{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:HREEYD5XMZZCLKUZX7P2U3453C","short_pith_number":"pith:HREEYD5X","canonical_record":{"source":{"id":"2506.01716","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T14:23:33Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"17c29d89e9e65373e1cdb827990aa112bebd39e96390dc2bf2d25b597cf740c5","abstract_canon_sha256":"9c981448de726c7c4bbf0e506b85b26f62f1b60ff2ae37dcf12b950463506b9b"},"schema_version":"1.0"},"canonical_sha256":"3c484c0fb7667225aa99bfdfaa6f9dd884debc5ff776f0937499018b27e63182","source":{"kind":"arxiv","id":"2506.01716","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.01716","created_at":"2026-07-05T11:14:18Z"},{"alias_kind":"arxiv_version","alias_value":"2506.01716v1","created_at":"2026-07-05T11:14:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01716","created_at":"2026-07-05T11:14:18Z"},{"alias_kind":"pith_short_12","alias_value":"HREEYD5XMZZC","created_at":"2026-07-05T11:14:18Z"},{"alias_kind":"pith_short_16","alias_value":"HREEYD5XMZZCLKUZ","created_at":"2026-07-05T11:14:18Z"},{"alias_kind":"pith_short_8","alias_value":"HREEYD5X","created_at":"2026-07-05T11:14:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:HREEYD5XMZZCLKUZX7P2U3453C","target":"record","payload":{"canonical_record":{"source":{"id":"2506.01716","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T14:23:33Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"17c29d89e9e65373e1cdb827990aa112bebd39e96390dc2bf2d25b597cf740c5","abstract_canon_sha256":"9c981448de726c7c4bbf0e506b85b26f62f1b60ff2ae37dcf12b950463506b9b"},"schema_version":"1.0"},"canonical_sha256":"3c484c0fb7667225aa99bfdfaa6f9dd884debc5ff776f0937499018b27e63182","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:18.282421Z","signature_b64":"T41UBOcctBAv0BXYnQ86Iquwb1GDffEedv8+4rZX5s9oTT4taToXcdQpvFPmJizTPe6TAEATLaKWWrPLDDtrAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3c484c0fb7667225aa99bfdfaa6f9dd884debc5ff776f0937499018b27e63182","last_reissued_at":"2026-07-05T11:14:18.281933Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:18.281933Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.01716","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":"2026-07-05T11:14:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/WY0uQaKVNifHGcI2IooRPI9Ux/UOv8G57oxltuuyxrUJy4yXXCRg2hHkE8wawIkVgmsbQHddRMosrQBt6FbDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-02T20:54:40.768155Z"},"content_sha256":"b95cbbbf8f3cdd1185a71b0219bd76e53ae2aaac8cd2930ba42703fe03b10f1d","schema_version":"1.0","event_id":"sha256:b95cbbbf8f3cdd1185a71b0219bd76e53ae2aaac8cd2930ba42703fe03b10f1d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:HREEYD5XMZZCLKUZX7P2U3453C","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Self-Challenging Language Model Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Jason Weston, Sainbayar Sukhbaatar, Sergey Levine, Xian Li, Yifei Zhou","submitted_at":"2025-06-02T14:23:33Z","abstract_excerpt":"Large language models are quickly becoming the foundation for intelligent agents that are capable of using tools. However, training such agents is challenging because it requires human creation and annotation of a diverse set of tasks, tools, and evaluation criteria. In this paper, we propose the Self-Challenging framework for training an agent on high-quality tasks that are generated by itself. The agent first plays the role of challenger and generates a task after interacting with the given tools. The tasks take the form of a novel general class of problems termed Code-as-Task, which are def"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01716","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/2506.01716/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":"2026-07-05T11:14:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WOk46/V5o6wH6X3ldT65x7HN75zibhmbtapzcoo3u3e2DciNmYp+HvPW7c2ZuqF8mQ2WolsRSOkuTDuKVAecCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-02T20:54:40.768815Z"},"content_sha256":"283ff7d924c10b05cc064275d12d64d6e1c2b6dafb336827e70386bdb7476bac","schema_version":"1.0","event_id":"sha256:283ff7d924c10b05cc064275d12d64d6e1c2b6dafb336827e70386bdb7476bac"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HREEYD5XMZZCLKUZX7P2U3453C/bundle.json","state_url":"https://pith.science/pith/HREEYD5XMZZCLKUZX7P2U3453C/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HREEYD5XMZZCLKUZX7P2U3453C/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":"2026-08-02T20:54:40Z","links":{"resolver":"https://pith.science/pith/HREEYD5XMZZCLKUZX7P2U3453C","bundle":"https://pith.science/pith/HREEYD5XMZZCLKUZX7P2U3453C/bundle.json","state":"https://pith.science/pith/HREEYD5XMZZCLKUZX7P2U3453C/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HREEYD5XMZZCLKUZX7P2U3453C/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:HREEYD5XMZZCLKUZX7P2U3453C","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":"9c981448de726c7c4bbf0e506b85b26f62f1b60ff2ae37dcf12b950463506b9b","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T14:23:33Z","title_canon_sha256":"17c29d89e9e65373e1cdb827990aa112bebd39e96390dc2bf2d25b597cf740c5"},"schema_version":"1.0","source":{"id":"2506.01716","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.01716","created_at":"2026-07-05T11:14:18Z"},{"alias_kind":"arxiv_version","alias_value":"2506.01716v1","created_at":"2026-07-05T11:14:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01716","created_at":"2026-07-05T11:14:18Z"},{"alias_kind":"pith_short_12","alias_value":"HREEYD5XMZZC","created_at":"2026-07-05T11:14:18Z"},{"alias_kind":"pith_short_16","alias_value":"HREEYD5XMZZCLKUZ","created_at":"2026-07-05T11:14:18Z"},{"alias_kind":"pith_short_8","alias_value":"HREEYD5X","created_at":"2026-07-05T11:14:18Z"}],"graph_snapshots":[{"event_id":"sha256:283ff7d924c10b05cc064275d12d64d6e1c2b6dafb336827e70386bdb7476bac","target":"graph","created_at":"2026-07-05T11:14:18Z","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/2506.01716/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models are quickly becoming the foundation for intelligent agents that are capable of using tools. However, training such agents is challenging because it requires human creation and annotation of a diverse set of tasks, tools, and evaluation criteria. In this paper, we propose the Self-Challenging framework for training an agent on high-quality tasks that are generated by itself. The agent first plays the role of challenger and generates a task after interacting with the given tools. The tasks take the form of a novel general class of problems termed Code-as-Task, which are def","authors_text":"Jason Weston, Sainbayar Sukhbaatar, Sergey Levine, Xian Li, Yifei Zhou","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T14:23:33Z","title":"Self-Challenging Language Model Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01716","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:b95cbbbf8f3cdd1185a71b0219bd76e53ae2aaac8cd2930ba42703fe03b10f1d","target":"record","created_at":"2026-07-05T11:14:18Z","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":"9c981448de726c7c4bbf0e506b85b26f62f1b60ff2ae37dcf12b950463506b9b","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-06-02T14:23:33Z","title_canon_sha256":"17c29d89e9e65373e1cdb827990aa112bebd39e96390dc2bf2d25b597cf740c5"},"schema_version":"1.0","source":{"id":"2506.01716","kind":"arxiv","version":1}},"canonical_sha256":"3c484c0fb7667225aa99bfdfaa6f9dd884debc5ff776f0937499018b27e63182","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3c484c0fb7667225aa99bfdfaa6f9dd884debc5ff776f0937499018b27e63182","first_computed_at":"2026-07-05T11:14:18.281933Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:18.281933Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"T41UBOcctBAv0BXYnQ86Iquwb1GDffEedv8+4rZX5s9oTT4taToXcdQpvFPmJizTPe6TAEATLaKWWrPLDDtrAw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:18.282421Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.01716","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b95cbbbf8f3cdd1185a71b0219bd76e53ae2aaac8cd2930ba42703fe03b10f1d","sha256:283ff7d924c10b05cc064275d12d64d6e1c2b6dafb336827e70386bdb7476bac"],"state_sha256":"7d6521919e8541a3f02a992a8befce636ff379940043df088a883642de6e5d0d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3SA1EVs0vslSvnSnF2HGop8fMfPmoIOANPK9mrKN8jo/gTPcmrwhtwQUczF9axqXd6rGTHuh4lsU447LcfLHDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-02T20:54:40.772366Z","bundle_sha256":"c02fcb74b78d57da54b660132c24605fe608342741a6eed90feb180ec4473252"}}