{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:IJCMR42PT2F22PFSLZJD5LZ7TX","short_pith_number":"pith:IJCMR42P","schema_version":"1.0","canonical_sha256":"4244c8f34f9e8bad3cb25e523eaf3f9de15cdc0bc6187d93c810f5eecbfeefaa","source":{"kind":"arxiv","id":"2406.16069","version":3},"attestation_state":"computed","paper":{"title":"FastMem: Fast Memorization of Prompt Improves Context Awareness of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bo Tang, Feiyu Xiong, Junyi Zhu, Matthew B. Blaschko, Shuochen Liu, Tong Xu, Yibo Yan, Yu Yu, Zhiyu Li","submitted_at":"2024-06-23T10:36:35Z","abstract_excerpt":"Large language models (LLMs) excel in generating coherent text, but they often struggle with context awareness, leading to inaccuracies in tasks requiring faithful adherence to provided information. We introduce FastMem, a novel method designed to enhance instruction fine-tuned LLMs' context awareness through fast memorization of the prompt. FastMem maximizes the likelihood of the prompt before inference by updating only the last Feed-Forward Network (FFN) module. This targeted approach ensures efficient optimization without overfitting, significantly improving the model's ability to comprehen"},"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":"2406.16069","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-23T10:36:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"64bb3a5c777457d3289d9cd055a79a00457e9bb84db95983ad46d5d288c48b0f","abstract_canon_sha256":"2c783b4880b2aa520fa4f55db03d9745e8b8b243e80ed1f642b2e1214f559b7e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:18.716619Z","signature_b64":"8J6B2Th9ftaqctfPcj1F4dhyOFZswAhxVs8THdwzmsrlRMBSowHIueLYhNAnU+ch96sbFPtuRpufqD6/nsvTCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4244c8f34f9e8bad3cb25e523eaf3f9de15cdc0bc6187d93c810f5eecbfeefaa","last_reissued_at":"2026-07-05T09:16:18.716161Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:18.716161Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FastMem: Fast Memorization of Prompt Improves Context Awareness of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bo Tang, Feiyu Xiong, Junyi Zhu, Matthew B. Blaschko, Shuochen Liu, Tong Xu, Yibo Yan, Yu Yu, Zhiyu Li","submitted_at":"2024-06-23T10:36:35Z","abstract_excerpt":"Large language models (LLMs) excel in generating coherent text, but they often struggle with context awareness, leading to inaccuracies in tasks requiring faithful adherence to provided information. We introduce FastMem, a novel method designed to enhance instruction fine-tuned LLMs' context awareness through fast memorization of the prompt. FastMem maximizes the likelihood of the prompt before inference by updating only the last Feed-Forward Network (FFN) module. This targeted approach ensures efficient optimization without overfitting, significantly improving the model's ability to comprehen"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.16069","kind":"arxiv","version":3},"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/2406.16069/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":"2406.16069","created_at":"2026-07-05T09:16:18.716217+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.16069v3","created_at":"2026-07-05T09:16:18.716217+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.16069","created_at":"2026-07-05T09:16:18.716217+00:00"},{"alias_kind":"pith_short_12","alias_value":"IJCMR42PT2F2","created_at":"2026-07-05T09:16:18.716217+00:00"},{"alias_kind":"pith_short_16","alias_value":"IJCMR42PT2F22PFS","created_at":"2026-07-05T09:16:18.716217+00:00"},{"alias_kind":"pith_short_8","alias_value":"IJCMR42P","created_at":"2026-07-05T09:16:18.716217+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.23231","citing_title":"PERMA: Benchmarking Personalized Memory Agents via Event-Driven Preference and Realistic Task Environments","ref_index":89,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IJCMR42PT2F22PFSLZJD5LZ7TX","json":"https://pith.science/pith/IJCMR42PT2F22PFSLZJD5LZ7TX.json","graph_json":"https://pith.science/api/pith-number/IJCMR42PT2F22PFSLZJD5LZ7TX/graph.json","events_json":"https://pith.science/api/pith-number/IJCMR42PT2F22PFSLZJD5LZ7TX/events.json","paper":"https://pith.science/paper/IJCMR42P"},"agent_actions":{"view_html":"https://pith.science/pith/IJCMR42PT2F22PFSLZJD5LZ7TX","download_json":"https://pith.science/pith/IJCMR42PT2F22PFSLZJD5LZ7TX.json","view_paper":"https://pith.science/paper/IJCMR42P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.16069&json=true","fetch_graph":"https://pith.science/api/pith-number/IJCMR42PT2F22PFSLZJD5LZ7TX/graph.json","fetch_events":"https://pith.science/api/pith-number/IJCMR42PT2F22PFSLZJD5LZ7TX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IJCMR42PT2F22PFSLZJD5LZ7TX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IJCMR42PT2F22PFSLZJD5LZ7TX/action/storage_attestation","attest_author":"https://pith.science/pith/IJCMR42PT2F22PFSLZJD5LZ7TX/action/author_attestation","sign_citation":"https://pith.science/pith/IJCMR42PT2F22PFSLZJD5LZ7TX/action/citation_signature","submit_replication":"https://pith.science/pith/IJCMR42PT2F22PFSLZJD5LZ7TX/action/replication_record"}},"created_at":"2026-07-05T09:16:18.716217+00:00","updated_at":"2026-07-05T09:16:18.716217+00:00"}