{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:P7QGPVCKM4XZEWUGSVAA73R6ND","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":"235b19173813cb2ea6e84cdc9f64b204838db2137c949b3c33d41902e8f99d2a","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T09:43:32Z","title_canon_sha256":"e211e0398ded964dfe019f994ba2549ab933a57ad6491ec3b0dd3e2ac7a4a5f2"},"schema_version":"1.0","source":{"id":"2605.14631","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14631","created_at":"2026-05-17T23:39:03Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14631v1","created_at":"2026-05-17T23:39:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14631","created_at":"2026-05-17T23:39:03Z"},{"alias_kind":"pith_short_12","alias_value":"P7QGPVCKM4XZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"P7QGPVCKM4XZEWUG","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"P7QGPVCK","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:12434fb08f36708f5f2698678c89d0d370fed5278380a70606583ac6d64cc9ff","target":"graph","created_at":"2026-05-17T23:39:03Z","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":"selectively penalising uninformative transport paths through the learned potential yields consistent improvements in generation quality across fidelity and coverage metrics."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"that the learned scalar potential V_φ can reliably distinguish structurally coherent trajectories from degenerate ones online during training without introducing instability or adversarial dynamics between the networks."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"AGMs use a lightweight learned potential V_phi with stop-gradient to selectively weight informative bridge samples in generative model training, yielding better fidelity and coverage."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A lightweight learned scalar potential reweights bridge samples during training to penalize uninformative paths and lift generative quality."}],"snapshot_sha256":"f463664fb5728c39bf20814f73f05d53b46b56ce643fa7d0ec4e59c33e61867c"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"aab9721818cbeebe914f4e6b88a1c4cb5d8538a2cc792de479445f31dea07958"},"paper":{"abstract_excerpt":"We introduce Action-Inspired Generative Models (AGMs), a dual-network generative framework motivated by the observation that existing bridge-matching methods assign uniform regression weight to every stochastic transition in the transport landscape, regardless of whether a given bridge sample lies along a structurally coherent trajectory or a degenerate one. We address this by introducing a lightweight learned scalar potential $V_\\phi$ that scores bridge samples online and modulates the drift objective via importance weights derived through a stop-gradient barrier -- preventing adversarial fee","authors_text":"Debnath Pal, Eshwar R. A.","cross_cats":["cs.AI","cs.CV"],"headline":"A lightweight learned scalar potential reweights bridge samples during training to penalize uninformative paths and lift generative quality.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T09:43:32Z","title":"Action-Inspired Generative Models"},"references":{"count":13,"internal_anchors":0,"resolved_work":13,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Feynman and Albert R","work_id":"06298579-db66-4529-8249-f8361c0b93ea","year":1965},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Denoising diffusion probabilistic models.Advances in Neural Information Processing Systems, 33:6840–6851","work_id":"b707e044-45b2-45ad-a843-aed07db67eff","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Kingma, Abhishek Kumar, Stefano Ermon, and Ben Poole","work_id":"9e587d0c-ff4f-4efe-b98b-2101a602cc3f","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Denoising diffusion implicit models","work_id":"884d5b66-10ac-4ea9-88b4-470a6e3404fc","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Yaron Lipman, Ricky T. Q. Chen, Heli Ben-Hamu, Maximilian Nickel, and Matt Le. Flow matching for generative modeling. InInternational Conference on Learning Representations, 2023","work_id":"60f63f80-86d4-48dd-9f27-ac664f461905","year":2023}],"snapshot_sha256":"2bb7921727cc173b25aff5db84fb0d4bf085576be87aa1567e51a3744e8da43a"},"source":{"id":"2605.14631","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T04:40:59.250196Z","id":"acb33bca-0b5e-4940-9e95-f296b65bf1f7","model_set":{"reader":"grok-4.3"},"one_line_summary":"AGMs use a lightweight learned potential V_phi with stop-gradient to selectively weight informative bridge samples in generative model training, yielding better fidelity and coverage.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A lightweight learned scalar potential reweights bridge samples during training to penalize uninformative paths and lift generative quality.","strongest_claim":"selectively penalising uninformative transport paths through the learned potential yields consistent improvements in generation quality across fidelity and coverage metrics.","weakest_assumption":"that the learned scalar potential V_φ can reliably distinguish structurally coherent trajectories from degenerate ones online during training without introducing instability or adversarial dynamics between the networks."}},"verdict_id":"acb33bca-0b5e-4940-9e95-f296b65bf1f7"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:91ee91f2616ecb9f72f6b3a4ebebb156f1c08a13616c00483e73034f6e78e0c9","target":"record","created_at":"2026-05-17T23:39:03Z","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":"235b19173813cb2ea6e84cdc9f64b204838db2137c949b3c33d41902e8f99d2a","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T09:43:32Z","title_canon_sha256":"e211e0398ded964dfe019f994ba2549ab933a57ad6491ec3b0dd3e2ac7a4a5f2"},"schema_version":"1.0","source":{"id":"2605.14631","kind":"arxiv","version":1}},"canonical_sha256":"7fe067d44a672f925a8695400fee3e68d16b4dfc5fd7343b585421be5d8a4319","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7fe067d44a672f925a8695400fee3e68d16b4dfc5fd7343b585421be5d8a4319","first_computed_at":"2026-05-17T23:39:03.970811Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:03.970811Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Xn9xKdLg8St42NM9NnAQDHmnUPSmB76Yyr5Sma2ThFkQWxsXNXMSTvpJzC/42PUt3UY10C0oysikYAQbi7RRDA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:03.971598Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14631","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:91ee91f2616ecb9f72f6b3a4ebebb156f1c08a13616c00483e73034f6e78e0c9","sha256:12434fb08f36708f5f2698678c89d0d370fed5278380a70606583ac6d64cc9ff"],"state_sha256":"322913cfa9069979851525e248befcced53bddd72b6cc71d342f4f195e86f10b"}