{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:QNSAHV2DOYDNQU3YXGRWCPZPNK","short_pith_number":"pith:QNSAHV2D","schema_version":"1.0","canonical_sha256":"836403d7437606d85378b9a3613f2f6ab194fbe9668b84c932175f1fd88a7ae9","source":{"kind":"arxiv","id":"2606.08875","version":1},"attestation_state":"computed","paper":{"title":"Can the Environment Speak for Itself? $T^{2}$-GRPO: A Turn-Trajectory Group Relative Policy Optimization for Caregiver Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Amir M. Rahmani, Honghui Xu, Jiang Wu, Nikil Dutt, Pengfei Zhang, Wenjun Huang, Yutong Song","submitted_at":"2026-06-07T23:04:55Z","abstract_excerpt":"Optimizing large language models (LLMs) for long-horizon caregiver agents requires balancing delayed task objectives with immediate environment dynamics, such as patient distress and resistance. In dementia care, this balance is especially difficult: trajectory level rewards are too sparse for turn level credit assignment, while external LLM-based evaluators are costly and can misread fragmented or indirect patient responses. To address this issue, we propose \\textbf{T}urn-\\textbf{T}rajectory \\textbf{G}roup \\textbf{R}elative \\textbf{P}olicy \\textbf{O}ptimization (\\textbf{T$^{2}$-GRPO}), a fram"},"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":"2606.08875","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-07T23:04:55Z","cross_cats_sorted":[],"title_canon_sha256":"9028dc5c89f22ec4ba65ea7d1d662b3ef7e83419b4a975d29bee72744b661653","abstract_canon_sha256":"2714a04c71769a27ff5ed62e6e2c4dc556ef149d9c28665481369a88e56c755f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T02:07:44.781810Z","signature_b64":"txa8x7q+2ec8G2L8TqRRIAL+XWVL9kTmpu4F36fyntSSrujxVzojQMPWhO1ene3JWQtmwvSQy37mBxAhHLPeCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"836403d7437606d85378b9a3613f2f6ab194fbe9668b84c932175f1fd88a7ae9","last_reissued_at":"2026-06-09T02:07:44.781334Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T02:07:44.781334Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Can the Environment Speak for Itself? $T^{2}$-GRPO: A Turn-Trajectory Group Relative Policy Optimization for Caregiver Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Amir M. Rahmani, Honghui Xu, Jiang Wu, Nikil Dutt, Pengfei Zhang, Wenjun Huang, Yutong Song","submitted_at":"2026-06-07T23:04:55Z","abstract_excerpt":"Optimizing large language models (LLMs) for long-horizon caregiver agents requires balancing delayed task objectives with immediate environment dynamics, such as patient distress and resistance. In dementia care, this balance is especially difficult: trajectory level rewards are too sparse for turn level credit assignment, while external LLM-based evaluators are costly and can misread fragmented or indirect patient responses. To address this issue, we propose \\textbf{T}urn-\\textbf{T}rajectory \\textbf{G}roup \\textbf{R}elative \\textbf{P}olicy \\textbf{O}ptimization (\\textbf{T$^{2}$-GRPO}), a fram"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.08875","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/2606.08875/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":"2606.08875","created_at":"2026-06-09T02:07:44.781406+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.08875v1","created_at":"2026-06-09T02:07:44.781406+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.08875","created_at":"2026-06-09T02:07:44.781406+00:00"},{"alias_kind":"pith_short_12","alias_value":"QNSAHV2DOYDN","created_at":"2026-06-09T02:07:44.781406+00:00"},{"alias_kind":"pith_short_16","alias_value":"QNSAHV2DOYDNQU3Y","created_at":"2026-06-09T02:07:44.781406+00:00"},{"alias_kind":"pith_short_8","alias_value":"QNSAHV2D","created_at":"2026-06-09T02:07:44.781406+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/QNSAHV2DOYDNQU3YXGRWCPZPNK","json":"https://pith.science/pith/QNSAHV2DOYDNQU3YXGRWCPZPNK.json","graph_json":"https://pith.science/api/pith-number/QNSAHV2DOYDNQU3YXGRWCPZPNK/graph.json","events_json":"https://pith.science/api/pith-number/QNSAHV2DOYDNQU3YXGRWCPZPNK/events.json","paper":"https://pith.science/paper/QNSAHV2D"},"agent_actions":{"view_html":"https://pith.science/pith/QNSAHV2DOYDNQU3YXGRWCPZPNK","download_json":"https://pith.science/pith/QNSAHV2DOYDNQU3YXGRWCPZPNK.json","view_paper":"https://pith.science/paper/QNSAHV2D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.08875&json=true","fetch_graph":"https://pith.science/api/pith-number/QNSAHV2DOYDNQU3YXGRWCPZPNK/graph.json","fetch_events":"https://pith.science/api/pith-number/QNSAHV2DOYDNQU3YXGRWCPZPNK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QNSAHV2DOYDNQU3YXGRWCPZPNK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QNSAHV2DOYDNQU3YXGRWCPZPNK/action/storage_attestation","attest_author":"https://pith.science/pith/QNSAHV2DOYDNQU3YXGRWCPZPNK/action/author_attestation","sign_citation":"https://pith.science/pith/QNSAHV2DOYDNQU3YXGRWCPZPNK/action/citation_signature","submit_replication":"https://pith.science/pith/QNSAHV2DOYDNQU3YXGRWCPZPNK/action/replication_record"}},"created_at":"2026-06-09T02:07:44.781406+00:00","updated_at":"2026-06-09T02:07:44.781406+00:00"}