{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HRZGXEDUI32HUUTAABDSFM7V63","short_pith_number":"pith:HRZGXEDU","schema_version":"1.0","canonical_sha256":"3c726b907446f47a5260004722b3f5f6d77a2e090c4f74b570830be4b28965cc","source":{"kind":"arxiv","id":"2606.09871","version":1},"attestation_state":"computed","paper":{"title":"SD-GRPO: Verifiable Segment Decomposition for Long-Form Vision-Language Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Hannah Yun, Hyunwoong Kim, Jonggwon Park, Junhyun Park, Seongeun Lee","submitted_at":"2026-06-02T07:50:50Z","abstract_excerpt":"Group Relative Policy Optimization (GRPO) and its variants, originally developed for Large Language Models (LLMs), have recently been applied to Multimodal LLMs and produced strong results. However, their coarse-grained holistic credit assignment from a single scalar advantage underfits vision-language (VL) tasks, where outputs are often long-form responses grounded in semantically rich images. To address this limitation, we exploit a structured signal that single-scalar formulations discard: the natural segmentation of long-form VL outputs. Concretely, we propose Segment-Decomposed GRPO (SD-G"},"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":"2606.09871","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-06-02T07:50:50Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"579ea1a17a1fe89b0d517d543d0dd6ad69530e04dc4caff3db530fd6900506dd","abstract_canon_sha256":"3e7a623a16ef174ea81ef441310f35985b40eed183cb82744818560bc7737cca"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T00:08:30.577662Z","signature_b64":"K5sNpLgBCqDRq44obKcYZQXS6ox8Msf+ic7VtH5SdYV30AxrU+X9C34eYqMxgsJwAWOoQMmrO0E99/vF/mJVBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3c726b907446f47a5260004722b3f5f6d77a2e090c4f74b570830be4b28965cc","last_reissued_at":"2026-06-10T00:08:30.576846Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T00:08:30.576846Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SD-GRPO: Verifiable Segment Decomposition for Long-Form Vision-Language Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Hannah Yun, Hyunwoong Kim, Jonggwon Park, Junhyun Park, Seongeun Lee","submitted_at":"2026-06-02T07:50:50Z","abstract_excerpt":"Group Relative Policy Optimization (GRPO) and its variants, originally developed for Large Language Models (LLMs), have recently been applied to Multimodal LLMs and produced strong results. However, their coarse-grained holistic credit assignment from a single scalar advantage underfits vision-language (VL) tasks, where outputs are often long-form responses grounded in semantically rich images. To address this limitation, we exploit a structured signal that single-scalar formulations discard: the natural segmentation of long-form VL outputs. Concretely, we propose Segment-Decomposed GRPO (SD-G"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09871","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/2606.09871/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":"2606.09871","created_at":"2026-06-10T00:08:30.576968+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.09871v1","created_at":"2026-06-10T00:08:30.576968+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09871","created_at":"2026-06-10T00:08:30.576968+00:00"},{"alias_kind":"pith_short_12","alias_value":"HRZGXEDUI32H","created_at":"2026-06-10T00:08:30.576968+00:00"},{"alias_kind":"pith_short_16","alias_value":"HRZGXEDUI32HUUTA","created_at":"2026-06-10T00:08:30.576968+00:00"},{"alias_kind":"pith_short_8","alias_value":"HRZGXEDU","created_at":"2026-06-10T00:08:30.576968+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/HRZGXEDUI32HUUTAABDSFM7V63","json":"https://pith.science/pith/HRZGXEDUI32HUUTAABDSFM7V63.json","graph_json":"https://pith.science/api/pith-number/HRZGXEDUI32HUUTAABDSFM7V63/graph.json","events_json":"https://pith.science/api/pith-number/HRZGXEDUI32HUUTAABDSFM7V63/events.json","paper":"https://pith.science/paper/HRZGXEDU"},"agent_actions":{"view_html":"https://pith.science/pith/HRZGXEDUI32HUUTAABDSFM7V63","download_json":"https://pith.science/pith/HRZGXEDUI32HUUTAABDSFM7V63.json","view_paper":"https://pith.science/paper/HRZGXEDU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.09871&json=true","fetch_graph":"https://pith.science/api/pith-number/HRZGXEDUI32HUUTAABDSFM7V63/graph.json","fetch_events":"https://pith.science/api/pith-number/HRZGXEDUI32HUUTAABDSFM7V63/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HRZGXEDUI32HUUTAABDSFM7V63/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HRZGXEDUI32HUUTAABDSFM7V63/action/storage_attestation","attest_author":"https://pith.science/pith/HRZGXEDUI32HUUTAABDSFM7V63/action/author_attestation","sign_citation":"https://pith.science/pith/HRZGXEDUI32HUUTAABDSFM7V63/action/citation_signature","submit_replication":"https://pith.science/pith/HRZGXEDUI32HUUTAABDSFM7V63/action/replication_record"}},"created_at":"2026-06-10T00:08:30.576968+00:00","updated_at":"2026-06-10T00:08:30.576968+00:00"}