{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:52PJEORW64UIJCRZUIHCP2Y2CJ","short_pith_number":"pith:52PJEORW","schema_version":"1.0","canonical_sha256":"ee9e923a36f728848a39a20e27eb1a124d03ba10c81624abf903c7f6df821b4d","source":{"kind":"arxiv","id":"2512.07805","version":6},"attestation_state":"computed","paper":{"title":"Group Representational Position Encoding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"GRAPE models positions as group actions on features, recovering RoPE and ALiBi exactly while adding low-cost extensions for cross-feature coupling.","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Andrew Chi-Chih Yao, Huizhuo Yuan, Kangping Xu, Quanquan Gu, Yang Yuan, Yifan Zhang, Yifeng Liu, Zhen Qin, Zixiang Chen","submitted_at":"2025-12-08T18:39:13Z","abstract_excerpt":"We present GRAPE (Group Representational Position Encoding), a unified framework for positional encoding based on group actions. GRAPE unifies two families of mechanisms: (i) multiplicative rotations (Multiplicative GRAPE) in $\\operatorname{SO}(d)$ and (ii) additive logit biases (Additive GRAPE) arising from unipotent actions in the general linear group $\\mathrm{GL}$. In Multiplicative GRAPE, a position $n \\in \\mathbb{Z}$ (or $t \\in \\mathbb{R}$) acts as $\\mathbf{G}(n) = \\exp(n \\, \\omega \\, \\mathbf{L})$ with a rank-2 skew-symmetric generator $\\mathbf{L} \\in \\mathbb{R}^{d \\times d}$, yielding a "},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2512.07805","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-08T18:39:13Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"2acc2b0c3893d070c6a889405076313e8c7c970709a99d2d17c1afc6b7f1758d","abstract_canon_sha256":"02316880bb419c7d3ae0adc454e4dc4fbcf5dda91d5ba2caf15ac660ea9091a1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:16.918493Z","signature_b64":"/wEYcgNXdoItcujtR1ImARBDM2hsfSBukszAUL/5QdfUoBrRUWb12nGwRBaj9VAmDrgV3ugWW4I4/dkf2jq5CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ee9e923a36f728848a39a20e27eb1a124d03ba10c81624abf903c7f6df821b4d","last_reissued_at":"2026-05-17T23:39:16.917857Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:16.917857Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Group Representational Position Encoding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"GRAPE models positions as group actions on features, recovering RoPE and ALiBi exactly while adding low-cost extensions for cross-feature coupling.","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Andrew Chi-Chih Yao, Huizhuo Yuan, Kangping Xu, Quanquan Gu, Yang Yuan, Yifan Zhang, Yifeng Liu, Zhen Qin, Zixiang Chen","submitted_at":"2025-12-08T18:39:13Z","abstract_excerpt":"We present GRAPE (Group Representational Position Encoding), a unified framework for positional encoding based on group actions. GRAPE unifies two families of mechanisms: (i) multiplicative rotations (Multiplicative GRAPE) in $\\operatorname{SO}(d)$ and (ii) additive logit biases (Additive GRAPE) arising from unipotent actions in the general linear group $\\mathrm{GL}$. In Multiplicative GRAPE, a position $n \\in \\mathbb{Z}$ (or $t \\in \\mathbb{R}$) acts as $\\mathbf{G}(n) = \\exp(n \\, \\omega \\, \\mathbf{L})$ with a rank-2 skew-symmetric generator $\\mathbf{L} \\in \\mathbb{R}^{d \\times d}$, yielding a "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"GRAPE provides a principled design space for positional geometry in long-context models, subsuming RoPE and ALiBi as special cases.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the proposed extensions (learned commuting subspaces and compact non-commuting mixtures) will capture useful cross-subspace feature coupling in practice without introducing new optimization difficulties or performance regressions.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"GRAPE unifies RoPE and ALiBi as special cases of group actions on positions, providing a principled design space for positional encodings via SO(d) rotations and GL unipotent transformations.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"GRAPE models positions as group actions on features, recovering RoPE and ALiBi exactly while adding low-cost extensions for cross-feature coupling.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"67e2f0594da038e93aa5f787cac1737a7b49ea4853111269620b9dcd2b94e971"},"source":{"id":"2512.07805","kind":"arxiv","version":6},"verdict":{"id":"7e2fece5-70b1-45d4-b106-30f9543d7e05","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T00:00:43.870987Z","strongest_claim":"GRAPE provides a principled design space for positional geometry in long-context models, subsuming RoPE and ALiBi as special cases.","one_line_summary":"GRAPE unifies RoPE and ALiBi as special cases of group actions on positions, providing a principled design space for positional encodings via SO(d) rotations and GL unipotent transformations.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the proposed extensions (learned commuting subspaces and compact non-commuting mixtures) will capture useful cross-subspace feature coupling in practice without introducing new optimization difficulties or performance regressions.","pith_extraction_headline":"GRAPE models positions as group actions on features, recovering RoPE and ALiBi exactly while adding low-cost extensions for cross-feature coupling."},"references":{"count":41,"sample":[{"doi":"","year":2025,"title":"Round and round we go! what makes rotary positional encodings useful? In International Conference on Learning Representations (ICLR 2025),","work_id":"870f04a1-b332-40d5-a979-59692692e391","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2004,"title":"Round and round we go! what makes rotary positional encodings useful?","work_id":"e8226205-3ecb-4aa3-a204-f042c63750bb","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2009,"title":"Extending Context Window of Large Language Models via Positional Interpolation","work_id":"c8b6df85-e7da-4bd8-90a4-d309cc2a0f60","ref_index":4,"cited_arxiv_id":"2306.15595","is_internal_anchor":true},{"doi":"","year":null,"title":"Contextual position encoding: Learning to count what’s important","work_id":"b3573740-adb9-4ab4-b55e-79456028e161","ref_index":6,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Transformer language models without positional encodings still learn positional information","work_id":"4ec68a0c-1f00-40be-8d28-278f90317c49","ref_index":7,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":41,"snapshot_sha256":"e5db67abf4f56546c1de092e98be3af59ed5d1f4fb23234a4ec32dfa86dcd0d1","internal_anchors":8},"formal_canon":{"evidence_count":2,"snapshot_sha256":"f710d7659e99f99c228ebd75adfebff4569a4c9f0f3048e3681da7c398dec3b7"},"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":"2512.07805","created_at":"2026-05-17T23:39:16.917967+00:00"},{"alias_kind":"arxiv_version","alias_value":"2512.07805v6","created_at":"2026-05-17T23:39:16.917967+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.07805","created_at":"2026-05-17T23:39:16.917967+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.04217","citing_title":"Jordan-RoPE: Non-Semisimple Relative Positional Encoding via Complex Jordan Blocks","ref_index":13,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ","json":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ.json","graph_json":"https://pith.science/api/pith-number/52PJEORW64UIJCRZUIHCP2Y2CJ/graph.json","events_json":"https://pith.science/api/pith-number/52PJEORW64UIJCRZUIHCP2Y2CJ/events.json","paper":"https://pith.science/paper/52PJEORW"},"agent_actions":{"view_html":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ","download_json":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ.json","view_paper":"https://pith.science/paper/52PJEORW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2512.07805&json=true","fetch_graph":"https://pith.science/api/pith-number/52PJEORW64UIJCRZUIHCP2Y2CJ/graph.json","fetch_events":"https://pith.science/api/pith-number/52PJEORW64UIJCRZUIHCP2Y2CJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ/action/storage_attestation","attest_author":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ/action/author_attestation","sign_citation":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ/action/citation_signature","submit_replication":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ/action/replication_record"}},"created_at":"2026-05-17T23:39:16.917967+00:00","updated_at":"2026-05-17T23:39:16.917967+00:00"}