{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:ZIAW3XPSPVXSEZRWZ4UZCULMLH","short_pith_number":"pith:ZIAW3XPS","canonical_record":{"source":{"id":"2312.11671","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-12-18T19:27:09Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"38e6c9bdb74fd3a0cc2a9488a9f09bb7aa517a92ab0313c2a8fc5b8b0ceae7b9","abstract_canon_sha256":"d65c00b912072f3b5d1bcce8d0eb4054ef256c724dbc180c1517d77b18d95da3"},"schema_version":"1.0"},"canonical_sha256":"ca016dddf27d6f226636cf2991516c59d1ac69270a1bdb9340aadd15e790349c","source":{"kind":"arxiv","id":"2312.11671","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.11671","created_at":"2026-07-05T07:30:06Z"},{"alias_kind":"arxiv_version","alias_value":"2312.11671v2","created_at":"2026-07-05T07:30:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.11671","created_at":"2026-07-05T07:30:06Z"},{"alias_kind":"pith_short_12","alias_value":"ZIAW3XPSPVXS","created_at":"2026-07-05T07:30:06Z"},{"alias_kind":"pith_short_16","alias_value":"ZIAW3XPSPVXSEZRW","created_at":"2026-07-05T07:30:06Z"},{"alias_kind":"pith_short_8","alias_value":"ZIAW3XPS","created_at":"2026-07-05T07:30:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:ZIAW3XPSPVXSEZRWZ4UZCULMLH","target":"record","payload":{"canonical_record":{"source":{"id":"2312.11671","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-12-18T19:27:09Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"38e6c9bdb74fd3a0cc2a9488a9f09bb7aa517a92ab0313c2a8fc5b8b0ceae7b9","abstract_canon_sha256":"d65c00b912072f3b5d1bcce8d0eb4054ef256c724dbc180c1517d77b18d95da3"},"schema_version":"1.0"},"canonical_sha256":"ca016dddf27d6f226636cf2991516c59d1ac69270a1bdb9340aadd15e790349c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:30:06.887571Z","signature_b64":"Qw8sno9/knOUv4MlXHKP2O/ITxAy1s6iNJ04nxoJk4lYTybRiWf2mXKzU8Q52FXNiM8KLoox5dpE5YPgeyg+BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ca016dddf27d6f226636cf2991516c59d1ac69270a1bdb9340aadd15e790349c","last_reissued_at":"2026-07-05T07:30:06.887084Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:30:06.887084Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2312.11671","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":"2026-07-05T07:30:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Rtcwa1FUO0p18LjOerKY6Oq18IT+ghoMuQq+s6L7MOJmcEB6MJpuES96zZ7L92aLn2xoogDnmrSGT256cVJTCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:54:04.719640Z"},"content_sha256":"0207d061beccd27df38f7b58bb0475e7f693beca036b909e94799b39f9adcb41","schema_version":"1.0","event_id":"sha256:0207d061beccd27df38f7b58bb0475e7f693beca036b909e94799b39f9adcb41"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:ZIAW3XPSPVXSEZRWZ4UZCULMLH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluating Language-Model Agents on Realistic Autonomous Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Aaron Ho, Brian Goodrich, Elizabeth Barnes, Haoxing Du, Hjalmar Wijk, Joel Burget, Lawrence Chan, Lucas Jun Koba Sato, Luke Harold Miles, Max Hasin, Megan Kinniment, Paul Christiano, Tao R. Lin","submitted_at":"2023-12-18T19:27:09Z","abstract_excerpt":"In this report, we explore the ability of language model agents to acquire resources, create copies of themselves, and adapt to novel challenges they encounter in the wild. We refer to this cluster of capabilities as \"autonomous replication and adaptation\" or ARA. We believe that systems capable of ARA could have wide-reaching and hard-to-anticipate consequences, and that measuring and forecasting ARA may be useful for informing measures around security, monitoring, and alignment. Additionally, once a system is capable of ARA, placing bounds on a system's capabilities may become significantly "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.11671","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/2312.11671/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":"2026-07-05T07:30:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"diTDX//Fx3BbZgWWqkgWFxB08w0EaTIBtNxeFh+HF78xp2gxGu9MIVO3x/bUtyp550W4ib40BnhSZ51be5zKCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:54:04.720016Z"},"content_sha256":"c55872096aa6bfa457f5252d3cd8dfd795672ff1bd51d3ce9a2dc13f43d15646","schema_version":"1.0","event_id":"sha256:c55872096aa6bfa457f5252d3cd8dfd795672ff1bd51d3ce9a2dc13f43d15646"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZIAW3XPSPVXSEZRWZ4UZCULMLH/bundle.json","state_url":"https://pith.science/pith/ZIAW3XPSPVXSEZRWZ4UZCULMLH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZIAW3XPSPVXSEZRWZ4UZCULMLH/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":"2026-07-07T06:54:04Z","links":{"resolver":"https://pith.science/pith/ZIAW3XPSPVXSEZRWZ4UZCULMLH","bundle":"https://pith.science/pith/ZIAW3XPSPVXSEZRWZ4UZCULMLH/bundle.json","state":"https://pith.science/pith/ZIAW3XPSPVXSEZRWZ4UZCULMLH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZIAW3XPSPVXSEZRWZ4UZCULMLH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:ZIAW3XPSPVXSEZRWZ4UZCULMLH","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":"d65c00b912072f3b5d1bcce8d0eb4054ef256c724dbc180c1517d77b18d95da3","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-12-18T19:27:09Z","title_canon_sha256":"38e6c9bdb74fd3a0cc2a9488a9f09bb7aa517a92ab0313c2a8fc5b8b0ceae7b9"},"schema_version":"1.0","source":{"id":"2312.11671","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.11671","created_at":"2026-07-05T07:30:06Z"},{"alias_kind":"arxiv_version","alias_value":"2312.11671v2","created_at":"2026-07-05T07:30:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.11671","created_at":"2026-07-05T07:30:06Z"},{"alias_kind":"pith_short_12","alias_value":"ZIAW3XPSPVXS","created_at":"2026-07-05T07:30:06Z"},{"alias_kind":"pith_short_16","alias_value":"ZIAW3XPSPVXSEZRW","created_at":"2026-07-05T07:30:06Z"},{"alias_kind":"pith_short_8","alias_value":"ZIAW3XPS","created_at":"2026-07-05T07:30:06Z"}],"graph_snapshots":[{"event_id":"sha256:c55872096aa6bfa457f5252d3cd8dfd795672ff1bd51d3ce9a2dc13f43d15646","target":"graph","created_at":"2026-07-05T07:30:06Z","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/2312.11671/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In this report, we explore the ability of language model agents to acquire resources, create copies of themselves, and adapt to novel challenges they encounter in the wild. We refer to this cluster of capabilities as \"autonomous replication and adaptation\" or ARA. We believe that systems capable of ARA could have wide-reaching and hard-to-anticipate consequences, and that measuring and forecasting ARA may be useful for informing measures around security, monitoring, and alignment. Additionally, once a system is capable of ARA, placing bounds on a system's capabilities may become significantly ","authors_text":"Aaron Ho, Brian Goodrich, Elizabeth Barnes, Haoxing Du, Hjalmar Wijk, Joel Burget, Lawrence Chan, Lucas Jun Koba Sato, Luke Harold Miles, Max Hasin, Megan Kinniment, Paul Christiano, Tao R. Lin","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-12-18T19:27:09Z","title":"Evaluating Language-Model Agents on Realistic Autonomous Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.11671","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":"sha256:0207d061beccd27df38f7b58bb0475e7f693beca036b909e94799b39f9adcb41","target":"record","created_at":"2026-07-05T07:30:06Z","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":"d65c00b912072f3b5d1bcce8d0eb4054ef256c724dbc180c1517d77b18d95da3","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-12-18T19:27:09Z","title_canon_sha256":"38e6c9bdb74fd3a0cc2a9488a9f09bb7aa517a92ab0313c2a8fc5b8b0ceae7b9"},"schema_version":"1.0","source":{"id":"2312.11671","kind":"arxiv","version":2}},"canonical_sha256":"ca016dddf27d6f226636cf2991516c59d1ac69270a1bdb9340aadd15e790349c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ca016dddf27d6f226636cf2991516c59d1ac69270a1bdb9340aadd15e790349c","first_computed_at":"2026-07-05T07:30:06.887084Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:30:06.887084Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Qw8sno9/knOUv4MlXHKP2O/ITxAy1s6iNJ04nxoJk4lYTybRiWf2mXKzU8Q52FXNiM8KLoox5dpE5YPgeyg+BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:30:06.887571Z","signed_message":"canonical_sha256_bytes"},"source_id":"2312.11671","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0207d061beccd27df38f7b58bb0475e7f693beca036b909e94799b39f9adcb41","sha256:c55872096aa6bfa457f5252d3cd8dfd795672ff1bd51d3ce9a2dc13f43d15646"],"state_sha256":"0d29dffda808ed446af3fd3fec7e677b1df1aa41cf228666e385ed2e6f577c83"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5hH77TQ5CBTvfZUxYCb1bDAdc0zFpoiMvC7U4W5YDvA5dJTsbKeFwx9IzNc0F22ej4m5S2YrpZXoXCbq+QQBCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T06:54:04.722049Z","bundle_sha256":"21a19b7d1c0243b3c2e42360391750df54618109f94905fc85c3c77985043e88"}}