{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:HIVF3FAC45LT4UKQ67SZNX7QPR","short_pith_number":"pith:HIVF3FAC","canonical_record":{"source":{"id":"2402.12419","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-19T09:55:32Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"f385892499068b6d79bb841be25fbea669038e7b36fcc2b8645c6c09b3afaf70","abstract_canon_sha256":"22f76d9e34afcabed5bac54b1e424bc3549a3c740c4f54e37e941b0d118d6a2e"},"schema_version":"1.0"},"canonical_sha256":"3a2a5d9402e7573e5150f7e596dff07c7bf7fd3ddfc8def794e76e1f0d8c60e9","source":{"kind":"arxiv","id":"2402.12419","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.12419","created_at":"2026-07-05T07:47:11Z"},{"alias_kind":"arxiv_version","alias_value":"2402.12419v1","created_at":"2026-07-05T07:47:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12419","created_at":"2026-07-05T07:47:11Z"},{"alias_kind":"pith_short_12","alias_value":"HIVF3FAC45LT","created_at":"2026-07-05T07:47:11Z"},{"alias_kind":"pith_short_16","alias_value":"HIVF3FAC45LT4UKQ","created_at":"2026-07-05T07:47:11Z"},{"alias_kind":"pith_short_8","alias_value":"HIVF3FAC","created_at":"2026-07-05T07:47:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:HIVF3FAC45LT4UKQ67SZNX7QPR","target":"record","payload":{"canonical_record":{"source":{"id":"2402.12419","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-19T09:55:32Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"f385892499068b6d79bb841be25fbea669038e7b36fcc2b8645c6c09b3afaf70","abstract_canon_sha256":"22f76d9e34afcabed5bac54b1e424bc3549a3c740c4f54e37e941b0d118d6a2e"},"schema_version":"1.0"},"canonical_sha256":"3a2a5d9402e7573e5150f7e596dff07c7bf7fd3ddfc8def794e76e1f0d8c60e9","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:47:11.036310Z","signature_b64":"1UbH0DtLcEYiwgA4JEtRP3iplF7DfyMECp1Uss/Lap55zL0qBFrZ1gr7SVabw+6jy0R9GWAnkq3z3KskH6iPAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a2a5d9402e7573e5150f7e596dff07c7bf7fd3ddfc8def794e76e1f0d8c60e9","last_reissued_at":"2026-07-05T07:47:11.035843Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:47:11.035843Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.12419","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:47:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2uTh9xkZw/8CCEGE1zbNWTzk6Za4WcQj1vCrTebYjZ33IQs21gBC/yWURwkXIvmVBNB4BUwaJlAGUzlITFB6Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T18:31:39.824845Z"},"content_sha256":"8aa14861271565d4308db67fb0714e53ec95d85e70297ecd54d9da231e44561b","schema_version":"1.0","event_id":"sha256:8aa14861271565d4308db67fb0714e53ec95d85e70297ecd54d9da231e44561b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:HIVF3FAC45LT4UKQ67SZNX7QPR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"EBFT: Effective and Block-Wise Fine-Tuning for Sparse LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Fan Wu, Fei Chao, Lei Zhang, Rongrong Ji, Shengchuan Zhang, Song Guo, Xiawu Zheng, Yiyu Shi","submitted_at":"2024-02-19T09:55:32Z","abstract_excerpt":"Existing methods for fine-tuning sparse LLMs often suffer from resource-intensive requirements and high retraining costs. Additionally, many fine-tuning methods often rely on approximations or heuristic optimization strategies, which may lead to suboptimal solutions. To address these issues, we propose an efficient and fast framework for fine-tuning sparse LLMs based on minimizing reconstruction error. Our approach involves sampling a small dataset for calibration and utilizing backpropagation to iteratively optimize block-wise reconstruction error, on a block-by-block basis, aiming for optima"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12419","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/2402.12419/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:47:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"npmif6rXeQhO3Mz+0X0/kOT2QO+Ogh+i7MlWLbJFFvbqk4dCgq/L59PaOW3Dszke/NTnHEgyG0kyHSXf+8WzAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T18:31:39.825798Z"},"content_sha256":"68ae489dfdce17307d9157258e7e1033b8d855a8ae4bbe229fbde7a45e782d08","schema_version":"1.0","event_id":"sha256:68ae489dfdce17307d9157258e7e1033b8d855a8ae4bbe229fbde7a45e782d08"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HIVF3FAC45LT4UKQ67SZNX7QPR/bundle.json","state_url":"https://pith.science/pith/HIVF3FAC45LT4UKQ67SZNX7QPR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HIVF3FAC45LT4UKQ67SZNX7QPR/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-05T18:31:39Z","links":{"resolver":"https://pith.science/pith/HIVF3FAC45LT4UKQ67SZNX7QPR","bundle":"https://pith.science/pith/HIVF3FAC45LT4UKQ67SZNX7QPR/bundle.json","state":"https://pith.science/pith/HIVF3FAC45LT4UKQ67SZNX7QPR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HIVF3FAC45LT4UKQ67SZNX7QPR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:HIVF3FAC45LT4UKQ67SZNX7QPR","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"22f76d9e34afcabed5bac54b1e424bc3549a3c740c4f54e37e941b0d118d6a2e","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-19T09:55:32Z","title_canon_sha256":"f385892499068b6d79bb841be25fbea669038e7b36fcc2b8645c6c09b3afaf70"},"schema_version":"1.0","source":{"id":"2402.12419","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.12419","created_at":"2026-07-05T07:47:11Z"},{"alias_kind":"arxiv_version","alias_value":"2402.12419v1","created_at":"2026-07-05T07:47:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12419","created_at":"2026-07-05T07:47:11Z"},{"alias_kind":"pith_short_12","alias_value":"HIVF3FAC45LT","created_at":"2026-07-05T07:47:11Z"},{"alias_kind":"pith_short_16","alias_value":"HIVF3FAC45LT4UKQ","created_at":"2026-07-05T07:47:11Z"},{"alias_kind":"pith_short_8","alias_value":"HIVF3FAC","created_at":"2026-07-05T07:47:11Z"}],"graph_snapshots":[{"event_id":"sha256:68ae489dfdce17307d9157258e7e1033b8d855a8ae4bbe229fbde7a45e782d08","target":"graph","created_at":"2026-07-05T07:47:11Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2402.12419/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Existing methods for fine-tuning sparse LLMs often suffer from resource-intensive requirements and high retraining costs. Additionally, many fine-tuning methods often rely on approximations or heuristic optimization strategies, which may lead to suboptimal solutions. To address these issues, we propose an efficient and fast framework for fine-tuning sparse LLMs based on minimizing reconstruction error. Our approach involves sampling a small dataset for calibration and utilizing backpropagation to iteratively optimize block-wise reconstruction error, on a block-by-block basis, aiming for optima","authors_text":"Fan Wu, Fei Chao, Lei Zhang, Rongrong Ji, Shengchuan Zhang, Song Guo, Xiawu Zheng, Yiyu Shi","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-19T09:55:32Z","title":"EBFT: Effective and Block-Wise Fine-Tuning for Sparse LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12419","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8aa14861271565d4308db67fb0714e53ec95d85e70297ecd54d9da231e44561b","target":"record","created_at":"2026-07-05T07:47:11Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"22f76d9e34afcabed5bac54b1e424bc3549a3c740c4f54e37e941b0d118d6a2e","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-19T09:55:32Z","title_canon_sha256":"f385892499068b6d79bb841be25fbea669038e7b36fcc2b8645c6c09b3afaf70"},"schema_version":"1.0","source":{"id":"2402.12419","kind":"arxiv","version":1}},"canonical_sha256":"3a2a5d9402e7573e5150f7e596dff07c7bf7fd3ddfc8def794e76e1f0d8c60e9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3a2a5d9402e7573e5150f7e596dff07c7bf7fd3ddfc8def794e76e1f0d8c60e9","first_computed_at":"2026-07-05T07:47:11.035843Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:47:11.035843Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1UbH0DtLcEYiwgA4JEtRP3iplF7DfyMECp1Uss/Lap55zL0qBFrZ1gr7SVabw+6jy0R9GWAnkq3z3KskH6iPAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:47:11.036310Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.12419","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8aa14861271565d4308db67fb0714e53ec95d85e70297ecd54d9da231e44561b","sha256:68ae489dfdce17307d9157258e7e1033b8d855a8ae4bbe229fbde7a45e782d08"],"state_sha256":"adc08b1cc7d0b91bee7322e98184f33534061639dc9dcc280c1e896bc6621893"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rptpAB8xnI9db/43FnKK6aMHToczVpavkr8NEHI2zzE3SFoWveK2Z/pckSBHH/vBHuZdCm7MpAC7DXCUjErLCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T18:31:39.831605Z","bundle_sha256":"f9d4442e2c6506266854c2031fdc26413ed948f499f632de4d8d9297a8b57ab5"}}