{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:FOBSE6NPZRW3OUDYZRF5Y55NGS","short_pith_number":"pith:FOBSE6NP","canonical_record":{"source":{"id":"2601.22900","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-01-30T12:19:54Z","cross_cats_sorted":[],"title_canon_sha256":"2558c79dfdb7b6302e8ab18ffb9eb8d6b5a77ac7610eb8b6eeadb7e517b2aba4","abstract_canon_sha256":"ae1d0127dc5056920fae5bdf70e79b77e0dec444fbde45ddcfe449c5ec8a9085"},"schema_version":"1.0"},"canonical_sha256":"2b832279afcc6db75078cc4bdc77ad34b75f6ece0631d2a5a9ee1ea98caccfb1","source":{"kind":"arxiv","id":"2601.22900","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.22900","created_at":"2026-06-02T01:03:41Z"},{"alias_kind":"arxiv_version","alias_value":"2601.22900v2","created_at":"2026-06-02T01:03:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.22900","created_at":"2026-06-02T01:03:41Z"},{"alias_kind":"pith_short_12","alias_value":"FOBSE6NPZRW3","created_at":"2026-06-02T01:03:41Z"},{"alias_kind":"pith_short_16","alias_value":"FOBSE6NPZRW3OUDY","created_at":"2026-06-02T01:03:41Z"},{"alias_kind":"pith_short_8","alias_value":"FOBSE6NP","created_at":"2026-06-02T01:03:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:FOBSE6NPZRW3OUDYZRF5Y55NGS","target":"record","payload":{"canonical_record":{"source":{"id":"2601.22900","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-01-30T12:19:54Z","cross_cats_sorted":[],"title_canon_sha256":"2558c79dfdb7b6302e8ab18ffb9eb8d6b5a77ac7610eb8b6eeadb7e517b2aba4","abstract_canon_sha256":"ae1d0127dc5056920fae5bdf70e79b77e0dec444fbde45ddcfe449c5ec8a9085"},"schema_version":"1.0"},"canonical_sha256":"2b832279afcc6db75078cc4bdc77ad34b75f6ece0631d2a5a9ee1ea98caccfb1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:41.589369Z","signature_b64":"3wSm66E7oNxLT1qfWrJkvoK9qCzIrSedHiWX+URGSnuQOwnZ/8EGy7Wh0hNNnY6nD9VQkrg9QZdhZzxlK57ABg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2b832279afcc6db75078cc4bdc77ad34b75f6ece0631d2a5a9ee1ea98caccfb1","last_reissued_at":"2026-06-02T01:03:41.588856Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:41.588856Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2601.22900","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:03:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yLMKTGCsw/Qit3taKIgsoXcCZrDtNchoCJ9V6QiqouyVN60RaJ531u8azSv4tHyXcFf6fE+pC6cdejL2Be5CCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T11:55:16.861983Z"},"content_sha256":"9539c065242739794559b774f395f9021dad3d79c2f6d6a4f66940c7c19a0a95","schema_version":"1.0","event_id":"sha256:9539c065242739794559b774f395f9021dad3d79c2f6d6a4f66940c7c19a0a95"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:FOBSE6NPZRW3OUDYZRF5Y55NGS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MulFeRL: Enhancing Reinforcement Learning with Verbal Feedback in a Multi-turn Loop","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Haitao Li, Qingyao Ai, Xuancheng Li, YiqunLiu, Yujia Zhou","submitted_at":"2026-01-30T12:19:54Z","abstract_excerpt":"Reinforcement Learning with Verifiable Rewards (RLVR) is widely used to improve reasoning across domains, but outcome-only scalar rewards are often sparse and uninformative. This limitation is especially severe for failed samples, where scalar rewards indicate only that a solution is incorrect without explaining why the reasoning breaks down. In this paper, we leverage richer verbal feedback to guide RLVR on failed samples and convert feedback-induced progress into trainable learning signals. We propose MulFeRL (Multi-turn Feedback-guided Reinforcement Learning), a multi-turn, event-triggered "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.22900","kind":"arxiv","version":2},"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/2601.22900/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":"2026-06-02T01:03:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WNKNOzeE8tslXKVLGC2vRlQcbETlrdQgDspLWknaa/HkxpYXseBfD+i63n9yyiALFS2JiMHo/ThR8nLkaWezAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T11:55:16.862721Z"},"content_sha256":"7afec483b1f0f30cf2fd70c34a9de248c21384a52f4c3b8121df1df9b1813ea4","schema_version":"1.0","event_id":"sha256:7afec483b1f0f30cf2fd70c34a9de248c21384a52f4c3b8121df1df9b1813ea4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FOBSE6NPZRW3OUDYZRF5Y55NGS/bundle.json","state_url":"https://pith.science/pith/FOBSE6NPZRW3OUDYZRF5Y55NGS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FOBSE6NPZRW3OUDYZRF5Y55NGS/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":"2026-08-14T11:55:16Z","links":{"resolver":"https://pith.science/pith/FOBSE6NPZRW3OUDYZRF5Y55NGS","bundle":"https://pith.science/pith/FOBSE6NPZRW3OUDYZRF5Y55NGS/bundle.json","state":"https://pith.science/pith/FOBSE6NPZRW3OUDYZRF5Y55NGS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FOBSE6NPZRW3OUDYZRF5Y55NGS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:FOBSE6NPZRW3OUDYZRF5Y55NGS","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":"ae1d0127dc5056920fae5bdf70e79b77e0dec444fbde45ddcfe449c5ec8a9085","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-01-30T12:19:54Z","title_canon_sha256":"2558c79dfdb7b6302e8ab18ffb9eb8d6b5a77ac7610eb8b6eeadb7e517b2aba4"},"schema_version":"1.0","source":{"id":"2601.22900","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.22900","created_at":"2026-06-02T01:03:41Z"},{"alias_kind":"arxiv_version","alias_value":"2601.22900v2","created_at":"2026-06-02T01:03:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.22900","created_at":"2026-06-02T01:03:41Z"},{"alias_kind":"pith_short_12","alias_value":"FOBSE6NPZRW3","created_at":"2026-06-02T01:03:41Z"},{"alias_kind":"pith_short_16","alias_value":"FOBSE6NPZRW3OUDY","created_at":"2026-06-02T01:03:41Z"},{"alias_kind":"pith_short_8","alias_value":"FOBSE6NP","created_at":"2026-06-02T01:03:41Z"}],"graph_snapshots":[{"event_id":"sha256:7afec483b1f0f30cf2fd70c34a9de248c21384a52f4c3b8121df1df9b1813ea4","target":"graph","created_at":"2026-06-02T01:03:41Z","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/2601.22900/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement Learning with Verifiable Rewards (RLVR) is widely used to improve reasoning across domains, but outcome-only scalar rewards are often sparse and uninformative. This limitation is especially severe for failed samples, where scalar rewards indicate only that a solution is incorrect without explaining why the reasoning breaks down. In this paper, we leverage richer verbal feedback to guide RLVR on failed samples and convert feedback-induced progress into trainable learning signals. We propose MulFeRL (Multi-turn Feedback-guided Reinforcement Learning), a multi-turn, event-triggered ","authors_text":"Haitao Li, Qingyao Ai, Xuancheng Li, YiqunLiu, Yujia Zhou","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-01-30T12:19:54Z","title":"MulFeRL: Enhancing Reinforcement Learning with Verbal Feedback in a Multi-turn Loop"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.22900","kind":"arxiv","version":2},"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":"sha256:9539c065242739794559b774f395f9021dad3d79c2f6d6a4f66940c7c19a0a95","target":"record","created_at":"2026-06-02T01:03:41Z","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":"ae1d0127dc5056920fae5bdf70e79b77e0dec444fbde45ddcfe449c5ec8a9085","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-01-30T12:19:54Z","title_canon_sha256":"2558c79dfdb7b6302e8ab18ffb9eb8d6b5a77ac7610eb8b6eeadb7e517b2aba4"},"schema_version":"1.0","source":{"id":"2601.22900","kind":"arxiv","version":2}},"canonical_sha256":"2b832279afcc6db75078cc4bdc77ad34b75f6ece0631d2a5a9ee1ea98caccfb1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2b832279afcc6db75078cc4bdc77ad34b75f6ece0631d2a5a9ee1ea98caccfb1","first_computed_at":"2026-06-02T01:03:41.588856Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:03:41.588856Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3wSm66E7oNxLT1qfWrJkvoK9qCzIrSedHiWX+URGSnuQOwnZ/8EGy7Wh0hNNnY6nD9VQkrg9QZdhZzxlK57ABg==","signature_status":"signed_v1","signed_at":"2026-06-02T01:03:41.589369Z","signed_message":"canonical_sha256_bytes"},"source_id":"2601.22900","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9539c065242739794559b774f395f9021dad3d79c2f6d6a4f66940c7c19a0a95","sha256:7afec483b1f0f30cf2fd70c34a9de248c21384a52f4c3b8121df1df9b1813ea4"],"state_sha256":"bbadefa0792fb2956a8ff84c483f761d175544a3f58670c7ea43f2de11041bfd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ka7GL+d5Iomkbi/kFyVZ097Dndug+NhCo7ZuFgz/vmdSh3RS4jI1icJso/viTXVeijex65R9OuWxLuaebShdBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T11:55:16.868325Z","bundle_sha256":"931bcf64f277de7f8eb9a8c2e327b8deb0ae34d938b69dcdd044201e4f13a90e"}}