{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:F5KVT75HMWYYHLWOUNHNBS5VFG","short_pith_number":"pith:F5KVT75H","schema_version":"1.0","canonical_sha256":"2f5559ffa765b183aecea34ed0cbb52989a086a21da910a5fbc1a581daef72a7","source":{"kind":"arxiv","id":"2607.26448","version":1},"attestation_state":"computed","paper":{"title":"Mergeable Model-Side Aggregation States for Long-Context Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dachuan Song, Junyu Yin, Xuan Wang, Zechen Hu","submitted_at":"2026-07-29T03:56:12Z","abstract_excerpt":"A known limitation of long-context language models is their increasingly unreliable performance in non-additive, set-based aggregation as context length grows. Examples include cardinality estimation, set relationships, and grouped statistics, which widely exist in logs, program outputs, tables, and multi-turn conversations. To provide the aggregation state required by these tasks, we introduce a model-side aggregation interface that maintains compact Hash-based HyperLogLog (HLL) sketch states alongside a frozen language model. While the model processes the context, an extractor maps each rele"},"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":"2607.26448","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-29T03:56:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3aabcf3f9676bbdc150156c276d3950b93b1a32d274742889ff086b0e7f76fb0","abstract_canon_sha256":"b72279f95ae36f6fd4c8f64d0bd59f6f2d15ccb4ed92d61835ad9803b99b2643"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f5559ffa765b183aecea34ed0cbb52989a086a21da910a5fbc1a581daef72a7","last_reissued_at":"2026-07-30T01:20:32.021585Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-30T01:20:32.021585Z"},"graph_snapshot":{"paper":{"title":"Mergeable Model-Side Aggregation States for Long-Context Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dachuan Song, Junyu Yin, Xuan Wang, Zechen Hu","submitted_at":"2026-07-29T03:56:12Z","abstract_excerpt":"A known limitation of long-context language models is their increasingly unreliable performance in non-additive, set-based aggregation as context length grows. Examples include cardinality estimation, set relationships, and grouped statistics, which widely exist in logs, program outputs, tables, and multi-turn conversations. To provide the aggregation state required by these tasks, we introduce a model-side aggregation interface that maintains compact Hash-based HyperLogLog (HLL) sketch states alongside a frozen language model. While the model processes the context, an extractor maps each rele"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.26448","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/2607.26448/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":"2607.26448","created_at":"2026-07-30T01:20:32.027618+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.26448v1","created_at":"2026-07-30T01:20:32.027618+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.26448","created_at":"2026-07-30T01:20:32.027618+00:00"},{"alias_kind":"pith_short_12","alias_value":"F5KVT75HMWYY","created_at":"2026-07-30T01:20:32.027618+00:00"},{"alias_kind":"pith_short_16","alias_value":"F5KVT75HMWYYHLWO","created_at":"2026-07-30T01:20:32.027618+00:00"},{"alias_kind":"pith_short_8","alias_value":"F5KVT75H","created_at":"2026-07-30T01:20:32.027618+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/F5KVT75HMWYYHLWOUNHNBS5VFG","json":"https://pith.science/pith/F5KVT75HMWYYHLWOUNHNBS5VFG.json","graph_json":"https://pith.science/api/pith-number/F5KVT75HMWYYHLWOUNHNBS5VFG/graph.json","events_json":"https://pith.science/api/pith-number/F5KVT75HMWYYHLWOUNHNBS5VFG/events.json","paper":"https://pith.science/paper/F5KVT75H"},"agent_actions":{"view_html":"https://pith.science/pith/F5KVT75HMWYYHLWOUNHNBS5VFG","download_json":"https://pith.science/pith/F5KVT75HMWYYHLWOUNHNBS5VFG.json","view_paper":"https://pith.science/paper/F5KVT75H","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.26448&json=true","fetch_graph":"https://pith.science/api/pith-number/F5KVT75HMWYYHLWOUNHNBS5VFG/graph.json","fetch_events":"https://pith.science/api/pith-number/F5KVT75HMWYYHLWOUNHNBS5VFG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F5KVT75HMWYYHLWOUNHNBS5VFG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F5KVT75HMWYYHLWOUNHNBS5VFG/action/storage_attestation","attest_author":"https://pith.science/pith/F5KVT75HMWYYHLWOUNHNBS5VFG/action/author_attestation","sign_citation":"https://pith.science/pith/F5KVT75HMWYYHLWOUNHNBS5VFG/action/citation_signature","submit_replication":"https://pith.science/pith/F5KVT75HMWYYHLWOUNHNBS5VFG/action/replication_record"}},"created_at":"2026-07-30T01:20:32.027618+00:00","updated_at":"2026-07-30T01:20:32.027618+00:00"}