{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:A3I55DDTOF2CDWISBRA7Q4RK43","short_pith_number":"pith:A3I55DDT","schema_version":"1.0","canonical_sha256":"06d1de8c73717421d9120c41f8722ae6dbd134fbdbd5c815230a3a00b4b7d0e6","source":{"kind":"arxiv","id":"2605.18856","version":1},"attestation_state":"computed","paper":{"title":"SPHERICAL KV: Angle-Domain Attention and Rate-Distortion Retention for Efficient Long-Context Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Aman Chadha, Amitava Das, Amit Dhanda, Anay Chauhan, Arion Das, Gurucharan Marthi Krishna Kumar, Vinija Jain","submitted_at":"2026-05-13T18:48:48Z","abstract_excerpt":"Long-context inference is increasingly constrained by the KV cache: resident memory grows with context length, and decoding becomes limited by repeated High Bandwidth Memory (HBM) streaming rather than arithmetic. Existing methods such as eviction, windowing, quantization, and offloading reduce footprint, but often leave the critical-path bottleneck only partially addressed, especially when compressed states must still be reconstructed into dense vectors during decoding.\n  We present Spherical KV, a long-context inference method that treats KV allocation as a rate-distortion problem grounded i"},"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":"2605.18856","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T18:48:48Z","cross_cats_sorted":["cs.CL","cs.IT","math.IT"],"title_canon_sha256":"4cfa7fee2d5850c35ddcb1a6439d8b614250a93cfcc5bbfe2ea9aef192cf2469","abstract_canon_sha256":"13296be9499ed0d9124fafd988338d50840eb181848e90f548933ffdd2ddb462"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:06:25.935993Z","signature_b64":"5+l0B29kg8fNc/TsOO84LWzPCdgo382PerWdTBG6SwcwstcURuaEgvn+3SXKAIkn3rTgwU/gttyrmVJpU94JCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"06d1de8c73717421d9120c41f8722ae6dbd134fbdbd5c815230a3a00b4b7d0e6","last_reissued_at":"2026-05-20T00:06:25.935094Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:06:25.935094Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SPHERICAL KV: Angle-Domain Attention and Rate-Distortion Retention for Efficient Long-Context Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Aman Chadha, Amitava Das, Amit Dhanda, Anay Chauhan, Arion Das, Gurucharan Marthi Krishna Kumar, Vinija Jain","submitted_at":"2026-05-13T18:48:48Z","abstract_excerpt":"Long-context inference is increasingly constrained by the KV cache: resident memory grows with context length, and decoding becomes limited by repeated High Bandwidth Memory (HBM) streaming rather than arithmetic. Existing methods such as eviction, windowing, quantization, and offloading reduce footprint, but often leave the critical-path bottleneck only partially addressed, especially when compressed states must still be reconstructed into dense vectors during decoding.\n  We present Spherical KV, a long-context inference method that treats KV allocation as a rate-distortion problem grounded i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18856","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/2605.18856/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":"2605.18856","created_at":"2026-05-20T00:06:25.935239+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.18856v1","created_at":"2026-05-20T00:06:25.935239+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18856","created_at":"2026-05-20T00:06:25.935239+00:00"},{"alias_kind":"pith_short_12","alias_value":"A3I55DDTOF2C","created_at":"2026-05-20T00:06:25.935239+00:00"},{"alias_kind":"pith_short_16","alias_value":"A3I55DDTOF2CDWIS","created_at":"2026-05-20T00:06:25.935239+00:00"},{"alias_kind":"pith_short_8","alias_value":"A3I55DDT","created_at":"2026-05-20T00:06:25.935239+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/A3I55DDTOF2CDWISBRA7Q4RK43","json":"https://pith.science/pith/A3I55DDTOF2CDWISBRA7Q4RK43.json","graph_json":"https://pith.science/api/pith-number/A3I55DDTOF2CDWISBRA7Q4RK43/graph.json","events_json":"https://pith.science/api/pith-number/A3I55DDTOF2CDWISBRA7Q4RK43/events.json","paper":"https://pith.science/paper/A3I55DDT"},"agent_actions":{"view_html":"https://pith.science/pith/A3I55DDTOF2CDWISBRA7Q4RK43","download_json":"https://pith.science/pith/A3I55DDTOF2CDWISBRA7Q4RK43.json","view_paper":"https://pith.science/paper/A3I55DDT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.18856&json=true","fetch_graph":"https://pith.science/api/pith-number/A3I55DDTOF2CDWISBRA7Q4RK43/graph.json","fetch_events":"https://pith.science/api/pith-number/A3I55DDTOF2CDWISBRA7Q4RK43/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A3I55DDTOF2CDWISBRA7Q4RK43/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A3I55DDTOF2CDWISBRA7Q4RK43/action/storage_attestation","attest_author":"https://pith.science/pith/A3I55DDTOF2CDWISBRA7Q4RK43/action/author_attestation","sign_citation":"https://pith.science/pith/A3I55DDTOF2CDWISBRA7Q4RK43/action/citation_signature","submit_replication":"https://pith.science/pith/A3I55DDTOF2CDWISBRA7Q4RK43/action/replication_record"}},"created_at":"2026-05-20T00:06:25.935239+00:00","updated_at":"2026-05-20T00:06:25.935239+00:00"}