{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AUXTTGYU7I5SBXBR3L7AUZJY4C","short_pith_number":"pith:AUXTTGYU","schema_version":"1.0","canonical_sha256":"052f399b14fa3b20dc31dafe0a6538e09ce62c54083027cd4487d1488551bf76","source":{"kind":"arxiv","id":"2404.03865","version":1},"attestation_state":"computed","paper":{"title":"FFN-SkipLLM: A Hidden Gem for Autoregressive Decoding with Adaptive Feed Forward Skipping","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Aditya Akella, Ajay Jaiswal, Bodun Hu, Lu Yin, Shiwei Liu, Tianlong Chen, Yeonju Ro","submitted_at":"2024-04-05T02:35:43Z","abstract_excerpt":"Autoregressive Large Language Models (e.g., LLaMa, GPTs) are omnipresent achieving remarkable success in language understanding and generation. However, such impressive capability typically comes with a substantial model size, which presents significant challenges for autoregressive token-by-token generation. To mitigate computation overload incurred during generation, several early-exit and layer-dropping strategies have been proposed. Despite some promising success due to the redundancy across LLMs layers on metrics like Rough-L/BLUE, our careful knowledge-intensive evaluation unveils issues"},"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":"2404.03865","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-05T02:35:43Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d0a8896754337ef56dddeaeffcf4b7430bf9301002c785f6b769deb737cfe26b","abstract_canon_sha256":"f80c9ba4efc6eff5a3b6e944eb837f1995991ed961f11c344d512b7da3127eaa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:04:42.364034Z","signature_b64":"kCMqINy89h4xcJwGHxtm9G7eSTwgC68/1snfkTs6xcVpCHVupHk7YL12+mFLo3aIBKhadutq8eUExje1Rp5ODg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"052f399b14fa3b20dc31dafe0a6538e09ce62c54083027cd4487d1488551bf76","last_reissued_at":"2026-07-05T08:04:42.363511Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:04:42.363511Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FFN-SkipLLM: A Hidden Gem for Autoregressive Decoding with Adaptive Feed Forward Skipping","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Aditya Akella, Ajay Jaiswal, Bodun Hu, Lu Yin, Shiwei Liu, Tianlong Chen, Yeonju Ro","submitted_at":"2024-04-05T02:35:43Z","abstract_excerpt":"Autoregressive Large Language Models (e.g., LLaMa, GPTs) are omnipresent achieving remarkable success in language understanding and generation. However, such impressive capability typically comes with a substantial model size, which presents significant challenges for autoregressive token-by-token generation. To mitigate computation overload incurred during generation, several early-exit and layer-dropping strategies have been proposed. Despite some promising success due to the redundancy across LLMs layers on metrics like Rough-L/BLUE, our careful knowledge-intensive evaluation unveils issues"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.03865","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/2404.03865/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":"2404.03865","created_at":"2026-07-05T08:04:42.363581+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.03865v1","created_at":"2026-07-05T08:04:42.363581+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.03865","created_at":"2026-07-05T08:04:42.363581+00:00"},{"alias_kind":"pith_short_12","alias_value":"AUXTTGYU7I5S","created_at":"2026-07-05T08:04:42.363581+00:00"},{"alias_kind":"pith_short_16","alias_value":"AUXTTGYU7I5SBXBR","created_at":"2026-07-05T08:04:42.363581+00:00"},{"alias_kind":"pith_short_8","alias_value":"AUXTTGYU","created_at":"2026-07-05T08:04:42.363581+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2505.17138","citing_title":"RAP: Runtime Adaptive Pruning for LLM Inference","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AUXTTGYU7I5SBXBR3L7AUZJY4C","json":"https://pith.science/pith/AUXTTGYU7I5SBXBR3L7AUZJY4C.json","graph_json":"https://pith.science/api/pith-number/AUXTTGYU7I5SBXBR3L7AUZJY4C/graph.json","events_json":"https://pith.science/api/pith-number/AUXTTGYU7I5SBXBR3L7AUZJY4C/events.json","paper":"https://pith.science/paper/AUXTTGYU"},"agent_actions":{"view_html":"https://pith.science/pith/AUXTTGYU7I5SBXBR3L7AUZJY4C","download_json":"https://pith.science/pith/AUXTTGYU7I5SBXBR3L7AUZJY4C.json","view_paper":"https://pith.science/paper/AUXTTGYU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.03865&json=true","fetch_graph":"https://pith.science/api/pith-number/AUXTTGYU7I5SBXBR3L7AUZJY4C/graph.json","fetch_events":"https://pith.science/api/pith-number/AUXTTGYU7I5SBXBR3L7AUZJY4C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AUXTTGYU7I5SBXBR3L7AUZJY4C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AUXTTGYU7I5SBXBR3L7AUZJY4C/action/storage_attestation","attest_author":"https://pith.science/pith/AUXTTGYU7I5SBXBR3L7AUZJY4C/action/author_attestation","sign_citation":"https://pith.science/pith/AUXTTGYU7I5SBXBR3L7AUZJY4C/action/citation_signature","submit_replication":"https://pith.science/pith/AUXTTGYU7I5SBXBR3L7AUZJY4C/action/replication_record"}},"created_at":"2026-07-05T08:04:42.363581+00:00","updated_at":"2026-07-05T08:04:42.363581+00:00"}