{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ZCMT7GW5BDOK5Q63FTI4NMGATG","short_pith_number":"pith:ZCMT7GW5","canonical_record":{"source":{"id":"2505.07773","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-12T17:23:34Z","cross_cats_sorted":[],"title_canon_sha256":"2a6c3dcfd1ce1ac75f768aa109db9618d58ca8349001ce39b6355e92950efd51","abstract_canon_sha256":"fe15f98a87556bcc425d997dca362c7f7b837734165da3a753d414be7e51b8eb"},"schema_version":"1.0"},"canonical_sha256":"c8993f9add08dcaec3db2cd1c6b0c099b14ba68d11e92d0d8b997cd404f1695a","source":{"kind":"arxiv","id":"2505.07773","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.07773","created_at":"2026-07-05T11:56:19Z"},{"alias_kind":"arxiv_version","alias_value":"2505.07773v4","created_at":"2026-07-05T11:56:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.07773","created_at":"2026-07-05T11:56:19Z"},{"alias_kind":"pith_short_12","alias_value":"ZCMT7GW5BDOK","created_at":"2026-07-05T11:56:19Z"},{"alias_kind":"pith_short_16","alias_value":"ZCMT7GW5BDOK5Q63","created_at":"2026-07-05T11:56:19Z"},{"alias_kind":"pith_short_8","alias_value":"ZCMT7GW5","created_at":"2026-07-05T11:56:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ZCMT7GW5BDOK5Q63FTI4NMGATG","target":"record","payload":{"canonical_record":{"source":{"id":"2505.07773","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-12T17:23:34Z","cross_cats_sorted":[],"title_canon_sha256":"2a6c3dcfd1ce1ac75f768aa109db9618d58ca8349001ce39b6355e92950efd51","abstract_canon_sha256":"fe15f98a87556bcc425d997dca362c7f7b837734165da3a753d414be7e51b8eb"},"schema_version":"1.0"},"canonical_sha256":"c8993f9add08dcaec3db2cd1c6b0c099b14ba68d11e92d0d8b997cd404f1695a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:56:19.556205Z","signature_b64":"Qs3JKQW2QzXev0Mmk2x9ipr7tEAHYXpZSOG+xJ2xeATIMzNWVroDWc3TGUK7gpj8D9LWP/X+s4rlYC1pl/tpAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c8993f9add08dcaec3db2cd1c6b0c099b14ba68d11e92d0d8b997cd404f1695a","last_reissued_at":"2026-07-05T11:56:19.555718Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:56:19.555718Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.07773","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":"2026-07-05T11:56:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"a4OJL+76qwZa29UHICwPmGMPIRpfyI/vCLfGNVKBmEOMRNrlLyMp+wjYumt9Wod9R9CSGQ48CbWwukejI+z6Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T12:21:42.068272Z"},"content_sha256":"58d39ebba14e503f360ef631d8a41ed7c7df613926474854cd69b74020226794","schema_version":"1.0","event_id":"sha256:58d39ebba14e503f360ef631d8a41ed7c7df613926474854cd69b74020226794"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ZCMT7GW5BDOK5Q63FTI4NMGATG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Agent RL Scaling Law: Agent RL with Spontaneous Code Execution for Mathematical Problem Solving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Haotian Xu, Jian Hu, Weinong Wang, Wenqiang Zhang, Xing W, Xinji Mai, Yingying Zhang, Zhong-Zhi Li","submitted_at":"2025-05-12T17:23:34Z","abstract_excerpt":"Large Language Models (LLMs) often struggle with mathematical reasoning tasks requiring precise, verifiable computation. While Reinforcement Learning (RL) from outcome-based rewards enhances text-based reasoning, understanding how agents autonomously learn to leverage external tools like code execution remains crucial. We investigate RL from outcome-based rewards for Tool-Integrated Reasoning, ZeroTIR, training base LLMs to spontaneously generate and execute Python code for mathematical problems without supervised tool-use examples. Our central contribution is we demonstrate that as RL trainin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.07773","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/2505.07773/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":"2026-07-05T11:56:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GurRKTMI3AfRnbA4j83bNjvB0qxlI10mutlRRqEr71SJ+G96dg+41u3z862XP4RkP4sy4s6hKw0JpKNnz+VnDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T12:21:42.069134Z"},"content_sha256":"1bb485ee98aca8bb00fee85a7021f9738d587ef6e0e605da6e878f0e4496ac40","schema_version":"1.0","event_id":"sha256:1bb485ee98aca8bb00fee85a7021f9738d587ef6e0e605da6e878f0e4496ac40"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZCMT7GW5BDOK5Q63FTI4NMGATG/bundle.json","state_url":"https://pith.science/pith/ZCMT7GW5BDOK5Q63FTI4NMGATG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZCMT7GW5BDOK5Q63FTI4NMGATG/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":"2026-08-03T12:21:42Z","links":{"resolver":"https://pith.science/pith/ZCMT7GW5BDOK5Q63FTI4NMGATG","bundle":"https://pith.science/pith/ZCMT7GW5BDOK5Q63FTI4NMGATG/bundle.json","state":"https://pith.science/pith/ZCMT7GW5BDOK5Q63FTI4NMGATG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZCMT7GW5BDOK5Q63FTI4NMGATG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZCMT7GW5BDOK5Q63FTI4NMGATG","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":"fe15f98a87556bcc425d997dca362c7f7b837734165da3a753d414be7e51b8eb","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-12T17:23:34Z","title_canon_sha256":"2a6c3dcfd1ce1ac75f768aa109db9618d58ca8349001ce39b6355e92950efd51"},"schema_version":"1.0","source":{"id":"2505.07773","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.07773","created_at":"2026-07-05T11:56:19Z"},{"alias_kind":"arxiv_version","alias_value":"2505.07773v4","created_at":"2026-07-05T11:56:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.07773","created_at":"2026-07-05T11:56:19Z"},{"alias_kind":"pith_short_12","alias_value":"ZCMT7GW5BDOK","created_at":"2026-07-05T11:56:19Z"},{"alias_kind":"pith_short_16","alias_value":"ZCMT7GW5BDOK5Q63","created_at":"2026-07-05T11:56:19Z"},{"alias_kind":"pith_short_8","alias_value":"ZCMT7GW5","created_at":"2026-07-05T11:56:19Z"}],"graph_snapshots":[{"event_id":"sha256:1bb485ee98aca8bb00fee85a7021f9738d587ef6e0e605da6e878f0e4496ac40","target":"graph","created_at":"2026-07-05T11:56:19Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2505.07773/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) often struggle with mathematical reasoning tasks requiring precise, verifiable computation. While Reinforcement Learning (RL) from outcome-based rewards enhances text-based reasoning, understanding how agents autonomously learn to leverage external tools like code execution remains crucial. We investigate RL from outcome-based rewards for Tool-Integrated Reasoning, ZeroTIR, training base LLMs to spontaneously generate and execute Python code for mathematical problems without supervised tool-use examples. Our central contribution is we demonstrate that as RL trainin","authors_text":"Haotian Xu, Jian Hu, Weinong Wang, Wenqiang Zhang, Xing W, Xinji Mai, Yingying Zhang, Zhong-Zhi Li","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-12T17:23:34Z","title":"Agent RL Scaling Law: Agent RL with Spontaneous Code Execution for Mathematical Problem Solving"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.07773","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":"sha256:58d39ebba14e503f360ef631d8a41ed7c7df613926474854cd69b74020226794","target":"record","created_at":"2026-07-05T11:56:19Z","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":"fe15f98a87556bcc425d997dca362c7f7b837734165da3a753d414be7e51b8eb","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-12T17:23:34Z","title_canon_sha256":"2a6c3dcfd1ce1ac75f768aa109db9618d58ca8349001ce39b6355e92950efd51"},"schema_version":"1.0","source":{"id":"2505.07773","kind":"arxiv","version":4}},"canonical_sha256":"c8993f9add08dcaec3db2cd1c6b0c099b14ba68d11e92d0d8b997cd404f1695a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c8993f9add08dcaec3db2cd1c6b0c099b14ba68d11e92d0d8b997cd404f1695a","first_computed_at":"2026-07-05T11:56:19.555718Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:56:19.555718Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Qs3JKQW2QzXev0Mmk2x9ipr7tEAHYXpZSOG+xJ2xeATIMzNWVroDWc3TGUK7gpj8D9LWP/X+s4rlYC1pl/tpAA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:56:19.556205Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.07773","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:58d39ebba14e503f360ef631d8a41ed7c7df613926474854cd69b74020226794","sha256:1bb485ee98aca8bb00fee85a7021f9738d587ef6e0e605da6e878f0e4496ac40"],"state_sha256":"aced95a93eaeda55f74971efca71638fbf65f0fd33bc33aad903453f6bf26ce8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OIv6Xl3F3XY6G6ckOrGj3UZI0DRk9kYw8S3KlPCV1xQ917NO7456NRz+XN9Xi/TGfLov3oWCBr4a5QXOdpSNCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T12:21:42.080128Z","bundle_sha256":"5936a3340ca7f78f7390880f0c95276ac6929c9542c22a52bcf221c35c0d2c1d"}}