{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:TQPC6CUXS7XXSQGSS6HPUNKU2U","short_pith_number":"pith:TQPC6CUX","canonical_record":{"source":{"id":"2607.07646","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-08T17:04:42Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"fbf40a49852233c9d84a14cfab55e5243e713f6e059981899352eccc9fcd1692","abstract_canon_sha256":"2c81278d8a27c7f627afed6dfd04543992dc6f390a7d1834f521ee61d07986f6"},"schema_version":"1.0"},"canonical_sha256":"9c1e2f0a9797ef7940d2978efa3554d51e159d20b0e6d4e47398ffbc3d50bad2","source":{"kind":"arxiv","id":"2607.07646","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.07646","created_at":"2026-07-09T01:20:37Z"},{"alias_kind":"arxiv_version","alias_value":"2607.07646v1","created_at":"2026-07-09T01:20:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.07646","created_at":"2026-07-09T01:20:37Z"},{"alias_kind":"pith_short_12","alias_value":"TQPC6CUXS7XX","created_at":"2026-07-09T01:20:37Z"},{"alias_kind":"pith_short_16","alias_value":"TQPC6CUXS7XXSQGS","created_at":"2026-07-09T01:20:37Z"},{"alias_kind":"pith_short_8","alias_value":"TQPC6CUX","created_at":"2026-07-09T01:20:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:TQPC6CUXS7XXSQGSS6HPUNKU2U","target":"record","payload":{"canonical_record":{"source":{"id":"2607.07646","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-08T17:04:42Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"fbf40a49852233c9d84a14cfab55e5243e713f6e059981899352eccc9fcd1692","abstract_canon_sha256":"2c81278d8a27c7f627afed6dfd04543992dc6f390a7d1834f521ee61d07986f6"},"schema_version":"1.0"},"canonical_sha256":"9c1e2f0a9797ef7940d2978efa3554d51e159d20b0e6d4e47398ffbc3d50bad2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-09T01:20:37.900037Z","signature_b64":"RfDchhjg0ZSjoBcXq9l//Ojbq/tYgESMkOXFfNM391suQ93LeDbnWJBvcwGi9DwMspc86HiuPW/tMZ6Qk4n0BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c1e2f0a9797ef7940d2978efa3554d51e159d20b0e6d4e47398ffbc3d50bad2","last_reissued_at":"2026-07-09T01:20:37.899510Z","signature_status":"signed_v1","first_computed_at":"2026-07-09T01:20:37.899510Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.07646","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":"2026-07-09T01:20:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"o0N9n9AT2V6oMuXA6f1LU9WS+X8LmT/c4Uk7sE+87SPjn9M/2/n4uKY0ldkLC+jZPbrV7iMhpk/PFXY0K32ECw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T15:46:57.394924Z"},"content_sha256":"5b1430063b1cd1fe86814bede8d7240bb2b14442568bd72d4365ce1ee32338d5","schema_version":"1.0","event_id":"sha256:5b1430063b1cd1fe86814bede8d7240bb2b14442568bd72d4365ce1ee32338d5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:TQPC6CUXS7XXSQGSS6HPUNKU2U","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RL Post-Training Builds Compositional Reasoning Strategies","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Andrew Saxe, Azwar Abdulsalam, Nishil Patel","submitted_at":"2026-07-08T17:04:42Z","abstract_excerpt":"Does RL post-training merely amplify primitive skills already latent in a base model, or can it compose primitive skills into new higher-level strategies? We study this question in a fully observable rewrite-grammar environment where the pretraining distribution is known and every generated rewrite can be audited. A Transformer is pretrained on primitive symbol-rewrite chains and post-trained on a Trace-based reasoning task with only a binary final-answer reward. RL solves held-out problems that remain rarely solved by the pretrained model even under much larger sampling budgets, while rejecti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.07646","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/2607.07646/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":"2026-07-09T01:20:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OaORlkvNBAEzBfySC5Z2fxhyNOrgLvbiuGLkrUH9A/iusjs0afJeWT35mvKHPVHhTwnUPCO0oWMSXm+pUo8bDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T15:46:57.395424Z"},"content_sha256":"0edd295c9ccbfcf3dfeef7d3bc2aa05dd50ef748c613fa78d16c62df47c7ed87","schema_version":"1.0","event_id":"sha256:0edd295c9ccbfcf3dfeef7d3bc2aa05dd50ef748c613fa78d16c62df47c7ed87"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TQPC6CUXS7XXSQGSS6HPUNKU2U/bundle.json","state_url":"https://pith.science/pith/TQPC6CUXS7XXSQGSS6HPUNKU2U/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TQPC6CUXS7XXSQGSS6HPUNKU2U/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":"2026-08-03T15:46:57Z","links":{"resolver":"https://pith.science/pith/TQPC6CUXS7XXSQGSS6HPUNKU2U","bundle":"https://pith.science/pith/TQPC6CUXS7XXSQGSS6HPUNKU2U/bundle.json","state":"https://pith.science/pith/TQPC6CUXS7XXSQGSS6HPUNKU2U/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TQPC6CUXS7XXSQGSS6HPUNKU2U/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TQPC6CUXS7XXSQGSS6HPUNKU2U","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":"2c81278d8a27c7f627afed6dfd04543992dc6f390a7d1834f521ee61d07986f6","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-08T17:04:42Z","title_canon_sha256":"fbf40a49852233c9d84a14cfab55e5243e713f6e059981899352eccc9fcd1692"},"schema_version":"1.0","source":{"id":"2607.07646","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.07646","created_at":"2026-07-09T01:20:37Z"},{"alias_kind":"arxiv_version","alias_value":"2607.07646v1","created_at":"2026-07-09T01:20:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.07646","created_at":"2026-07-09T01:20:37Z"},{"alias_kind":"pith_short_12","alias_value":"TQPC6CUXS7XX","created_at":"2026-07-09T01:20:37Z"},{"alias_kind":"pith_short_16","alias_value":"TQPC6CUXS7XXSQGS","created_at":"2026-07-09T01:20:37Z"},{"alias_kind":"pith_short_8","alias_value":"TQPC6CUX","created_at":"2026-07-09T01:20:37Z"}],"graph_snapshots":[{"event_id":"sha256:0edd295c9ccbfcf3dfeef7d3bc2aa05dd50ef748c613fa78d16c62df47c7ed87","target":"graph","created_at":"2026-07-09T01:20:37Z","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/2607.07646/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Does RL post-training merely amplify primitive skills already latent in a base model, or can it compose primitive skills into new higher-level strategies? We study this question in a fully observable rewrite-grammar environment where the pretraining distribution is known and every generated rewrite can be audited. A Transformer is pretrained on primitive symbol-rewrite chains and post-trained on a Trace-based reasoning task with only a binary final-answer reward. RL solves held-out problems that remain rarely solved by the pretrained model even under much larger sampling budgets, while rejecti","authors_text":"Andrew Saxe, Azwar Abdulsalam, Nishil Patel","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-08T17:04:42Z","title":"RL Post-Training Builds Compositional Reasoning Strategies"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.07646","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":"sha256:5b1430063b1cd1fe86814bede8d7240bb2b14442568bd72d4365ce1ee32338d5","target":"record","created_at":"2026-07-09T01:20:37Z","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":"2c81278d8a27c7f627afed6dfd04543992dc6f390a7d1834f521ee61d07986f6","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-08T17:04:42Z","title_canon_sha256":"fbf40a49852233c9d84a14cfab55e5243e713f6e059981899352eccc9fcd1692"},"schema_version":"1.0","source":{"id":"2607.07646","kind":"arxiv","version":1}},"canonical_sha256":"9c1e2f0a9797ef7940d2978efa3554d51e159d20b0e6d4e47398ffbc3d50bad2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9c1e2f0a9797ef7940d2978efa3554d51e159d20b0e6d4e47398ffbc3d50bad2","first_computed_at":"2026-07-09T01:20:37.899510Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-09T01:20:37.899510Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RfDchhjg0ZSjoBcXq9l//Ojbq/tYgESMkOXFfNM391suQ93LeDbnWJBvcwGi9DwMspc86HiuPW/tMZ6Qk4n0BA==","signature_status":"signed_v1","signed_at":"2026-07-09T01:20:37.900037Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.07646","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5b1430063b1cd1fe86814bede8d7240bb2b14442568bd72d4365ce1ee32338d5","sha256:0edd295c9ccbfcf3dfeef7d3bc2aa05dd50ef748c613fa78d16c62df47c7ed87"],"state_sha256":"c9b890a8cea4231525f9231d8c57840ce48256bb867e0f04633e0aabf1b1f856"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"G462pgdew8sa4fO00eAtLvpfHzqI6uoY6rcJ1jcHYdNib+HWWeKcWPXc9JKKAhZ9QbFQcApbFaKLQqBGnZBZDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T15:46:57.398845Z","bundle_sha256":"d40d9ec685e8e5ccebd5799f19d3210abdcaa98c0f6a2a747c7fb4b8e36bf6da"}}