{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:BEFQERO5LHCFQBA2YFUJCYKAMS","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":"68a69dbb823b02c186d7d5d111ff4d7260b7e85b049a205c70c05bac37fdfd7a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-30T16:17:21Z","title_canon_sha256":"5bc33420f94caab4993bbc8ef26ed4c8e36eec909e4e4de5786e73294215c2a3"},"schema_version":"1.0","source":{"id":"2509.26469","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.26469","created_at":"2026-05-27T01:05:38Z"},{"alias_kind":"arxiv_version","alias_value":"2509.26469v4","created_at":"2026-05-27T01:05:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.26469","created_at":"2026-05-27T01:05:38Z"},{"alias_kind":"pith_short_12","alias_value":"BEFQERO5LHCF","created_at":"2026-05-27T01:05:38Z"},{"alias_kind":"pith_short_16","alias_value":"BEFQERO5LHCFQBA2","created_at":"2026-05-27T01:05:38Z"},{"alias_kind":"pith_short_8","alias_value":"BEFQERO5","created_at":"2026-05-27T01:05:38Z"}],"graph_snapshots":[{"event_id":"sha256:86f5c7699821f138e291c637fd3caca1f52be93563f13cd479ab8ea173905b8e","target":"graph","created_at":"2026-05-27T01:05:38Z","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":"In VQ-VAE image compression, VQGAN image generation, and DAC speech coding tasks across various data sets, our proposed methods improve reconstruction and sample quality over alternative quantization approaches."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That modeling quantization distortion as an additive error vector (and the space-filling curve assignment) produces gradients that are both stable and unbiased enough to replace auxiliary losses and temperature schedules without degrading the hard forward-pass behavior."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"DiVeQ and SF-DiVeQ make vector quantization differentiable through error-vector reparameterization and space-filling curves, yielding better reconstruction in VQ-VAE, VQGAN, and DAC tasks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"DiVeQ makes vector quantization differentiable by reparameterizing it as an additive error vector."}],"snapshot_sha256":"45f129a11702c01e06b9a39c42cc34b3ca90abfbb74a2d4827eacf5ee120cad0"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"8e883833f512c9600f1ebc0eb73d006788d75bc76a8a761eb992d24a45ad7707"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2509.26469/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Vector quantization is common in deep models, yet its hard assignments block gradients and hinder end-to-end training. We propose DiVeQ, which treats quantization as adding an error vector that mimics the quantization distortion, keeping the forward pass hard while letting gradients flow. We also present a space-filling variant (SF-DiVeQ) that assigns input to a curve constructed by the lines connecting codewords, resulting in less quantization error and full codebook usage. Both methods train end-to-end without requiring auxiliary losses or temperature schedules. In VQ-VAE image compression, ","authors_text":"Arno Solin, Mohammad Hassan Vali, Tom B\\\"ackstr\\\"om","cross_cats":[],"headline":"DiVeQ makes vector quantization differentiable by reparameterizing it as an additive error vector.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-30T16:17:21Z","title":"DiVeQ: Differentiable Vector Quantization Using the Reparameterization Trick"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.26469","kind":"arxiv","version":4},"verdict":{"created_at":"2026-05-18T11:53:17.007194Z","id":"bea943f8-8673-4a7f-a5b1-c4808b6df145","model_set":{"reader":"grok-4.3"},"one_line_summary":"DiVeQ and SF-DiVeQ make vector quantization differentiable through error-vector reparameterization and space-filling curves, yielding better reconstruction in VQ-VAE, VQGAN, and DAC tasks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"DiVeQ makes vector quantization differentiable by reparameterizing it as an additive error vector.","strongest_claim":"In VQ-VAE image compression, VQGAN image generation, and DAC speech coding tasks across various data sets, our proposed methods improve reconstruction and sample quality over alternative quantization approaches.","weakest_assumption":"That modeling quantization distortion as an additive error vector (and the space-filling curve assignment) produces gradients that are both stable and unbiased enough to replace auxiliary losses and temperature schedules without degrading the hard forward-pass behavior."}},"verdict_id":"bea943f8-8673-4a7f-a5b1-c4808b6df145"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:4f6acd7d4398d095d91be6542af84d34a8adfa3e068c757856aba42f9fef1092","target":"record","created_at":"2026-05-27T01:05:38Z","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":"68a69dbb823b02c186d7d5d111ff4d7260b7e85b049a205c70c05bac37fdfd7a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-30T16:17:21Z","title_canon_sha256":"5bc33420f94caab4993bbc8ef26ed4c8e36eec909e4e4de5786e73294215c2a3"},"schema_version":"1.0","source":{"id":"2509.26469","kind":"arxiv","version":4}},"canonical_sha256":"090b0245dd59c458041ac16891614064a3e0d21d3464e2e844ffd25787c17f0b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"090b0245dd59c458041ac16891614064a3e0d21d3464e2e844ffd25787c17f0b","first_computed_at":"2026-05-27T01:05:38.245132Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-27T01:05:38.245132Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/l07bFFmCbg2c/cQ1kg6zGkfgC6haef9nV/96AIKC6O6NHJqG++YwiY9M8AXNFHPz/+CuCHYAOl1tOYTv8gPBQ==","signature_status":"signed_v1","signed_at":"2026-05-27T01:05:38.245807Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.26469","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4f6acd7d4398d095d91be6542af84d34a8adfa3e068c757856aba42f9fef1092","sha256:86f5c7699821f138e291c637fd3caca1f52be93563f13cd479ab8ea173905b8e"],"state_sha256":"cb6d8b79a52ad4cdf8dc0d6683a426afd9b09a0d04bbb09b255474cb5f984c2b"}