. CLIN: A continually learning language agent for rapid task adaptation and generalization.arXiv, ","arxiv_id":"2605.13880","detector":"doi_compliance","evidence":{"ref_index":9,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"Bodhisattwa Prasad Majumder, Bhavana Dalvi Mishra, Peter Jansen, Oyvind Tafjord, Niket Tandon, Li Zhang, Chris Callison-Burch, and Peter Clark. CLIN: A continually learning language agent for rapid task adaptation and generalization.arXiv, ","reconstructed_doi":"10.48550/arXiv.2310.10134.URLhttps://arxiv.org/abs/2310.10134"},"severity":"advisory","ref_index":9,"audited_at":"2026-05-19T09:47:59.513531Z","event_type":"pith.integrity.v1","detected_doi":"10.48550/arXiv.2310.10134.URLhttps://arxiv.org/abs/2310.10134","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"b943619eaf60d084cf87aaf3f97c5619039a67ca827c84e4d7042c3bbd823f16","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.0.0","detected_arxiv_id":null,"integrity_event_id":677,"payload_sha256":"badb1faa1d94103f510afccf23b8bf7a6a04749747aae8670325a7871857826d","signature_b64":"mph/X7HEovmUxoKOP3n7Z5F1Xy654DOwSrrZneSVst795EMqmTP/heYDpnI8+MGTIXH6NuXQr4L0sYzKeMtxBg==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-19T09:51:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9WTPbFStjfVkCTbNzZnyt0NAScOBnA9xqHgwqkhmReY1hFvsACFnXrW7M1wpP0yGId+DPjGvoXBlbmUquXn6Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T16:50:30.338365Z"},"content_sha256":"248434f41ef39110b7ada3d0fcc9a9cc2efef10f6b5df7c2466c273e3afe61ab","schema_version":"1.0","event_id":"sha256:248434f41ef39110b7ada3d0fcc9a9cc2efef10f6b5df7c2466c273e3afe61ab"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/V5LIG6OZF5PER336NLLXVNRM3S/bundle.json","state_url":"https://pith.science/pith/V5LIG6OZF5PER336NLLXVNRM3S/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/V5LIG6OZF5PER336NLLXVNRM3S/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":"2026-05-21T16:50:30Z","links":{"resolver":"https://pith.science/pith/V5LIG6OZF5PER336NLLXVNRM3S","bundle":"https://pith.science/pith/V5LIG6OZF5PER336NLLXVNRM3S/bundle.json","state":"https://pith.science/pith/V5LIG6OZF5PER336NLLXVNRM3S/state.json","well_known_bundle":"https://pith.science/.well-known/pith/V5LIG6OZF5PER336NLLXVNRM3S/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:V5LIG6OZF5PER336NLLXVNRM3S","merge_version":"pith-open-graph-merge-v1","event_count":5,"valid_event_count":5,"invalid_event_count":0,"equivocation_count":1,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"cff4abde4570b51cd1946a9a209b315530b2fe54f925b569f88b961b7b457649","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T04:34:43Z","title_canon_sha256":"f33fac511cef95d8d740b902d062a69cd90aca34d371ab437bb74f51cd762dea"},"schema_version":"1.0","source":{"id":"2605.13880","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13880","created_at":"2026-05-17T23:39:19Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13880v1","created_at":"2026-05-17T23:39:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13880","created_at":"2026-05-17T23:39:19Z"},{"alias_kind":"pith_short_12","alias_value":"V5LIG6OZF5PE","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"V5LIG6OZF5PER336","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"V5LIG6OZ","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:372f8115648ac64b9af005a538f7804c04574b81393349352c41e135c8082df5","target":"graph","created_at":"2026-05-17T23:39:19Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Preping substantially improves over a no-memory baseline and achieves performance competitive with strong playbook-based methods built from offline or online experience, with deployment cost $2.99× lower on AppWorld and $2.23× lower on BFCL v3 than online memory construction."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That synthetic tasks generated and filtered by the proposer-validator loop will transfer to real target-environment tasks without any direct experience of those tasks."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Preping builds agent memory via proposer-guided synthetic practice and selective validation, matching offline/online methods at 2-3x lower deployment cost."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Agents can construct competitive procedural memory for new environments using only self-generated synthetic tasks before any real experience."}],"snapshot_sha256":"6d51bf3eb1a40f87addfa0a6437e69e7717fa2c8358465e2b5b4b19fce4c94d6"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Agent memory is typically constructed either offline from curated demonstrations or online from post-deployment interactions. However, regardless of how it is built, an agent faces a cold-start gap when first introduced to a new environment without any task-specific experience available. In this paper, we study pre-task memory construction: whether an agent can build procedural memory before observing any target-environment tasks, using only self-generated synthetic practice. Yet, synthetic interaction alone is insufficient, as without controlling what to practice and what to store, synthetic ","authors_text":"Jinheon Baek, Minki Kang, Sangwoo Park, Sung Ju Hwang, Yumin Choi","cross_cats":["cs.CL"],"headline":"Agents can construct competitive procedural memory for new environments using only self-generated synthetic tasks before any real experience.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T04:34:43Z","title":"PREPING: Building Agent Memory without Tasks"},"references":{"count":53,"internal_anchors":7,"resolved_work":53,"sample":[{"cited_arxiv_id":"","doi":"10.48550/arxiv.2602.21320","is_internal_anchor":false,"ref_index":1,"title":"Tool-r0: Self-evolving llm agents for tool-learning from zero data","work_id":"682365bd-b781-4d4c-a78a-a0351adbdc31","year":2026},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Introducing the model context protocol, November 2024","work_id":"f0cdfef8-384f-4258-b107-3db84142d5ca","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"DeepSeek-V3.2: Pushing the frontier of open large language models, 2025","work_id":"812ba0f2-05ea-49f6-a758-c2ae6d0fc49f","year":2025},{"cited_arxiv_id":"","doi":"10.48550/arxiv.2510.04851","is_internal_anchor":false,"ref_index":4,"title":"LEGOMem: Modular Procedural Memory for Multi-agent LLM Systems for Workflow Automation","work_id":"81b5a1e3-7f36-4854-a305-10e386c1b07f","year":2025},{"cited_arxiv_id":"2508.05004","doi":"10.48550/arxiv.2508.05004","is_internal_anchor":true,"ref_index":5,"title":"R-Zero: Self-Evolving Reasoning LLM from Zero Data","work_id":"45b39aae-1151-4087-9b48-dd16313e6306","year":2025}],"snapshot_sha256":"320ef9a362b9f83b5cba69ff837dea30c9ab8274d36be88aa5764a1597a07c53"},"source":{"id":"2605.13880","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T06:10:55.020382Z","id":"11a37770-dc45-4a86-9c31-d818ca02939b","model_set":{"reader":"grok-4.3"},"one_line_summary":"Preping builds agent memory via proposer-guided synthetic practice and selective validation, matching offline/online methods at 2-3x lower deployment cost.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Agents can construct competitive procedural memory for new environments using only self-generated synthetic tasks before any real experience.","strongest_claim":"Preping substantially improves over a no-memory baseline and achieves performance competitive with strong playbook-based methods built from offline or online experience, with deployment cost $2.99× lower on AppWorld and $2.23× lower on BFCL v3 than online memory construction.","weakest_assumption":"That synthetic tasks generated and filtered by the proposer-validator loop will transfer to real target-environment tasks without any direct experience of those tasks."}},"verdict_id":"11a37770-dc45-4a86-9c31-d818ca02939b"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:659e51e1abf561f3987e50a681d58675377df78d0614e7b942fbbc4433ba8eff","target":"record","created_at":"2026-05-17T23:39:19Z","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":"cff4abde4570b51cd1946a9a209b315530b2fe54f925b569f88b961b7b457649","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T04:34:43Z","title_canon_sha256":"f33fac511cef95d8d740b902d062a69cd90aca34d371ab437bb74f51cd762dea"},"schema_version":"1.0","source":{"id":"2605.13880","kind":"arxiv","version":1}},"canonical_sha256":"af568379d92f5e48ef7e6ad77ab62cdca2dbc4dc0147d965637b81954cb16714","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"af568379d92f5e48ef7e6ad77ab62cdca2dbc4dc0147d965637b81954cb16714","first_computed_at":"2026-05-17T23:39:19.197500Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:19.197500Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oHgD6sYlvFwe/H32cLb1UMfbazvp7HScGMnXnfVG53ciFECcJalsIhQm7jX0teSNuxU1AKTWFIjgulFhfHtUDA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:19.198215Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.13880","source_kind":"arxiv","source_version":1}}},"equivocations":[{"signer_id":"pith.science","event_type":"integrity_finding","target":"integrity","event_ids":["sha256:248434f41ef39110b7ada3d0fcc9a9cc2efef10f6b5df7c2466c273e3afe61ab","sha256:548b973e1845378a0e83f10ab2392f5527ad148f272591dd7d33d4a832e914e0","sha256:be93ddb88081722dad1ad1dd82caaa6c1f6ddaa4ea5aa9df40954847c222befc"]}],"invalid_events":[],"applied_event_ids":["sha256:659e51e1abf561f3987e50a681d58675377df78d0614e7b942fbbc4433ba8eff","sha256:372f8115648ac64b9af005a538f7804c04574b81393349352c41e135c8082df5"],"state_sha256":"38a545c80224bfd0cdab6bbdac93830f63ac32721202586038e1cf9573a4e337"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pNR6JsFjhRcUjwuzZmaARcpcpRHVoANWxrTfOryWVWiiAPEa1cPGRKd/EOlQ5nPSKTGdJXxIPHzosjp5TOj3Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-21T16:50:30.341761Z","bundle_sha256":"56347eed9b8ced728999124a6884676409829bb9a7b3e2f32f7d8ceb9053ae7b"}}