{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:P43VXGCJD4UVEGBSZERP6VHXJ3","short_pith_number":"pith:P43VXGCJ","schema_version":"1.0","canonical_sha256":"7f375b98491f29521832c922ff54f74ec5543bbea67a633d5f02c0fe1d188e6f","source":{"kind":"arxiv","id":"2605.28699","version":1},"attestation_state":"computed","paper":{"title":"TRACER: Turn-level Regret Matching with Inner Reinforcement Credit for Cooperative Multi-LLM Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chusen Li, Shuigeng Zhou, Wentao Zhang, Zhou Liu","submitted_at":"2026-05-27T16:25:21Z","abstract_excerpt":"Large language models increasingly rely on either reinforcement learning or multi-agent prompting to improve reasoning, yet these two paradigms remain difficult to combine. Directly applying single-agent reinforcement learning to multi-turn multi-agent systems faces following dilemmas: i) Sparse rewards, role-level free-riding and excessive training overhead. ii) Agents only imitate to collaborate. iii) Fixed collaboration protocol falls into oscillating local optimum. We introduce TRACER, a turn-level reinforcement framework for cooperative multi-LLM reasoning. TRACER separates collaborative "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.28699","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-27T16:25:21Z","cross_cats_sorted":[],"title_canon_sha256":"3d100b924859e06a7fbcc37776e79d936e053f20cc7815db282f86fe131b4cf6","abstract_canon_sha256":"1dee5d604f05cdb1a5049c48b170de194dc782902d7f78a4091c05b23e268f5b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T02:05:00.317712Z","signature_b64":"kUd5Z/3NfJr05ovjlnuqOaQMK8wBYmv2CFBml1pRlwOy++dCKkHM0pc5GyzWjq6suG4HfL+w41pRG2bu+AfBBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7f375b98491f29521832c922ff54f74ec5543bbea67a633d5f02c0fe1d188e6f","last_reissued_at":"2026-05-28T02:05:00.317289Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T02:05:00.317289Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TRACER: Turn-level Regret Matching with Inner Reinforcement Credit for Cooperative Multi-LLM Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chusen Li, Shuigeng Zhou, Wentao Zhang, Zhou Liu","submitted_at":"2026-05-27T16:25:21Z","abstract_excerpt":"Large language models increasingly rely on either reinforcement learning or multi-agent prompting to improve reasoning, yet these two paradigms remain difficult to combine. Directly applying single-agent reinforcement learning to multi-turn multi-agent systems faces following dilemmas: i) Sparse rewards, role-level free-riding and excessive training overhead. ii) Agents only imitate to collaborate. iii) Fixed collaboration protocol falls into oscillating local optimum. We introduce TRACER, a turn-level reinforcement framework for cooperative multi-LLM reasoning. TRACER separates collaborative "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28699","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/2605.28699/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.28699","created_at":"2026-05-28T02:05:00.317356+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.28699v1","created_at":"2026-05-28T02:05:00.317356+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28699","created_at":"2026-05-28T02:05:00.317356+00:00"},{"alias_kind":"pith_short_12","alias_value":"P43VXGCJD4UV","created_at":"2026-05-28T02:05:00.317356+00:00"},{"alias_kind":"pith_short_16","alias_value":"P43VXGCJD4UVEGBS","created_at":"2026-05-28T02:05:00.317356+00:00"},{"alias_kind":"pith_short_8","alias_value":"P43VXGCJ","created_at":"2026-05-28T02:05:00.317356+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P43VXGCJD4UVEGBSZERP6VHXJ3","json":"https://pith.science/pith/P43VXGCJD4UVEGBSZERP6VHXJ3.json","graph_json":"https://pith.science/api/pith-number/P43VXGCJD4UVEGBSZERP6VHXJ3/graph.json","events_json":"https://pith.science/api/pith-number/P43VXGCJD4UVEGBSZERP6VHXJ3/events.json","paper":"https://pith.science/paper/P43VXGCJ"},"agent_actions":{"view_html":"https://pith.science/pith/P43VXGCJD4UVEGBSZERP6VHXJ3","download_json":"https://pith.science/pith/P43VXGCJD4UVEGBSZERP6VHXJ3.json","view_paper":"https://pith.science/paper/P43VXGCJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.28699&json=true","fetch_graph":"https://pith.science/api/pith-number/P43VXGCJD4UVEGBSZERP6VHXJ3/graph.json","fetch_events":"https://pith.science/api/pith-number/P43VXGCJD4UVEGBSZERP6VHXJ3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P43VXGCJD4UVEGBSZERP6VHXJ3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P43VXGCJD4UVEGBSZERP6VHXJ3/action/storage_attestation","attest_author":"https://pith.science/pith/P43VXGCJD4UVEGBSZERP6VHXJ3/action/author_attestation","sign_citation":"https://pith.science/pith/P43VXGCJD4UVEGBSZERP6VHXJ3/action/citation_signature","submit_replication":"https://pith.science/pith/P43VXGCJD4UVEGBSZERP6VHXJ3/action/replication_record"}},"created_at":"2026-05-28T02:05:00.317356+00:00","updated_at":"2026-05-28T02:05:00.317356+00:00"}