{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:5CRXFXR7FZKLAO35KHZYGZ2CGW","short_pith_number":"pith:5CRXFXR7","canonical_record":{"source":{"id":"2303.14409","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-25T09:22:59Z","cross_cats_sorted":[],"title_canon_sha256":"0549f0d5c4ae5be3336faf0db922424e35bfbdb78a95fd178efbcde9b38d916f","abstract_canon_sha256":"cde63f806711975455cdb1903c4993a112fc2b1c7b9478ebb66ca26a7466843b"},"schema_version":"1.0"},"canonical_sha256":"e8a372de3f2e54b03b7d51f3836742359f71b1d334f05483b11ff5312ab88dfc","source":{"kind":"arxiv","id":"2303.14409","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.14409","created_at":"2026-07-05T05:54:41Z"},{"alias_kind":"arxiv_version","alias_value":"2303.14409v1","created_at":"2026-07-05T05:54:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.14409","created_at":"2026-07-05T05:54:41Z"},{"alias_kind":"pith_short_12","alias_value":"5CRXFXR7FZKL","created_at":"2026-07-05T05:54:41Z"},{"alias_kind":"pith_short_16","alias_value":"5CRXFXR7FZKLAO35","created_at":"2026-07-05T05:54:41Z"},{"alias_kind":"pith_short_8","alias_value":"5CRXFXR7","created_at":"2026-07-05T05:54:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:5CRXFXR7FZKLAO35KHZYGZ2CGW","target":"record","payload":{"canonical_record":{"source":{"id":"2303.14409","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-25T09:22:59Z","cross_cats_sorted":[],"title_canon_sha256":"0549f0d5c4ae5be3336faf0db922424e35bfbdb78a95fd178efbcde9b38d916f","abstract_canon_sha256":"cde63f806711975455cdb1903c4993a112fc2b1c7b9478ebb66ca26a7466843b"},"schema_version":"1.0"},"canonical_sha256":"e8a372de3f2e54b03b7d51f3836742359f71b1d334f05483b11ff5312ab88dfc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:54:41.194430Z","signature_b64":"gUH4bgFtE8anljRV8o4JpVkkWjmlI5wW8ZUx6K4iRe0Sy2rmI8iSkm/+8trZQ72JC990HXZfTyeBnulCyytrDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e8a372de3f2e54b03b7d51f3836742359f71b1d334f05483b11ff5312ab88dfc","last_reissued_at":"2026-07-05T05:54:41.193996Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:54:41.193996Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2303.14409","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":"2026-07-05T05:54:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ER9Vhg6uRlwk6xvCK0yFz76m4YK2BnHxFNE/qn66gCa74VAj0jNasrhBFk0uNfORL+ts8Tvs6R+feif6TudxDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T11:34:05.636943Z"},"content_sha256":"a4e7fa6e6297a0f9c701adad67cd64e3f11574a344fc1f66c50ba02efb31faa8","schema_version":"1.0","event_id":"sha256:a4e7fa6e6297a0f9c701adad67cd64e3f11574a344fc1f66c50ba02efb31faa8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:5CRXFXR7FZKLAO35KHZYGZ2CGW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Vision Models Can Be Efficiently Specialized via Few-Shot Task-Aware Compression","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dan Alistarh, Denis Kuznedelev, Eldar Kurtic, Elias Frantar, Kimia Noorbakhsh, Sara Beery, Soroush Tabesh","submitted_at":"2023-03-25T09:22:59Z","abstract_excerpt":"Recent vision architectures and self-supervised training methods enable vision models that are extremely accurate and general, but come with massive parameter and computational costs. In practical settings, such as camera traps, users have limited resources, and may fine-tune a pretrained model on (often limited) data from a small set of specific categories of interest. These users may wish to make use of modern, highly-accurate models, but are often computationally constrained. To address this, we ask: can we quickly compress large generalist models into accurate and efficient specialists? Fo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.14409","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/2303.14409/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":"2026-07-05T05:54:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+NNaloB1YvGTIsuyQ6CG4XVLRldVTnSpGjN0BrEVhImDtxiZUmupCbI3EoTgRSLObMtDlRTpAs7y5E4npe3cDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T11:34:05.637532Z"},"content_sha256":"f709ad746fb3963481231712c02e549c3a6920f65c0db19257a18580ac4ae506","schema_version":"1.0","event_id":"sha256:f709ad746fb3963481231712c02e549c3a6920f65c0db19257a18580ac4ae506"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5CRXFXR7FZKLAO35KHZYGZ2CGW/bundle.json","state_url":"https://pith.science/pith/5CRXFXR7FZKLAO35KHZYGZ2CGW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5CRXFXR7FZKLAO35KHZYGZ2CGW/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":"2026-08-15T11:34:05Z","links":{"resolver":"https://pith.science/pith/5CRXFXR7FZKLAO35KHZYGZ2CGW","bundle":"https://pith.science/pith/5CRXFXR7FZKLAO35KHZYGZ2CGW/bundle.json","state":"https://pith.science/pith/5CRXFXR7FZKLAO35KHZYGZ2CGW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5CRXFXR7FZKLAO35KHZYGZ2CGW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:5CRXFXR7FZKLAO35KHZYGZ2CGW","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":"cde63f806711975455cdb1903c4993a112fc2b1c7b9478ebb66ca26a7466843b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-25T09:22:59Z","title_canon_sha256":"0549f0d5c4ae5be3336faf0db922424e35bfbdb78a95fd178efbcde9b38d916f"},"schema_version":"1.0","source":{"id":"2303.14409","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.14409","created_at":"2026-07-05T05:54:41Z"},{"alias_kind":"arxiv_version","alias_value":"2303.14409v1","created_at":"2026-07-05T05:54:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.14409","created_at":"2026-07-05T05:54:41Z"},{"alias_kind":"pith_short_12","alias_value":"5CRXFXR7FZKL","created_at":"2026-07-05T05:54:41Z"},{"alias_kind":"pith_short_16","alias_value":"5CRXFXR7FZKLAO35","created_at":"2026-07-05T05:54:41Z"},{"alias_kind":"pith_short_8","alias_value":"5CRXFXR7","created_at":"2026-07-05T05:54:41Z"}],"graph_snapshots":[{"event_id":"sha256:f709ad746fb3963481231712c02e549c3a6920f65c0db19257a18580ac4ae506","target":"graph","created_at":"2026-07-05T05:54:41Z","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/2303.14409/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent vision architectures and self-supervised training methods enable vision models that are extremely accurate and general, but come with massive parameter and computational costs. In practical settings, such as camera traps, users have limited resources, and may fine-tune a pretrained model on (often limited) data from a small set of specific categories of interest. These users may wish to make use of modern, highly-accurate models, but are often computationally constrained. To address this, we ask: can we quickly compress large generalist models into accurate and efficient specialists? Fo","authors_text":"Dan Alistarh, Denis Kuznedelev, Eldar Kurtic, Elias Frantar, Kimia Noorbakhsh, Sara Beery, Soroush Tabesh","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-25T09:22:59Z","title":"Vision Models Can Be Efficiently Specialized via Few-Shot Task-Aware Compression"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.14409","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:a4e7fa6e6297a0f9c701adad67cd64e3f11574a344fc1f66c50ba02efb31faa8","target":"record","created_at":"2026-07-05T05:54:41Z","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":"cde63f806711975455cdb1903c4993a112fc2b1c7b9478ebb66ca26a7466843b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-25T09:22:59Z","title_canon_sha256":"0549f0d5c4ae5be3336faf0db922424e35bfbdb78a95fd178efbcde9b38d916f"},"schema_version":"1.0","source":{"id":"2303.14409","kind":"arxiv","version":1}},"canonical_sha256":"e8a372de3f2e54b03b7d51f3836742359f71b1d334f05483b11ff5312ab88dfc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e8a372de3f2e54b03b7d51f3836742359f71b1d334f05483b11ff5312ab88dfc","first_computed_at":"2026-07-05T05:54:41.193996Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:54:41.193996Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"gUH4bgFtE8anljRV8o4JpVkkWjmlI5wW8ZUx6K4iRe0Sy2rmI8iSkm/+8trZQ72JC990HXZfTyeBnulCyytrDw==","signature_status":"signed_v1","signed_at":"2026-07-05T05:54:41.194430Z","signed_message":"canonical_sha256_bytes"},"source_id":"2303.14409","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a4e7fa6e6297a0f9c701adad67cd64e3f11574a344fc1f66c50ba02efb31faa8","sha256:f709ad746fb3963481231712c02e549c3a6920f65c0db19257a18580ac4ae506"],"state_sha256":"b39e60fcde60073e1e4d672749f1c92e39b2037f9dc53b0958c5619b7b2c2acc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Oy7lm8KnldX4zekqho6RcAvPbS5HZrsJpbqkFKVONc3A0sPMx0LIWS2oi6X9EaTW/bxaLzUB7aITBVh6NjvUDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T11:34:05.642219Z","bundle_sha256":"14e9a434ab68685b9a92b06dd3fe34df0b6013db8da39a296e2007c7bfa39d35"}}