{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:DJYFIDYJCQYEOXCIM4CKJL6NFC","short_pith_number":"pith:DJYFIDYJ","canonical_record":{"source":{"id":"2610.23344","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-10-16T11:51:00Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b3dfdea22bc4f567dd9ca0d621fbcd9ec26abc467a9ac0e177f24886bd613b0f","abstract_canon_sha256":"5fde4e3c5ed8be39b7773fcf7aaad30fe46db42a192b7cce598ae93b896ba681"},"schema_version":"1.0"},"canonical_sha256":"1a70540f091430475c486704a4afcd28a2d788553fdf7d63915b9cff6ff19ef4","source":{"kind":"arxiv","id":"2610.23344","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2610.23344","created_at":"2026-10-19T11:16:22Z"},{"alias_kind":"arxiv_version","alias_value":"2610.23344v1","created_at":"2026-10-19T11:16:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2610.23344","created_at":"2026-10-19T11:16:22Z"},{"alias_kind":"pith_short_12","alias_value":"DJYFIDYJCQYE","created_at":"2026-10-19T11:16:22Z"},{"alias_kind":"pith_short_16","alias_value":"DJYFIDYJCQYEOXCI","created_at":"2026-10-19T11:16:22Z"},{"alias_kind":"pith_short_8","alias_value":"DJYFIDYJ","created_at":"2026-10-19T11:16:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:DJYFIDYJCQYEOXCIM4CKJL6NFC","target":"record","payload":{"canonical_record":{"source":{"id":"2610.23344","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-10-16T11:51:00Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b3dfdea22bc4f567dd9ca0d621fbcd9ec26abc467a9ac0e177f24886bd613b0f","abstract_canon_sha256":"5fde4e3c5ed8be39b7773fcf7aaad30fe46db42a192b7cce598ae93b896ba681"},"schema_version":"1.0"},"canonical_sha256":"1a70540f091430475c486704a4afcd28a2d788553fdf7d63915b9cff6ff19ef4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-10-19T11:16:22.961089Z","signature_b64":"7cDXcf0EfkHuDzp2yaq7y+IHE5ZomMS3LJLUIOYej0NINioc+rvNW3N7L9SCLt2IzXYPVzqZuFleB6eD4qp3Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1a70540f091430475c486704a4afcd28a2d788553fdf7d63915b9cff6ff19ef4","last_reissued_at":"2026-10-19T11:16:22.959411Z","signature_status":"signed_v1","first_computed_at":"2026-10-19T11:16:22.959411Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2610.23344","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":"2026-10-19T11:16:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"f1LBkXR4QcOBRrWnbu2AUultTqisRGth7wg9y+vl/pL1ecNJhbSi3DbqgdQ9Fqu8xcckosaCJBoal1jfJMnoDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-11-22T02:14:27.260529Z"},"content_sha256":"2ee38d71458b43daeacd0cc3dcaa1015b44e857a10ae13a05a650cbc7f2b32d8","schema_version":"1.0","event_id":"sha256:2ee38d71458b43daeacd0cc3dcaa1015b44e857a10ae13a05a650cbc7f2b32d8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:DJYFIDYJCQYEOXCIM4CKJL6NFC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Language Models Learn to Improve Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Benoit Dumoulin, Bing He, Hanqing Lu, Minhua Lin, Suhang Wang, Xiang Zhang, Zewen Liu, Zhan Shi, Zhaotian Weng, Zijun Wang","submitted_at":"2026-10-16T11:51:00Z","abstract_excerpt":"Large language models (LLMs) are increasingly used to automate AI research, including language-model (LM) development, offering a potential path toward recursive self-improvement. However, turning experimental feedback into a learned ability to propose effective improvements across models faces two challenges. First, training such an ability with real feedback is expensive: reinforcement learning (RL) requires evaluating tens of thousands of proposals, each potentially requiring a full model-training run. Even at ten GPU-hours per run, reward collection alone would consume hundreds of thousand"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2610.23344","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/2610.23344/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"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":"2026-10-19T11:16:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F6wUPiRNG1Lhk/haWPN6s1Lj9b8wR1P5pCi7/56beOfUWU0bioSc4+/tlWh3i7unvgmsHEl6RgMEBvcfC4HdAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-11-22T02:14:27.260864Z"},"content_sha256":"109f5168ec813832af4921812cd02431a9c4cc039ff31d34d45a0c5591ad0d6a","schema_version":"1.0","event_id":"sha256:109f5168ec813832af4921812cd02431a9c4cc039ff31d34d45a0c5591ad0d6a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DJYFIDYJCQYEOXCIM4CKJL6NFC/bundle.json","state_url":"https://pith.science/pith/DJYFIDYJCQYEOXCIM4CKJL6NFC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DJYFIDYJCQYEOXCIM4CKJL6NFC/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":"2026-11-22T02:14:27Z","links":{"resolver":"https://pith.science/pith/DJYFIDYJCQYEOXCIM4CKJL6NFC","bundle":"https://pith.science/pith/DJYFIDYJCQYEOXCIM4CKJL6NFC/bundle.json","state":"https://pith.science/pith/DJYFIDYJCQYEOXCIM4CKJL6NFC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DJYFIDYJCQYEOXCIM4CKJL6NFC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DJYFIDYJCQYEOXCIM4CKJL6NFC","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":"5fde4e3c5ed8be39b7773fcf7aaad30fe46db42a192b7cce598ae93b896ba681","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-10-16T11:51:00Z","title_canon_sha256":"b3dfdea22bc4f567dd9ca0d621fbcd9ec26abc467a9ac0e177f24886bd613b0f"},"schema_version":"1.0","source":{"id":"2610.23344","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2610.23344","created_at":"2026-10-19T11:16:22Z"},{"alias_kind":"arxiv_version","alias_value":"2610.23344v1","created_at":"2026-10-19T11:16:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2610.23344","created_at":"2026-10-19T11:16:22Z"},{"alias_kind":"pith_short_12","alias_value":"DJYFIDYJCQYE","created_at":"2026-10-19T11:16:22Z"},{"alias_kind":"pith_short_16","alias_value":"DJYFIDYJCQYEOXCI","created_at":"2026-10-19T11:16:22Z"},{"alias_kind":"pith_short_8","alias_value":"DJYFIDYJ","created_at":"2026-10-19T11:16:22Z"}],"graph_snapshots":[{"event_id":"sha256:109f5168ec813832af4921812cd02431a9c4cc039ff31d34d45a0c5591ad0d6a","target":"graph","created_at":"2026-10-19T11:16:22Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2610.23344/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) are increasingly used to automate AI research, including language-model (LM) development, offering a potential path toward recursive self-improvement. However, turning experimental feedback into a learned ability to propose effective improvements across models faces two challenges. First, training such an ability with real feedback is expensive: reinforcement learning (RL) requires evaluating tens of thousands of proposals, each potentially requiring a full model-training run. Even at ten GPU-hours per run, reward collection alone would consume hundreds of thousand","authors_text":"Benoit Dumoulin, Bing He, Hanqing Lu, Minhua Lin, Suhang Wang, Xiang Zhang, Zewen Liu, Zhan Shi, Zhaotian Weng, Zijun Wang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-10-16T11:51:00Z","title":"Language Models Learn to Improve Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2610.23344","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":"sha256:2ee38d71458b43daeacd0cc3dcaa1015b44e857a10ae13a05a650cbc7f2b32d8","target":"record","created_at":"2026-10-19T11:16:22Z","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":"5fde4e3c5ed8be39b7773fcf7aaad30fe46db42a192b7cce598ae93b896ba681","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-10-16T11:51:00Z","title_canon_sha256":"b3dfdea22bc4f567dd9ca0d621fbcd9ec26abc467a9ac0e177f24886bd613b0f"},"schema_version":"1.0","source":{"id":"2610.23344","kind":"arxiv","version":1}},"canonical_sha256":"1a70540f091430475c486704a4afcd28a2d788553fdf7d63915b9cff6ff19ef4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1a70540f091430475c486704a4afcd28a2d788553fdf7d63915b9cff6ff19ef4","first_computed_at":"2026-10-19T11:16:22.959411Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-10-19T11:16:22.959411Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7cDXcf0EfkHuDzp2yaq7y+IHE5ZomMS3LJLUIOYej0NINioc+rvNW3N7L9SCLt2IzXYPVzqZuFleB6eD4qp3Bw==","signature_status":"signed_v1","signed_at":"2026-10-19T11:16:22.961089Z","signed_message":"canonical_sha256_bytes"},"source_id":"2610.23344","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2ee38d71458b43daeacd0cc3dcaa1015b44e857a10ae13a05a650cbc7f2b32d8","sha256:109f5168ec813832af4921812cd02431a9c4cc039ff31d34d45a0c5591ad0d6a"],"state_sha256":"98e95decf475513bfbb34f8b7373a6079113a796344ea8e872367592a9ea1320"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"usimEfJTBalMvcGG5Nb2vd6ggHWU8St5xRDvlSum6Hl4DJK05Sm2KCr5ce/W1azxYnTv41Di65L61Y/5qcMdAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-11-22T02:14:27.267209Z","bundle_sha256":"4a136f93c1285f3ae4a1d91b7bb7351b982efbffd2e1f05fa43554bc3126f75a"}}