{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:PI7DO5NWRCMOEM745WDBN73K7Y","short_pith_number":"pith:PI7DO5NW","canonical_record":{"source":{"id":"2511.15572","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-19T16:03:21Z","cross_cats_sorted":[],"title_canon_sha256":"f1c33b217f031c57c0dbc57daea71266dd59ce186ae3ee4396e325d8ce3cd7cb","abstract_canon_sha256":"119b4c5982a6a6abd874ce894e60e69d93cad4c8d187812e7ea4e5ba9e01feea"},"schema_version":"1.0"},"canonical_sha256":"7a3e3775b68898e233fced8616ff6afe29a87f6c6b5fc11c8108fda753e063fc","source":{"kind":"arxiv","id":"2511.15572","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2511.15572","created_at":"2026-05-27T02:05:08Z"},{"alias_kind":"arxiv_version","alias_value":"2511.15572v3","created_at":"2026-05-27T02:05:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.15572","created_at":"2026-05-27T02:05:08Z"},{"alias_kind":"pith_short_12","alias_value":"PI7DO5NWRCMO","created_at":"2026-05-27T02:05:08Z"},{"alias_kind":"pith_short_16","alias_value":"PI7DO5NWRCMOEM74","created_at":"2026-05-27T02:05:08Z"},{"alias_kind":"pith_short_8","alias_value":"PI7DO5NW","created_at":"2026-05-27T02:05:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:PI7DO5NWRCMOEM745WDBN73K7Y","target":"record","payload":{"canonical_record":{"source":{"id":"2511.15572","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-19T16:03:21Z","cross_cats_sorted":[],"title_canon_sha256":"f1c33b217f031c57c0dbc57daea71266dd59ce186ae3ee4396e325d8ce3cd7cb","abstract_canon_sha256":"119b4c5982a6a6abd874ce894e60e69d93cad4c8d187812e7ea4e5ba9e01feea"},"schema_version":"1.0"},"canonical_sha256":"7a3e3775b68898e233fced8616ff6afe29a87f6c6b5fc11c8108fda753e063fc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T02:05:08.394554Z","signature_b64":"f9FEdV6GGM1l0PNvn/uUo5g9qEWEFLB+TdFmKMP4/mDQoQz+ABpQfRwG20FK3pc6N/5S8mz+2Gif8kl6nGiNDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a3e3775b68898e233fced8616ff6afe29a87f6c6b5fc11c8108fda753e063fc","last_reissued_at":"2026-05-27T02:05:08.393777Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T02:05:08.393777Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2511.15572","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-27T02:05:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JNFpje8wAAeoxjZoIqJtDZhhtDrhNvS4P+AUnmn9lLL2aLJhdmj3L9ptmTznSyu6atyJ0XF/ouN1eYpK5qf8Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T07:39:55.613813Z"},"content_sha256":"ffba636bda9ca55ef9eaa36d64d7745d9db20cf62bbd555697b8da2d9d89e3a3","schema_version":"1.0","event_id":"sha256:ffba636bda9ca55ef9eaa36d64d7745d9db20cf62bbd555697b8da2d9d89e3a3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:PI7DO5NWRCMOEM745WDBN73K7Y","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"From Per-Image Low-Rank to Encoding Mismatch: Rethinking Feature Distillation in Vision Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"An encoding mismatch from per-image low-rank features and rotating dataset subspaces blocks feature-map distillation for compressing Vision Transformers.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bonan Xu, Huiyuan Tian, Shijian Li","submitted_at":"2025-11-19T16:03:21Z","abstract_excerpt":"Feature-map knowledge distillation (KD) transfers internal representations well between comparably sized Vision Transformers (ViTs), but it often fails in compression. We revisit this failure and uncover a paradox. Sample-wise SVD shows that each image is highly compressible, which seems to suggest that a narrow student with a linear projector should match the teacher \"in principle\". However, a dataset-level view contradicts this intuition: PCA shows that the teacher is a union of low-rank subspaces with significant subspace rotation across inputs. We further introduce token-level Spectral Ene"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We refer to this combined phenomenon as an encoding mismatch. We propose two minimal remedies, Lift or WideLast... On ImageNet-1K, these fixes revive feature KD for ViT compression, improving DeiT-Tiny distilled from CaiT-S24 from 74.86% to 77.53%/78.23% top-1 accuracy.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that the observed per-image low-rank structure, dataset-level subspace rotations, and token-level spectral energy patterns are the primary and causal reasons for feature-map KD failure in compression, rather than other factors such as optimization dynamics or capacity limits.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"The paper identifies an encoding mismatch in ViT feature distillation from per-image compressibility versus dataset subspace rotations and broad spectral energy patterns, proposing Lift and WideLast remedies that improve DeiT-Tiny accuracy from 74.86% to 77.53-78.23% on ImageNet-1K.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"An encoding mismatch from per-image low-rank features and rotating dataset subspaces blocks feature-map distillation for compressing Vision Transformers.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"131f4edf674c8156b3ccbdc07db66a596225e0a2b54b99c187b1abe371a97035"},"source":{"id":"2511.15572","kind":"arxiv","version":3},"verdict":{"id":"279a9c32-b2cf-4cb6-8506-5765e92914fc","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T20:19:20.348005Z","strongest_claim":"We refer to this combined phenomenon as an encoding mismatch. We propose two minimal remedies, Lift or WideLast... On ImageNet-1K, these fixes revive feature KD for ViT compression, improving DeiT-Tiny distilled from CaiT-S24 from 74.86% to 77.53%/78.23% top-1 accuracy.","one_line_summary":"The paper identifies an encoding mismatch in ViT feature distillation from per-image compressibility versus dataset subspace rotations and broad spectral energy patterns, proposing Lift and WideLast remedies that improve DeiT-Tiny accuracy from 74.86% to 77.53-78.23% on ImageNet-1K.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that the observed per-image low-rank structure, dataset-level subspace rotations, and token-level spectral energy patterns are the primary and causal reasons for feature-map KD failure in compression, rather than other factors such as optimization dynamics or capacity limits.","pith_extraction_headline":"An encoding mismatch from per-image low-rank features and rotating dataset subspaces blocks feature-map distillation for compressing Vision Transformers."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2511.15572/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":"279a9c32-b2cf-4cb6-8506-5765e92914fc"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-27T02:05:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5LFyvrL1pZDTSVs1wEJNPVyS+YedPKNOFN+r+Aijt8BG+YCdT7gUZqdHZVcYERaKQoaTlahtPxHy1cJJtnHeBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T07:39:55.614822Z"},"content_sha256":"15520201a5820d22c3fcad1583478a9b13e40e49eb6079ec1903c51bb6f1ae1b","schema_version":"1.0","event_id":"sha256:15520201a5820d22c3fcad1583478a9b13e40e49eb6079ec1903c51bb6f1ae1b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PI7DO5NWRCMOEM745WDBN73K7Y/bundle.json","state_url":"https://pith.science/pith/PI7DO5NWRCMOEM745WDBN73K7Y/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PI7DO5NWRCMOEM745WDBN73K7Y/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":"2026-06-10T07:39:55Z","links":{"resolver":"https://pith.science/pith/PI7DO5NWRCMOEM745WDBN73K7Y","bundle":"https://pith.science/pith/PI7DO5NWRCMOEM745WDBN73K7Y/bundle.json","state":"https://pith.science/pith/PI7DO5NWRCMOEM745WDBN73K7Y/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PI7DO5NWRCMOEM745WDBN73K7Y/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PI7DO5NWRCMOEM745WDBN73K7Y","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":"119b4c5982a6a6abd874ce894e60e69d93cad4c8d187812e7ea4e5ba9e01feea","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-19T16:03:21Z","title_canon_sha256":"f1c33b217f031c57c0dbc57daea71266dd59ce186ae3ee4396e325d8ce3cd7cb"},"schema_version":"1.0","source":{"id":"2511.15572","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2511.15572","created_at":"2026-05-27T02:05:08Z"},{"alias_kind":"arxiv_version","alias_value":"2511.15572v3","created_at":"2026-05-27T02:05:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.15572","created_at":"2026-05-27T02:05:08Z"},{"alias_kind":"pith_short_12","alias_value":"PI7DO5NWRCMO","created_at":"2026-05-27T02:05:08Z"},{"alias_kind":"pith_short_16","alias_value":"PI7DO5NWRCMOEM74","created_at":"2026-05-27T02:05:08Z"},{"alias_kind":"pith_short_8","alias_value":"PI7DO5NW","created_at":"2026-05-27T02:05:08Z"}],"graph_snapshots":[{"event_id":"sha256:15520201a5820d22c3fcad1583478a9b13e40e49eb6079ec1903c51bb6f1ae1b","target":"graph","created_at":"2026-05-27T02:05:08Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"We refer to this combined phenomenon as an encoding mismatch. We propose two minimal remedies, Lift or WideLast... On ImageNet-1K, these fixes revive feature KD for ViT compression, improving DeiT-Tiny distilled from CaiT-S24 from 74.86% to 77.53%/78.23% top-1 accuracy."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The assumption that the observed per-image low-rank structure, dataset-level subspace rotations, and token-level spectral energy patterns are the primary and causal reasons for feature-map KD failure in compression, rather than other factors such as optimization dynamics or capacity limits."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"The paper identifies an encoding mismatch in ViT feature distillation from per-image compressibility versus dataset subspace rotations and broad spectral energy patterns, proposing Lift and WideLast remedies that improve DeiT-Tiny accuracy from 74.86% to 77.53-78.23% on ImageNet-1K."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"An encoding mismatch from per-image low-rank features and rotating dataset subspaces blocks feature-map distillation for compressing Vision Transformers."}],"snapshot_sha256":"131f4edf674c8156b3ccbdc07db66a596225e0a2b54b99c187b1abe371a97035"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2511.15572/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Feature-map knowledge distillation (KD) transfers internal representations well between comparably sized Vision Transformers (ViTs), but it often fails in compression. We revisit this failure and uncover a paradox. Sample-wise SVD shows that each image is highly compressible, which seems to suggest that a narrow student with a linear projector should match the teacher \"in principle\". However, a dataset-level view contradicts this intuition: PCA shows that the teacher is a union of low-rank subspaces with significant subspace rotation across inputs. We further introduce token-level Spectral Ene","authors_text":"Bonan Xu, Huiyuan Tian, Shijian Li","cross_cats":[],"headline":"An encoding mismatch from per-image low-rank features and rotating dataset subspaces blocks feature-map distillation for compressing Vision Transformers.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-19T16:03:21Z","title":"From Per-Image Low-Rank to Encoding Mismatch: Rethinking Feature Distillation in Vision Transformers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.15572","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-17T20:19:20.348005Z","id":"279a9c32-b2cf-4cb6-8506-5765e92914fc","model_set":{"reader":"grok-4.3"},"one_line_summary":"The paper identifies an encoding mismatch in ViT feature distillation from per-image compressibility versus dataset subspace rotations and broad spectral energy patterns, proposing Lift and WideLast remedies that improve DeiT-Tiny accuracy from 74.86% to 77.53-78.23% on ImageNet-1K.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"An encoding mismatch from per-image low-rank features and rotating dataset subspaces blocks feature-map distillation for compressing Vision Transformers.","strongest_claim":"We refer to this combined phenomenon as an encoding mismatch. We propose two minimal remedies, Lift or WideLast... On ImageNet-1K, these fixes revive feature KD for ViT compression, improving DeiT-Tiny distilled from CaiT-S24 from 74.86% to 77.53%/78.23% top-1 accuracy.","weakest_assumption":"The assumption that the observed per-image low-rank structure, dataset-level subspace rotations, and token-level spectral energy patterns are the primary and causal reasons for feature-map KD failure in compression, rather than other factors such as optimization dynamics or capacity limits."}},"verdict_id":"279a9c32-b2cf-4cb6-8506-5765e92914fc"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ffba636bda9ca55ef9eaa36d64d7745d9db20cf62bbd555697b8da2d9d89e3a3","target":"record","created_at":"2026-05-27T02:05:08Z","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":"119b4c5982a6a6abd874ce894e60e69d93cad4c8d187812e7ea4e5ba9e01feea","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-19T16:03:21Z","title_canon_sha256":"f1c33b217f031c57c0dbc57daea71266dd59ce186ae3ee4396e325d8ce3cd7cb"},"schema_version":"1.0","source":{"id":"2511.15572","kind":"arxiv","version":3}},"canonical_sha256":"7a3e3775b68898e233fced8616ff6afe29a87f6c6b5fc11c8108fda753e063fc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7a3e3775b68898e233fced8616ff6afe29a87f6c6b5fc11c8108fda753e063fc","first_computed_at":"2026-05-27T02:05:08.393777Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-27T02:05:08.393777Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"f9FEdV6GGM1l0PNvn/uUo5g9qEWEFLB+TdFmKMP4/mDQoQz+ABpQfRwG20FK3pc6N/5S8mz+2Gif8kl6nGiNDw==","signature_status":"signed_v1","signed_at":"2026-05-27T02:05:08.394554Z","signed_message":"canonical_sha256_bytes"},"source_id":"2511.15572","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ffba636bda9ca55ef9eaa36d64d7745d9db20cf62bbd555697b8da2d9d89e3a3","sha256:15520201a5820d22c3fcad1583478a9b13e40e49eb6079ec1903c51bb6f1ae1b"],"state_sha256":"6674c5636601156dadae71123d3b0eadeb8ca8756cf192e92ff4d2537a55fe8e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XJOLb6POrNBd5ZoiAu+14etTbUP7tuCA0FDZ9Zq/pH8ggw4F2DjTroq8rQz+FMi/l81qtLV6p9kYOQTFF8qjCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-10T07:39:55.619036Z","bundle_sha256":"e1e781ba0b85817a88ba3c693e799910c68ee6f9ed2e4143a5f2ae3fd5c178c9"}}