{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:52PJEORW64UIJCRZUIHCP2Y2CJ","short_pith_number":"pith:52PJEORW","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"},"canonical_sha256":"ee9e923a36f728848a39a20e27eb1a124d03ba10c81624abf903c7f6df821b4d","source":{"kind":"arxiv","id":"2512.07805","version":6},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.07805","created_at":"2026-05-17T23:39:16Z"},{"alias_kind":"arxiv_version","alias_value":"2512.07805v6","created_at":"2026-05-17T23:39:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.07805","created_at":"2026-05-17T23:39:16Z"},{"alias_kind":"pith_short_12","alias_value":"52PJEORW64UI","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"52PJEORW64UIJCRZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"52PJEORW","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:52PJEORW64UIJCRZUIHCP2Y2CJ","target":"record","payload":{"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"},"canonical_sha256":"ee9e923a36f728848a39a20e27eb1a124d03ba10c81624abf903c7f6df821b4d","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"},"source_kind":"arxiv","source_id":"2512.07805","source_version":6,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VBAiU/LHh3eeE2GUqxVHpHhB14vTUZfnCLn8IcKgxUq9Jxvtgg8Wb9t4dCPQVy6TGgw4DTbqB17Ga6e4fwGeCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-19T05:55:17.071551Z"},"content_sha256":"f16ae3c046168370b65e8f9a0464aa5a647b37236dd8dc4cf3ae71e3b94e2404","schema_version":"1.0","event_id":"sha256:f16ae3c046168370b65e8f9a0464aa5a647b37236dd8dc4cf3ae71e3b94e2404"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:52PJEORW64UIJCRZUIHCP2Y2CJ","target":"graph","payload":{"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"},"verdict_id":"7e2fece5-70b1-45d4-b106-30f9543d7e05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xa5n14BZyLY5X4SvKecOsMeG91rBkNgS50z6e+9jDmUbFwWzJAa7JsJxmXxUoiTY0zQypYFL7WbknvT8jpm4CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-19T05:55:17.072525Z"},"content_sha256":"016c1f96800c6d5ad2180bad74764d8a6a316eb2b37f4ab4503b42f386105705","schema_version":"1.0","event_id":"sha256:016c1f96800c6d5ad2180bad74764d8a6a316eb2b37f4ab4503b42f386105705"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ/bundle.json","state_url":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/52PJEORW64UIJCRZUIHCP2Y2CJ/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":"2026-05-19T05:55:17Z","links":{"resolver":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ","bundle":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ/bundle.json","state":"https://pith.science/pith/52PJEORW64UIJCRZUIHCP2Y2CJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/52PJEORW64UIJCRZUIHCP2Y2CJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:52PJEORW64UIJCRZUIHCP2Y2CJ","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":"02316880bb419c7d3ae0adc454e4dc4fbcf5dda91d5ba2caf15ac660ea9091a1","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-08T18:39:13Z","title_canon_sha256":"2acc2b0c3893d070c6a889405076313e8c7c970709a99d2d17c1afc6b7f1758d"},"schema_version":"1.0","source":{"id":"2512.07805","kind":"arxiv","version":6}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.07805","created_at":"2026-05-17T23:39:16Z"},{"alias_kind":"arxiv_version","alias_value":"2512.07805v6","created_at":"2026-05-17T23:39:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.07805","created_at":"2026-05-17T23:39:16Z"},{"alias_kind":"pith_short_12","alias_value":"52PJEORW64UI","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"52PJEORW64UIJCRZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"52PJEORW","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:016c1f96800c6d5ad2180bad74764d8a6a316eb2b37f4ab4503b42f386105705","target":"graph","created_at":"2026-05-17T23:39:16Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"GRAPE provides a principled design space for positional geometry in long-context models, subsuming RoPE and ALiBi as special cases."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"GRAPE models positions as group actions on features, recovering RoPE and ALiBi exactly while adding low-cost extensions for cross-feature coupling."}],"snapshot_sha256":"67e2f0594da038e93aa5f787cac1737a7b49ea4853111269620b9dcd2b94e971"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"f710d7659e99f99c228ebd75adfebff4569a4c9f0f3048e3681da7c398dec3b7"},"paper":{"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 ","authors_text":"Andrew Chi-Chih Yao, Huizhuo Yuan, Kangping Xu, Quanquan Gu, Yang Yuan, Yifan Zhang, Yifeng Liu, Zhen Qin, Zixiang Chen","cross_cats":["cs.AI","cs.CL"],"headline":"GRAPE models positions as group actions on features, recovering RoPE and ALiBi exactly while adding low-cost extensions for cross-feature coupling.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-08T18:39:13Z","title":"Group Representational Position Encoding"},"references":{"count":41,"internal_anchors":8,"resolved_work":41,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"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","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Round and round we go! what makes rotary positional encodings useful?","work_id":"e8226205-3ecb-4aa3-a204-f042c63750bb","year":2004},{"cited_arxiv_id":"2306.15595","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Extending Context Window of Large Language Models via Positional Interpolation","work_id":"c8b6df85-e7da-4bd8-90a4-d309cc2a0f60","year":2009},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":6,"title":"Contextual position encoding: Learning to count what’s important","work_id":"b3573740-adb9-4ab4-b55e-79456028e161","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":7,"title":"Transformer language models without positional encodings still learn positional information","work_id":"4ec68a0c-1f00-40be-8d28-278f90317c49","year":null}],"snapshot_sha256":"e5db67abf4f56546c1de092e98be3af59ed5d1f4fb23234a4ec32dfa86dcd0d1"},"source":{"id":"2512.07805","kind":"arxiv","version":6},"verdict":{"created_at":"2026-05-17T00:00:43.870987Z","id":"7e2fece5-70b1-45d4-b106-30f9543d7e05","model_set":{"reader":"grok-4.3"},"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","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.","strongest_claim":"GRAPE provides a principled design space for positional geometry in long-context models, subsuming RoPE and ALiBi as special cases.","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."}},"verdict_id":"7e2fece5-70b1-45d4-b106-30f9543d7e05"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f16ae3c046168370b65e8f9a0464aa5a647b37236dd8dc4cf3ae71e3b94e2404","target":"record","created_at":"2026-05-17T23:39:16Z","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":"02316880bb419c7d3ae0adc454e4dc4fbcf5dda91d5ba2caf15ac660ea9091a1","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-08T18:39:13Z","title_canon_sha256":"2acc2b0c3893d070c6a889405076313e8c7c970709a99d2d17c1afc6b7f1758d"},"schema_version":"1.0","source":{"id":"2512.07805","kind":"arxiv","version":6}},"canonical_sha256":"ee9e923a36f728848a39a20e27eb1a124d03ba10c81624abf903c7f6df821b4d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ee9e923a36f728848a39a20e27eb1a124d03ba10c81624abf903c7f6df821b4d","first_computed_at":"2026-05-17T23:39:16.917857Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:16.917857Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/wEYcgNXdoItcujtR1ImARBDM2hsfSBukszAUL/5QdfUoBrRUWb12nGwRBaj9VAmDrgV3ugWW4I4/dkf2jq5CQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:16.918493Z","signed_message":"canonical_sha256_bytes"},"source_id":"2512.07805","source_kind":"arxiv","source_version":6}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f16ae3c046168370b65e8f9a0464aa5a647b37236dd8dc4cf3ae71e3b94e2404","sha256:016c1f96800c6d5ad2180bad74764d8a6a316eb2b37f4ab4503b42f386105705"],"state_sha256":"2ad932123c71895461fdef6498b3491a21d3b826f494a0b8e84ed71fecd9d22a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zsglrsbbVHuRS6jUK+zcyTZV7YAYrgZDcujWNKcoeTOUw0nJsupogm7Nn1YL1aP88hwa7xdhHXEqJNpttjN/Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-19T05:55:17.076114Z","bundle_sha256":"6f0cac03e070b95a4209c69f6684de11f180cda5b226f21164e5cb1cd9e1e1df"}}