{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:BFDCNHZG44MPXGEQBIHNYM4ZVS","short_pith_number":"pith:BFDCNHZG","canonical_record":{"source":{"id":"2410.20650","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-28T01:12:20Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"e8b600a81d455467a32c1492c8d888b0ce0232b149c02dbccacd10bf2dff3694","abstract_canon_sha256":"554cea1be40db3b1af396c214ee197e41f3897f509449d98364b34e167971680"},"schema_version":"1.0"},"canonical_sha256":"0946269f26e718fb98900a0edc3399ac89b4d275b8236502d13cc70a8710fbcc","source":{"kind":"arxiv","id":"2410.20650","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.20650","created_at":"2026-07-05T09:27:08Z"},{"alias_kind":"arxiv_version","alias_value":"2410.20650v1","created_at":"2026-07-05T09:27:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.20650","created_at":"2026-07-05T09:27:08Z"},{"alias_kind":"pith_short_12","alias_value":"BFDCNHZG44MP","created_at":"2026-07-05T09:27:08Z"},{"alias_kind":"pith_short_16","alias_value":"BFDCNHZG44MPXGEQ","created_at":"2026-07-05T09:27:08Z"},{"alias_kind":"pith_short_8","alias_value":"BFDCNHZG","created_at":"2026-07-05T09:27:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:BFDCNHZG44MPXGEQBIHNYM4ZVS","target":"record","payload":{"canonical_record":{"source":{"id":"2410.20650","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-28T01:12:20Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"e8b600a81d455467a32c1492c8d888b0ce0232b149c02dbccacd10bf2dff3694","abstract_canon_sha256":"554cea1be40db3b1af396c214ee197e41f3897f509449d98364b34e167971680"},"schema_version":"1.0"},"canonical_sha256":"0946269f26e718fb98900a0edc3399ac89b4d275b8236502d13cc70a8710fbcc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:27:08.822663Z","signature_b64":"zvNOSDLSnOJRDITWJvJFpY/0hUrQUct0QlMb72cB40w5JF10HFjHeLfOYuzqA5U5n4dOWddKJqNu2G9nqQDkDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0946269f26e718fb98900a0edc3399ac89b4d275b8236502d13cc70a8710fbcc","last_reissued_at":"2026-07-05T09:27:08.822199Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:27:08.822199Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.20650","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":"2026-07-05T09:27:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"64FYM58plawkISLPJgY10dQn/MU8Rb4gBlI26sn8OVDdUmFF44/mhB74k57sYtTgDF5Nriqbi8DOGVGBas5OCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T12:59:02.471084Z"},"content_sha256":"f16450bd986c869a4324b33b0d7a832f7b18940dd60b5a4c0f1949755fdbdb90","schema_version":"1.0","event_id":"sha256:f16450bd986c869a4324b33b0d7a832f7b18940dd60b5a4c0f1949755fdbdb90"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:BFDCNHZG44MPXGEQBIHNYM4ZVS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"NeuZip: Memory-Efficient Training and Inference with Dynamic Compression of Neural Networks","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Lili Mou, Yanshuai Cao, Yongchang Hao","submitted_at":"2024-10-28T01:12:20Z","abstract_excerpt":"The performance of neural networks improves when more parameters are used. However, the model sizes are constrained by the available on-device memory during training and inference. Although applying techniques like quantization can alleviate the constraint, they suffer from performance degradation. In this work, we introduce NeuZip, a new weight compression scheme based on the entropy of floating-point numbers in neural networks. With NeuZip, we are able to achieve memory-efficient training and inference without sacrificing performance. Notably, we significantly reduce the memory footprint of "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.20650","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/2410.20650/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":"2026-07-05T09:27:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oOaQ0d9vyBAvx0RGC1A6cS1y56jS1vyJsAb7Gfa+YlGsiFGzGfVd6CSjB56qxu6GAQ0yRkdyhQYPi1r/EDtlDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T12:59:02.471975Z"},"content_sha256":"1278699b46d8870d6ea3c1ea0a5a144d2e5490a489073ce6d4afa7614dd978df","schema_version":"1.0","event_id":"sha256:1278699b46d8870d6ea3c1ea0a5a144d2e5490a489073ce6d4afa7614dd978df"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BFDCNHZG44MPXGEQBIHNYM4ZVS/bundle.json","state_url":"https://pith.science/pith/BFDCNHZG44MPXGEQBIHNYM4ZVS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BFDCNHZG44MPXGEQBIHNYM4ZVS/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":"2026-08-10T12:59:02Z","links":{"resolver":"https://pith.science/pith/BFDCNHZG44MPXGEQBIHNYM4ZVS","bundle":"https://pith.science/pith/BFDCNHZG44MPXGEQBIHNYM4ZVS/bundle.json","state":"https://pith.science/pith/BFDCNHZG44MPXGEQBIHNYM4ZVS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BFDCNHZG44MPXGEQBIHNYM4ZVS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:BFDCNHZG44MPXGEQBIHNYM4ZVS","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":"554cea1be40db3b1af396c214ee197e41f3897f509449d98364b34e167971680","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-28T01:12:20Z","title_canon_sha256":"e8b600a81d455467a32c1492c8d888b0ce0232b149c02dbccacd10bf2dff3694"},"schema_version":"1.0","source":{"id":"2410.20650","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.20650","created_at":"2026-07-05T09:27:08Z"},{"alias_kind":"arxiv_version","alias_value":"2410.20650v1","created_at":"2026-07-05T09:27:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.20650","created_at":"2026-07-05T09:27:08Z"},{"alias_kind":"pith_short_12","alias_value":"BFDCNHZG44MP","created_at":"2026-07-05T09:27:08Z"},{"alias_kind":"pith_short_16","alias_value":"BFDCNHZG44MPXGEQ","created_at":"2026-07-05T09:27:08Z"},{"alias_kind":"pith_short_8","alias_value":"BFDCNHZG","created_at":"2026-07-05T09:27:08Z"}],"graph_snapshots":[{"event_id":"sha256:1278699b46d8870d6ea3c1ea0a5a144d2e5490a489073ce6d4afa7614dd978df","target":"graph","created_at":"2026-07-05T09:27:08Z","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/2410.20650/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The performance of neural networks improves when more parameters are used. However, the model sizes are constrained by the available on-device memory during training and inference. Although applying techniques like quantization can alleviate the constraint, they suffer from performance degradation. In this work, we introduce NeuZip, a new weight compression scheme based on the entropy of floating-point numbers in neural networks. With NeuZip, we are able to achieve memory-efficient training and inference without sacrificing performance. Notably, we significantly reduce the memory footprint of ","authors_text":"Lili Mou, Yanshuai Cao, Yongchang Hao","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-28T01:12:20Z","title":"NeuZip: Memory-Efficient Training and Inference with Dynamic Compression of Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.20650","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:f16450bd986c869a4324b33b0d7a832f7b18940dd60b5a4c0f1949755fdbdb90","target":"record","created_at":"2026-07-05T09:27:08Z","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":"554cea1be40db3b1af396c214ee197e41f3897f509449d98364b34e167971680","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-28T01:12:20Z","title_canon_sha256":"e8b600a81d455467a32c1492c8d888b0ce0232b149c02dbccacd10bf2dff3694"},"schema_version":"1.0","source":{"id":"2410.20650","kind":"arxiv","version":1}},"canonical_sha256":"0946269f26e718fb98900a0edc3399ac89b4d275b8236502d13cc70a8710fbcc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0946269f26e718fb98900a0edc3399ac89b4d275b8236502d13cc70a8710fbcc","first_computed_at":"2026-07-05T09:27:08.822199Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:27:08.822199Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zvNOSDLSnOJRDITWJvJFpY/0hUrQUct0QlMb72cB40w5JF10HFjHeLfOYuzqA5U5n4dOWddKJqNu2G9nqQDkDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:27:08.822663Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.20650","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f16450bd986c869a4324b33b0d7a832f7b18940dd60b5a4c0f1949755fdbdb90","sha256:1278699b46d8870d6ea3c1ea0a5a144d2e5490a489073ce6d4afa7614dd978df"],"state_sha256":"abe14945dcd7fd8fda406ad57fe705c0ffe8c2b1a6145fbed423c16fe3acf028"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mBN44W8TptCZcnaVPMaaHmZjIc0/V+B3XzEiA7xOmNW0cA4lw396uaH8u6i8GVhJM84a3LA/BkdCFH6jFm/XBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T12:59:02.479139Z","bundle_sha256":"8270a2127597ded27f317fecb486984688d3110b2bdebbca2c066a9a790b41ea"}}