{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JAME3G7RIMDRK7GOAWPLXEMNSP","short_pith_number":"pith:JAME3G7R","schema_version":"1.0","canonical_sha256":"48184d9bf14307157cce059ebb918d93e70f0573807eafe5645d5d758ee62990","source":{"kind":"arxiv","id":"2607.29065","version":1},"attestation_state":"computed","paper":{"title":"Tokenizer-Agnostic Engram Module","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hai Leong Chieu, Jia Peng Lim","submitted_at":"2026-07-31T06:32:15Z","abstract_excerpt":"Deepseek's Engram, a conditional memory module, was introduced to trade-off storage versus reasoning in large language models. However, the module relies on token-level $N$-gram hashing for Engram embedding lookup, introducing a tight coupling to the tokenizer used: a model with a different tokenizer would have to train its own Engram embeddings from scratch. To improve the reusability of Engram embeddings, we propose a change to the hashing routine, enabling compatibility between Engram models using different tokenizers. Instead of modelling disjoint $N$-gram spaces, we treat $N$-gram as a me"},"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":"2607.29065","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-31T06:32:15Z","cross_cats_sorted":[],"title_canon_sha256":"897f59af75406c8dc333821aaf4194471e4ef3bf3d2657ba5889ceeb46a4972b","abstract_canon_sha256":"1d7e26d0177c3ce65e85ad7deea7f8282bad16d996e65b31b932953682153a10"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-03T01:18:13.339537Z","signature_b64":"95s3DweP34Zup0gaV78lfOuqcPlLtBeXfpxV52FP7f+NfpxzZm7afIG1cr7aubjUPfcPdGDAycBavCfx8xdeCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"48184d9bf14307157cce059ebb918d93e70f0573807eafe5645d5d758ee62990","last_reissued_at":"2026-08-03T01:18:13.337917Z","signature_status":"signed_v1","first_computed_at":"2026-08-03T01:18:13.337917Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tokenizer-Agnostic Engram Module","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hai Leong Chieu, Jia Peng Lim","submitted_at":"2026-07-31T06:32:15Z","abstract_excerpt":"Deepseek's Engram, a conditional memory module, was introduced to trade-off storage versus reasoning in large language models. However, the module relies on token-level $N$-gram hashing for Engram embedding lookup, introducing a tight coupling to the tokenizer used: a model with a different tokenizer would have to train its own Engram embeddings from scratch. To improve the reusability of Engram embeddings, we propose a change to the hashing routine, enabling compatibility between Engram models using different tokenizers. Instead of modelling disjoint $N$-gram spaces, we treat $N$-gram as a me"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.29065","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/2607.29065/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":"2607.29065","created_at":"2026-08-03T01:18:13.338869+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.29065v1","created_at":"2026-08-03T01:18:13.338869+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.29065","created_at":"2026-08-03T01:18:13.338869+00:00"},{"alias_kind":"pith_short_12","alias_value":"JAME3G7RIMDR","created_at":"2026-08-03T01:18:13.338869+00:00"},{"alias_kind":"pith_short_16","alias_value":"JAME3G7RIMDRK7GO","created_at":"2026-08-03T01:18:13.338869+00:00"},{"alias_kind":"pith_short_8","alias_value":"JAME3G7R","created_at":"2026-08-03T01:18:13.338869+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/JAME3G7RIMDRK7GOAWPLXEMNSP","json":"https://pith.science/pith/JAME3G7RIMDRK7GOAWPLXEMNSP.json","graph_json":"https://pith.science/api/pith-number/JAME3G7RIMDRK7GOAWPLXEMNSP/graph.json","events_json":"https://pith.science/api/pith-number/JAME3G7RIMDRK7GOAWPLXEMNSP/events.json","paper":"https://pith.science/paper/JAME3G7R"},"agent_actions":{"view_html":"https://pith.science/pith/JAME3G7RIMDRK7GOAWPLXEMNSP","download_json":"https://pith.science/pith/JAME3G7RIMDRK7GOAWPLXEMNSP.json","view_paper":"https://pith.science/paper/JAME3G7R","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.29065&json=true","fetch_graph":"https://pith.science/api/pith-number/JAME3G7RIMDRK7GOAWPLXEMNSP/graph.json","fetch_events":"https://pith.science/api/pith-number/JAME3G7RIMDRK7GOAWPLXEMNSP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JAME3G7RIMDRK7GOAWPLXEMNSP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JAME3G7RIMDRK7GOAWPLXEMNSP/action/storage_attestation","attest_author":"https://pith.science/pith/JAME3G7RIMDRK7GOAWPLXEMNSP/action/author_attestation","sign_citation":"https://pith.science/pith/JAME3G7RIMDRK7GOAWPLXEMNSP/action/citation_signature","submit_replication":"https://pith.science/pith/JAME3G7RIMDRK7GOAWPLXEMNSP/action/replication_record"}},"created_at":"2026-08-03T01:18:13.338869+00:00","updated_at":"2026-08-03T01:18:13.338869+00:00"}