{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:RSZLNFYKALAHPMZ7ISJ24CMOQW","short_pith_number":"pith:RSZLNFYK","canonical_record":{"source":{"id":"2506.17286","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-15T07:19:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6d7b2cc9ee4761f9df5a158a97833b39071e47e59b1f67ed4c5a932bd8c50fc8","abstract_canon_sha256":"e5b9e272cce35077b2d4f3eea0c1a67ec24b17596faa95a2ad8fa0e22fbc4cc2"},"schema_version":"1.0"},"canonical_sha256":"8cb2b6970a02c077b33f4493ae098e858783b280bc2890ea867f34045d7209e2","source":{"kind":"arxiv","id":"2506.17286","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.17286","created_at":"2026-07-05T11:41:43Z"},{"alias_kind":"arxiv_version","alias_value":"2506.17286v2","created_at":"2026-07-05T11:41:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.17286","created_at":"2026-07-05T11:41:43Z"},{"alias_kind":"pith_short_12","alias_value":"RSZLNFYKALAH","created_at":"2026-07-05T11:41:43Z"},{"alias_kind":"pith_short_16","alias_value":"RSZLNFYKALAHPMZ7","created_at":"2026-07-05T11:41:43Z"},{"alias_kind":"pith_short_8","alias_value":"RSZLNFYK","created_at":"2026-07-05T11:41:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:RSZLNFYKALAHPMZ7ISJ24CMOQW","target":"record","payload":{"canonical_record":{"source":{"id":"2506.17286","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-15T07:19:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6d7b2cc9ee4761f9df5a158a97833b39071e47e59b1f67ed4c5a932bd8c50fc8","abstract_canon_sha256":"e5b9e272cce35077b2d4f3eea0c1a67ec24b17596faa95a2ad8fa0e22fbc4cc2"},"schema_version":"1.0"},"canonical_sha256":"8cb2b6970a02c077b33f4493ae098e858783b280bc2890ea867f34045d7209e2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:41:43.483157Z","signature_b64":"cJlmn/rBj4fNe25/hyLMRS3bITwtFlRXJFtq/AtwqJGsJyWTfzBqAUOairzLR6GS45UDAq59P3rle/ytml8DAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8cb2b6970a02c077b33f4493ae098e858783b280bc2890ea867f34045d7209e2","last_reissued_at":"2026-07-05T11:41:43.482610Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:41:43.482610Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.17286","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:41:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vYTeNlALPgnHlZ57qBLOvzvU2s9geZfZaCA793aNe7DT/VFGRKvzf5nPwkGU8rH6VsPRO8trhdJMxHCbm64YAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T03:15:12.660125Z"},"content_sha256":"9d216811fea3fd2cfc68dd86582ab66400dea50e8a9664ceb6f3dd4aaa541270","schema_version":"1.0","event_id":"sha256:9d216811fea3fd2cfc68dd86582ab66400dea50e8a9664ceb6f3dd4aaa541270"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:RSZLNFYKALAHPMZ7ISJ24CMOQW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"GTA: Grouped-head latenT Attention","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Cheng Deng, Haifeng Zhang, Jiwen Jiang, Jun Wang, Lei Chen, Lionel Ni, Luoyang Sun, Xinjian Wu","submitted_at":"2025-06-15T07:19:33Z","abstract_excerpt":"Attention mechanisms underpin the success of large language models (LLMs), yet their substantial computational and memory overhead poses challenges for optimizing efficiency and performance. A critical bottleneck arises as KV cache and attention computations scale rapidly with text length, challenging deployment on hardware with limited computational and memory resources. We observe that attention mechanisms exhibit substantial redundancy, since the KV cache can be significantly compressed and attention maps across heads display high similarity, revealing that much of the computation and stora"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.17286","kind":"arxiv","version":2},"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/2506.17286/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:41:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TU4k+NuFW1Tycs1ltdZVcQgKVuJogf2y36jv4LTLMdTa4gMRQpAW4zLh9N6ZMl5KbPAlgqxNFhmdgiaoYIwxDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T03:15:12.660647Z"},"content_sha256":"f11e5241e37e3b9c83cf9aafad9f31107d47c3bca20539175ad3bdeec7af4a3b","schema_version":"1.0","event_id":"sha256:f11e5241e37e3b9c83cf9aafad9f31107d47c3bca20539175ad3bdeec7af4a3b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RSZLNFYKALAHPMZ7ISJ24CMOQW/bundle.json","state_url":"https://pith.science/pith/RSZLNFYKALAHPMZ7ISJ24CMOQW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RSZLNFYKALAHPMZ7ISJ24CMOQW/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":"2026-08-05T03:15:12Z","links":{"resolver":"https://pith.science/pith/RSZLNFYKALAHPMZ7ISJ24CMOQW","bundle":"https://pith.science/pith/RSZLNFYKALAHPMZ7ISJ24CMOQW/bundle.json","state":"https://pith.science/pith/RSZLNFYKALAHPMZ7ISJ24CMOQW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RSZLNFYKALAHPMZ7ISJ24CMOQW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:RSZLNFYKALAHPMZ7ISJ24CMOQW","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":"e5b9e272cce35077b2d4f3eea0c1a67ec24b17596faa95a2ad8fa0e22fbc4cc2","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-15T07:19:33Z","title_canon_sha256":"6d7b2cc9ee4761f9df5a158a97833b39071e47e59b1f67ed4c5a932bd8c50fc8"},"schema_version":"1.0","source":{"id":"2506.17286","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.17286","created_at":"2026-07-05T11:41:43Z"},{"alias_kind":"arxiv_version","alias_value":"2506.17286v2","created_at":"2026-07-05T11:41:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.17286","created_at":"2026-07-05T11:41:43Z"},{"alias_kind":"pith_short_12","alias_value":"RSZLNFYKALAH","created_at":"2026-07-05T11:41:43Z"},{"alias_kind":"pith_short_16","alias_value":"RSZLNFYKALAHPMZ7","created_at":"2026-07-05T11:41:43Z"},{"alias_kind":"pith_short_8","alias_value":"RSZLNFYK","created_at":"2026-07-05T11:41:43Z"}],"graph_snapshots":[{"event_id":"sha256:f11e5241e37e3b9c83cf9aafad9f31107d47c3bca20539175ad3bdeec7af4a3b","target":"graph","created_at":"2026-07-05T11:41:43Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.17286/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Attention mechanisms underpin the success of large language models (LLMs), yet their substantial computational and memory overhead poses challenges for optimizing efficiency and performance. A critical bottleneck arises as KV cache and attention computations scale rapidly with text length, challenging deployment on hardware with limited computational and memory resources. We observe that attention mechanisms exhibit substantial redundancy, since the KV cache can be significantly compressed and attention maps across heads display high similarity, revealing that much of the computation and stora","authors_text":"Cheng Deng, Haifeng Zhang, Jiwen Jiang, Jun Wang, Lei Chen, Lionel Ni, Luoyang Sun, Xinjian Wu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-15T07:19:33Z","title":"GTA: Grouped-head latenT Attention"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.17286","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:9d216811fea3fd2cfc68dd86582ab66400dea50e8a9664ceb6f3dd4aaa541270","target":"record","created_at":"2026-07-05T11:41:43Z","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":"e5b9e272cce35077b2d4f3eea0c1a67ec24b17596faa95a2ad8fa0e22fbc4cc2","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-15T07:19:33Z","title_canon_sha256":"6d7b2cc9ee4761f9df5a158a97833b39071e47e59b1f67ed4c5a932bd8c50fc8"},"schema_version":"1.0","source":{"id":"2506.17286","kind":"arxiv","version":2}},"canonical_sha256":"8cb2b6970a02c077b33f4493ae098e858783b280bc2890ea867f34045d7209e2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8cb2b6970a02c077b33f4493ae098e858783b280bc2890ea867f34045d7209e2","first_computed_at":"2026-07-05T11:41:43.482610Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:41:43.482610Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cJlmn/rBj4fNe25/hyLMRS3bITwtFlRXJFtq/AtwqJGsJyWTfzBqAUOairzLR6GS45UDAq59P3rle/ytml8DAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:41:43.483157Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.17286","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9d216811fea3fd2cfc68dd86582ab66400dea50e8a9664ceb6f3dd4aaa541270","sha256:f11e5241e37e3b9c83cf9aafad9f31107d47c3bca20539175ad3bdeec7af4a3b"],"state_sha256":"62cf741d8995f801812daa4eabf640580cebc652a5f4f264931cfce8962f6db9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tVfW3uM32A98cFn00intkTwIG7NLQ2T85n8qivverH5s5Y8ruP3lVg8wan00IbkqQ/O8Tlkt971ty+hAv4R7Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T03:15:12.665466Z","bundle_sha256":"8282af041c84d3232a702de010060d4369742e24487219ad73974849798b23af"}}