{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:ILHP74PGQZOS4SYTST5X4FFPFB","short_pith_number":"pith:ILHP74PG","canonical_record":{"source":{"id":"2601.09085","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-01-14T02:35:19Z","cross_cats_sorted":["cs.AI","cs.CL","cs.IR"],"title_canon_sha256":"6f520fcecbe667c45e4c757cea65e983242b195c1fb05ecf8f84113013888846","abstract_canon_sha256":"2404a6976c56612648fbac1a774c415748f525bb59a70ab3994ea333fc14c44d"},"schema_version":"1.0"},"canonical_sha256":"42cefff1e6865d2e4b1394fb7e14af284946465ef890543292792aa0c5bbd6d0","source":{"kind":"arxiv","id":"2601.09085","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.09085","created_at":"2026-06-09T01:05:12Z"},{"alias_kind":"arxiv_version","alias_value":"2601.09085v2","created_at":"2026-06-09T01:05:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.09085","created_at":"2026-06-09T01:05:12Z"},{"alias_kind":"pith_short_12","alias_value":"ILHP74PGQZOS","created_at":"2026-06-09T01:05:12Z"},{"alias_kind":"pith_short_16","alias_value":"ILHP74PGQZOS4SYT","created_at":"2026-06-09T01:05:12Z"},{"alias_kind":"pith_short_8","alias_value":"ILHP74PG","created_at":"2026-06-09T01:05:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:ILHP74PGQZOS4SYTST5X4FFPFB","target":"record","payload":{"canonical_record":{"source":{"id":"2601.09085","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-01-14T02:35:19Z","cross_cats_sorted":["cs.AI","cs.CL","cs.IR"],"title_canon_sha256":"6f520fcecbe667c45e4c757cea65e983242b195c1fb05ecf8f84113013888846","abstract_canon_sha256":"2404a6976c56612648fbac1a774c415748f525bb59a70ab3994ea333fc14c44d"},"schema_version":"1.0"},"canonical_sha256":"42cefff1e6865d2e4b1394fb7e14af284946465ef890543292792aa0c5bbd6d0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:05:12.604457Z","signature_b64":"cJ5FcDsyq1aGXcTQ0nIbEdUYln/vV4z6pe2yeqdXwX+dvjuWlgPb+VZianDwE2eYZzN/BhEihelBpgQbu0JpAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"42cefff1e6865d2e4b1394fb7e14af284946465ef890543292792aa0c5bbd6d0","last_reissued_at":"2026-06-09T01:05:12.603979Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:05:12.603979Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2601.09085","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":"2026-06-09T01:05:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CJhzH4FC7NSa1B8JFkcGOJcQg0WD/Pu5K05nv3kE0a1+wuA2IVj57F0vb5Zdl/5aXEZaoraanKUoTL9u3wm2AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T09:16:52.448699Z"},"content_sha256":"03271d28f36e696900f9fb94565dcb7c55aff683e6d1c544b7eca01b2c2da05d","schema_version":"1.0","event_id":"sha256:03271d28f36e696900f9fb94565dcb7c55aff683e6d1c544b7eca01b2c2da05d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:ILHP74PGQZOS4SYTST5X4FFPFB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MMR-GRPO: Accelerating GRPO-Style Training through Diversity-Aware Reward Reweighting","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.IR"],"primary_cat":"cs.LG","authors_text":"Kangda Wei, Ruihong Huang","submitted_at":"2026-01-14T02:35:19Z","abstract_excerpt":"Group Relative Policy Optimization (GRPO) has become a standard approach for training mathematical reasoning models; however, its reliance on multiple completions per prompt makes training computationally expensive. Although recent work has reduced the number of training steps required to reach peak performance, the overall wall-clock training time often remains unchanged or even increases due to higher per-step cost. We propose MMR-GRPO, which integrates Maximal Marginal Relevance to reweigh rewards based on completion diversity. Our key insight is that semantically redundant completions cont"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.09085","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/2601.09085/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":"2026-06-09T01:05:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JCUbr5xGUxffUUqBvKPJrySToXUCH7syyNQMJoZ3wUTCqbWG5gsaB5ZG1FIJnNDDCMxWxjicRBVFSyr14Z3jBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T09:16:52.449566Z"},"content_sha256":"21ca6570d72757d5b7d2c2ca85d10a0c80f28a4b477ea77d0d09c230dcda96aa","schema_version":"1.0","event_id":"sha256:21ca6570d72757d5b7d2c2ca85d10a0c80f28a4b477ea77d0d09c230dcda96aa"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ILHP74PGQZOS4SYTST5X4FFPFB/bundle.json","state_url":"https://pith.science/pith/ILHP74PGQZOS4SYTST5X4FFPFB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ILHP74PGQZOS4SYTST5X4FFPFB/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":"2026-06-10T09:16:52Z","links":{"resolver":"https://pith.science/pith/ILHP74PGQZOS4SYTST5X4FFPFB","bundle":"https://pith.science/pith/ILHP74PGQZOS4SYTST5X4FFPFB/bundle.json","state":"https://pith.science/pith/ILHP74PGQZOS4SYTST5X4FFPFB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ILHP74PGQZOS4SYTST5X4FFPFB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ILHP74PGQZOS4SYTST5X4FFPFB","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":"2404a6976c56612648fbac1a774c415748f525bb59a70ab3994ea333fc14c44d","cross_cats_sorted":["cs.AI","cs.CL","cs.IR"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-01-14T02:35:19Z","title_canon_sha256":"6f520fcecbe667c45e4c757cea65e983242b195c1fb05ecf8f84113013888846"},"schema_version":"1.0","source":{"id":"2601.09085","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.09085","created_at":"2026-06-09T01:05:12Z"},{"alias_kind":"arxiv_version","alias_value":"2601.09085v2","created_at":"2026-06-09T01:05:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.09085","created_at":"2026-06-09T01:05:12Z"},{"alias_kind":"pith_short_12","alias_value":"ILHP74PGQZOS","created_at":"2026-06-09T01:05:12Z"},{"alias_kind":"pith_short_16","alias_value":"ILHP74PGQZOS4SYT","created_at":"2026-06-09T01:05:12Z"},{"alias_kind":"pith_short_8","alias_value":"ILHP74PG","created_at":"2026-06-09T01:05:12Z"}],"graph_snapshots":[{"event_id":"sha256:21ca6570d72757d5b7d2c2ca85d10a0c80f28a4b477ea77d0d09c230dcda96aa","target":"graph","created_at":"2026-06-09T01:05:12Z","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/2601.09085/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Group Relative Policy Optimization (GRPO) has become a standard approach for training mathematical reasoning models; however, its reliance on multiple completions per prompt makes training computationally expensive. Although recent work has reduced the number of training steps required to reach peak performance, the overall wall-clock training time often remains unchanged or even increases due to higher per-step cost. We propose MMR-GRPO, which integrates Maximal Marginal Relevance to reweigh rewards based on completion diversity. Our key insight is that semantically redundant completions cont","authors_text":"Kangda Wei, Ruihong Huang","cross_cats":["cs.AI","cs.CL","cs.IR"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-01-14T02:35:19Z","title":"MMR-GRPO: Accelerating GRPO-Style Training through Diversity-Aware Reward Reweighting"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.09085","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":"sha256:03271d28f36e696900f9fb94565dcb7c55aff683e6d1c544b7eca01b2c2da05d","target":"record","created_at":"2026-06-09T01:05:12Z","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":"2404a6976c56612648fbac1a774c415748f525bb59a70ab3994ea333fc14c44d","cross_cats_sorted":["cs.AI","cs.CL","cs.IR"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-01-14T02:35:19Z","title_canon_sha256":"6f520fcecbe667c45e4c757cea65e983242b195c1fb05ecf8f84113013888846"},"schema_version":"1.0","source":{"id":"2601.09085","kind":"arxiv","version":2}},"canonical_sha256":"42cefff1e6865d2e4b1394fb7e14af284946465ef890543292792aa0c5bbd6d0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"42cefff1e6865d2e4b1394fb7e14af284946465ef890543292792aa0c5bbd6d0","first_computed_at":"2026-06-09T01:05:12.603979Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-09T01:05:12.603979Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cJ5FcDsyq1aGXcTQ0nIbEdUYln/vV4z6pe2yeqdXwX+dvjuWlgPb+VZianDwE2eYZzN/BhEihelBpgQbu0JpAw==","signature_status":"signed_v1","signed_at":"2026-06-09T01:05:12.604457Z","signed_message":"canonical_sha256_bytes"},"source_id":"2601.09085","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:03271d28f36e696900f9fb94565dcb7c55aff683e6d1c544b7eca01b2c2da05d","sha256:21ca6570d72757d5b7d2c2ca85d10a0c80f28a4b477ea77d0d09c230dcda96aa"],"state_sha256":"fd10b13a324aee7d15390fcc0a40f2f24056921e294c94f369f1f9e4c4790b58"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kfiL5MZ1pRvgQn64cBl/Ii7ErL6VxFtmkoxqsGQ0/4LpG+yqfL7VKmDmJrC8M8jy85tssMDiP2RHMAjehbBlBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-10T09:16:52.453842Z","bundle_sha256":"60aa661108d714d7660b797d85330227dcaf1a08706399cf842169b90c3b41d2"}}