{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:5TOP3QL74RXK76JBTNLPE77PDN","short_pith_number":"pith:5TOP3QL7","canonical_record":{"source":{"id":"2503.09501","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-12T16:05:31Z","cross_cats_sorted":["cs.CL","cs.LG","cs.MA"],"title_canon_sha256":"b7c0eba3a40500deb0e28334be45be2db8924fe1bbf551e87d8e9ca1d67e294c","abstract_canon_sha256":"c8e09b324596917b2d8f95786197dd905ff11d31e71c6d83bfef9ea73321f458"},"schema_version":"1.0"},"canonical_sha256":"ecdcfdc17fe46eaff9219b56f27fef1b4e1fec7faffec222d9be669b498f40fd","source":{"kind":"arxiv","id":"2503.09501","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.09501","created_at":"2026-07-05T11:10:06Z"},{"alias_kind":"arxiv_version","alias_value":"2503.09501v3","created_at":"2026-07-05T11:10:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.09501","created_at":"2026-07-05T11:10:06Z"},{"alias_kind":"pith_short_12","alias_value":"5TOP3QL74RXK","created_at":"2026-07-05T11:10:06Z"},{"alias_kind":"pith_short_16","alias_value":"5TOP3QL74RXK76JB","created_at":"2026-07-05T11:10:06Z"},{"alias_kind":"pith_short_8","alias_value":"5TOP3QL7","created_at":"2026-07-05T11:10:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:5TOP3QL74RXK76JBTNLPE77PDN","target":"record","payload":{"canonical_record":{"source":{"id":"2503.09501","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-12T16:05:31Z","cross_cats_sorted":["cs.CL","cs.LG","cs.MA"],"title_canon_sha256":"b7c0eba3a40500deb0e28334be45be2db8924fe1bbf551e87d8e9ca1d67e294c","abstract_canon_sha256":"c8e09b324596917b2d8f95786197dd905ff11d31e71c6d83bfef9ea73321f458"},"schema_version":"1.0"},"canonical_sha256":"ecdcfdc17fe46eaff9219b56f27fef1b4e1fec7faffec222d9be669b498f40fd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:06.361782Z","signature_b64":"5aelTVMOwE9RnAzNjKGyQEiFtk5w/vpz8apkTfJ7yUsg40oiM01QkSkOUkqTmvNVxfq6UT9SvDPC6L2ukkkgAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ecdcfdc17fe46eaff9219b56f27fef1b4e1fec7faffec222d9be669b498f40fd","last_reissued_at":"2026-07-05T11:10:06.361246Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:06.361246Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.09501","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:10:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pKQxOiDyhHC3NAWRD8M/ksWT9sr6k0DjTzeB/WFoAgcmXHteUES2yby0Y9p9jDOEqDAXC4RSH3gAx6BiPb93CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T21:47:54.372953Z"},"content_sha256":"b10bf72033add58ecd3b782a4ca3072355dfc1851d311e4b23b14e4e7bdd1458","schema_version":"1.0","event_id":"sha256:b10bf72033add58ecd3b782a4ca3072355dfc1851d311e4b23b14e4e7bdd1458"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:5TOP3QL74RXK76JBTNLPE77PDN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ReMA: Learning to Meta-think for LLMs with Multi-Agent Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG","cs.MA"],"primary_cat":"cs.AI","authors_text":"Hanjing Wang, Jun Wang, Linyi Yang, Mark Schmidt, Shuyue Hu, Weinan Zhang, Xiaoyu Wen, Yan Song, Ying Wen, Yunxiang Li, Ziyu Wan","submitted_at":"2025-03-12T16:05:31Z","abstract_excerpt":"Recent research on Reasoning of Large Language Models (LLMs) has sought to further enhance their performance by integrating meta-thinking -- enabling models to monitor, evaluate, and control their reasoning processes for more adaptive and effective problem-solving. However, current single-agent work lacks a specialized design for acquiring meta-thinking, resulting in low efficacy. To address this challenge, we introduce Reinforced Meta-thinking Agents (ReMA), a novel framework that leverages Multi-Agent Reinforcement Learning (MARL) to elicit meta-thinking behaviors, encouraging LLMs to think "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.09501","kind":"arxiv","version":3},"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/2503.09501/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":"2026-07-05T11:10:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wIEWzvddlQLJuH5JQ0qePAnYXDU4piQjGv8fbL0Z4TGyD3P0nummfVoFv4yoL56mQGeBLXgQ80DDBza5oIH2Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T21:47:54.373372Z"},"content_sha256":"48c0c05e9b58bd598f711c97c35c30d42625462e45feb10f768e4c1209e4879d","schema_version":"1.0","event_id":"sha256:48c0c05e9b58bd598f711c97c35c30d42625462e45feb10f768e4c1209e4879d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5TOP3QL74RXK76JBTNLPE77PDN/bundle.json","state_url":"https://pith.science/pith/5TOP3QL74RXK76JBTNLPE77PDN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5TOP3QL74RXK76JBTNLPE77PDN/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":"2026-07-06T21:47:54Z","links":{"resolver":"https://pith.science/pith/5TOP3QL74RXK76JBTNLPE77PDN","bundle":"https://pith.science/pith/5TOP3QL74RXK76JBTNLPE77PDN/bundle.json","state":"https://pith.science/pith/5TOP3QL74RXK76JBTNLPE77PDN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5TOP3QL74RXK76JBTNLPE77PDN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:5TOP3QL74RXK76JBTNLPE77PDN","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":"c8e09b324596917b2d8f95786197dd905ff11d31e71c6d83bfef9ea73321f458","cross_cats_sorted":["cs.CL","cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-12T16:05:31Z","title_canon_sha256":"b7c0eba3a40500deb0e28334be45be2db8924fe1bbf551e87d8e9ca1d67e294c"},"schema_version":"1.0","source":{"id":"2503.09501","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.09501","created_at":"2026-07-05T11:10:06Z"},{"alias_kind":"arxiv_version","alias_value":"2503.09501v3","created_at":"2026-07-05T11:10:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.09501","created_at":"2026-07-05T11:10:06Z"},{"alias_kind":"pith_short_12","alias_value":"5TOP3QL74RXK","created_at":"2026-07-05T11:10:06Z"},{"alias_kind":"pith_short_16","alias_value":"5TOP3QL74RXK76JB","created_at":"2026-07-05T11:10:06Z"},{"alias_kind":"pith_short_8","alias_value":"5TOP3QL7","created_at":"2026-07-05T11:10:06Z"}],"graph_snapshots":[{"event_id":"sha256:48c0c05e9b58bd598f711c97c35c30d42625462e45feb10f768e4c1209e4879d","target":"graph","created_at":"2026-07-05T11:10:06Z","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/2503.09501/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent research on Reasoning of Large Language Models (LLMs) has sought to further enhance their performance by integrating meta-thinking -- enabling models to monitor, evaluate, and control their reasoning processes for more adaptive and effective problem-solving. However, current single-agent work lacks a specialized design for acquiring meta-thinking, resulting in low efficacy. To address this challenge, we introduce Reinforced Meta-thinking Agents (ReMA), a novel framework that leverages Multi-Agent Reinforcement Learning (MARL) to elicit meta-thinking behaviors, encouraging LLMs to think ","authors_text":"Hanjing Wang, Jun Wang, Linyi Yang, Mark Schmidt, Shuyue Hu, Weinan Zhang, Xiaoyu Wen, Yan Song, Ying Wen, Yunxiang Li, Ziyu Wan","cross_cats":["cs.CL","cs.LG","cs.MA"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-12T16:05:31Z","title":"ReMA: Learning to Meta-think for LLMs with Multi-Agent Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.09501","kind":"arxiv","version":3},"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:b10bf72033add58ecd3b782a4ca3072355dfc1851d311e4b23b14e4e7bdd1458","target":"record","created_at":"2026-07-05T11:10:06Z","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":"c8e09b324596917b2d8f95786197dd905ff11d31e71c6d83bfef9ea73321f458","cross_cats_sorted":["cs.CL","cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-12T16:05:31Z","title_canon_sha256":"b7c0eba3a40500deb0e28334be45be2db8924fe1bbf551e87d8e9ca1d67e294c"},"schema_version":"1.0","source":{"id":"2503.09501","kind":"arxiv","version":3}},"canonical_sha256":"ecdcfdc17fe46eaff9219b56f27fef1b4e1fec7faffec222d9be669b498f40fd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ecdcfdc17fe46eaff9219b56f27fef1b4e1fec7faffec222d9be669b498f40fd","first_computed_at":"2026-07-05T11:10:06.361246Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:10:06.361246Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"5aelTVMOwE9RnAzNjKGyQEiFtk5w/vpz8apkTfJ7yUsg40oiM01QkSkOUkqTmvNVxfq6UT9SvDPC6L2ukkkgAg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:10:06.361782Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.09501","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b10bf72033add58ecd3b782a4ca3072355dfc1851d311e4b23b14e4e7bdd1458","sha256:48c0c05e9b58bd598f711c97c35c30d42625462e45feb10f768e4c1209e4879d"],"state_sha256":"525c9aa34fcdd7a74854efa807a65966ecc8271495ff56b2b6b2d6b61947da85"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UlA8stRR9TModQ3yDtD8EObHEbeO2ruCO2kEko38a+0WCvIbg9gChM3ssau1TS2jTLf6O8HBG4nZRngTDwAvAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T21:47:54.375288Z","bundle_sha256":"ccc5bf1fa44c111bf8a361ee8918aeeff6580d7a1d64caf399451a136c0d0b17"}}