{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YZN2JPPE367O2C34K4BPAFT6DC","short_pith_number":"pith:YZN2JPPE","schema_version":"1.0","canonical_sha256":"c65ba4bde4dfbeed0b7c5702f0167e189ffe09b18203dbf9298cde4ca319c25b","source":{"kind":"arxiv","id":"2606.09916","version":1},"attestation_state":"computed","paper":{"title":"IntentKV: Cross-Turn Intent-Aware KV Cache Pruning for Agent Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jie Li, Jiong Lou, Junjie Li","submitted_at":"2026-06-06T15:54:48Z","abstract_excerpt":"Multi-turn LLM agents fan short queries into long trajectories of tool calls, search results, and intermediate reasoning. Both KV memory and KV read bandwidth grow by orders of magnitude across a single trajectory, making the key-value (KV) cache, not parameter compute, the dominant serving bottleneck for long-horizon agents. We introduce IntentKV, learned KV pruning that keeps the base LLM frozen. IntentKV maintains a session-level QueryMemory of cross-turn intent, scores live history tokens with a memory-attention rule, and adds a zero-initialized residual head with cross-attention over curr"},"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":"2606.09916","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-06T15:54:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"30ebc6fa9e90475f31a45170d5444e29268eb69f4845a8b116380c0cc6314cf0","abstract_canon_sha256":"29cc16966f973d1d8d8592bc1a9068ae73cd6fd16cd3733b2bef152eac4b7413"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T00:08:32.871416Z","signature_b64":"+WbJlrIO55zbNjVaF9Q4s6gPIWQzibPdK/N2xNTQvsvImKVIhCJyExo+2fFeyeqwK9dOXbVrwIS8HtQ0ad9fCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c65ba4bde4dfbeed0b7c5702f0167e189ffe09b18203dbf9298cde4ca319c25b","last_reissued_at":"2026-06-10T00:08:32.870619Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T00:08:32.870619Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"IntentKV: Cross-Turn Intent-Aware KV Cache Pruning for Agent Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jie Li, Jiong Lou, Junjie Li","submitted_at":"2026-06-06T15:54:48Z","abstract_excerpt":"Multi-turn LLM agents fan short queries into long trajectories of tool calls, search results, and intermediate reasoning. Both KV memory and KV read bandwidth grow by orders of magnitude across a single trajectory, making the key-value (KV) cache, not parameter compute, the dominant serving bottleneck for long-horizon agents. We introduce IntentKV, learned KV pruning that keeps the base LLM frozen. IntentKV maintains a session-level QueryMemory of cross-turn intent, scores live history tokens with a memory-attention rule, and adds a zero-initialized residual head with cross-attention over curr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09916","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/2606.09916/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":"2606.09916","created_at":"2026-06-10T00:08:32.870736+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.09916v1","created_at":"2026-06-10T00:08:32.870736+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09916","created_at":"2026-06-10T00:08:32.870736+00:00"},{"alias_kind":"pith_short_12","alias_value":"YZN2JPPE367O","created_at":"2026-06-10T00:08:32.870736+00:00"},{"alias_kind":"pith_short_16","alias_value":"YZN2JPPE367O2C34","created_at":"2026-06-10T00:08:32.870736+00:00"},{"alias_kind":"pith_short_8","alias_value":"YZN2JPPE","created_at":"2026-06-10T00:08:32.870736+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/YZN2JPPE367O2C34K4BPAFT6DC","json":"https://pith.science/pith/YZN2JPPE367O2C34K4BPAFT6DC.json","graph_json":"https://pith.science/api/pith-number/YZN2JPPE367O2C34K4BPAFT6DC/graph.json","events_json":"https://pith.science/api/pith-number/YZN2JPPE367O2C34K4BPAFT6DC/events.json","paper":"https://pith.science/paper/YZN2JPPE"},"agent_actions":{"view_html":"https://pith.science/pith/YZN2JPPE367O2C34K4BPAFT6DC","download_json":"https://pith.science/pith/YZN2JPPE367O2C34K4BPAFT6DC.json","view_paper":"https://pith.science/paper/YZN2JPPE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.09916&json=true","fetch_graph":"https://pith.science/api/pith-number/YZN2JPPE367O2C34K4BPAFT6DC/graph.json","fetch_events":"https://pith.science/api/pith-number/YZN2JPPE367O2C34K4BPAFT6DC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YZN2JPPE367O2C34K4BPAFT6DC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YZN2JPPE367O2C34K4BPAFT6DC/action/storage_attestation","attest_author":"https://pith.science/pith/YZN2JPPE367O2C34K4BPAFT6DC/action/author_attestation","sign_citation":"https://pith.science/pith/YZN2JPPE367O2C34K4BPAFT6DC/action/citation_signature","submit_replication":"https://pith.science/pith/YZN2JPPE367O2C34K4BPAFT6DC/action/replication_record"}},"created_at":"2026-06-10T00:08:32.870736+00:00","updated_at":"2026-06-10T00:08:32.870736+00:00"}