{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:KZ26O2L2Z2BJB26ZRAG4DJ4Z3L","short_pith_number":"pith:KZ26O2L2","canonical_record":{"source":{"id":"2405.16012","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-25T02:30:46Z","cross_cats_sorted":[],"title_canon_sha256":"26a6efc2146342efc57a5f35886204d89292ef5422c2cdbd416fd9d37406e06d","abstract_canon_sha256":"357a22316533092c38d1b3abfb0e8470d8ff5f6a4fc1b3bae4068f728846f0ec"},"schema_version":"1.0"},"canonical_sha256":"5675e7697ace8290ebd9880dc1a799dac4f638ced8907c0507eecddc41f21106","source":{"kind":"arxiv","id":"2405.16012","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.16012","created_at":"2026-07-05T09:27:22Z"},{"alias_kind":"arxiv_version","alias_value":"2405.16012v3","created_at":"2026-07-05T09:27:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.16012","created_at":"2026-07-05T09:27:22Z"},{"alias_kind":"pith_short_12","alias_value":"KZ26O2L2Z2BJ","created_at":"2026-07-05T09:27:22Z"},{"alias_kind":"pith_short_16","alias_value":"KZ26O2L2Z2BJB26Z","created_at":"2026-07-05T09:27:22Z"},{"alias_kind":"pith_short_8","alias_value":"KZ26O2L2","created_at":"2026-07-05T09:27:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:KZ26O2L2Z2BJB26ZRAG4DJ4Z3L","target":"record","payload":{"canonical_record":{"source":{"id":"2405.16012","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-25T02:30:46Z","cross_cats_sorted":[],"title_canon_sha256":"26a6efc2146342efc57a5f35886204d89292ef5422c2cdbd416fd9d37406e06d","abstract_canon_sha256":"357a22316533092c38d1b3abfb0e8470d8ff5f6a4fc1b3bae4068f728846f0ec"},"schema_version":"1.0"},"canonical_sha256":"5675e7697ace8290ebd9880dc1a799dac4f638ced8907c0507eecddc41f21106","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:27:22.303794Z","signature_b64":"NrfsLUf/Ko9MbjQe5NEnmGOB3+hR8wAVgGEqxCLSyi0YmGcw3/IXit49E8rNEnBAmx5kDdCld+GI807iUzaoDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5675e7697ace8290ebd9880dc1a799dac4f638ced8907c0507eecddc41f21106","last_reissued_at":"2026-07-05T09:27:22.303013Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:27:22.303013Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.16012","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":"2026-07-05T09:27:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"i67Rhqw8AnSn9uahOj1uy0oZnBc3vpIce4adAxWcu7vk5mcYgYmHLKkNbCGA9nD32lUPvuAlnhR+HrxlLogcDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T14:23:19.118299Z"},"content_sha256":"0f2dfdbd39c1649e9021e7c745f54d8b1e6ee42721b6de80ba4e3379f261c2a8","schema_version":"1.0","event_id":"sha256:0f2dfdbd39c1649e9021e7c745f54d8b1e6ee42721b6de80ba4e3379f261c2a8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:KZ26O2L2Z2BJB26ZRAG4DJ4Z3L","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Pessimistic Backward Policy for GFlowNets","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Hyosoon Jang, Jinkyoo Park, Minsu Kim, Sungsoo Ahn, Yunhui Jang","submitted_at":"2024-05-25T02:30:46Z","abstract_excerpt":"This paper studies Generative Flow Networks (GFlowNets), which learn to sample objects proportionally to a given reward function through the trajectory of state transitions. In this work, we observe that GFlowNets tend to under-exploit the high-reward objects due to training on insufficient number of trajectories, which may lead to a large gap between the estimated flow and the (known) reward value. In response to this challenge, we propose a pessimistic backward policy for GFlowNets (PBP-GFN), which maximizes the observed flow to align closely with the true reward for the object. We extensive"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.16012","kind":"arxiv","version":3},"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/2405.16012/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":"2026-07-05T09:27:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mUh9V4S1Om8Ult0nb63R0SQB3tJM1A/5zj45c54BTGXXDXlgpqv/VESWqYSW6Ga4SU8+pOV+znQcIhwYjkKrDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T14:23:19.118798Z"},"content_sha256":"3a00d7a74fa52f8587520aaa0db3940d562d59008fc08457a4c49a049aaaf02b","schema_version":"1.0","event_id":"sha256:3a00d7a74fa52f8587520aaa0db3940d562d59008fc08457a4c49a049aaaf02b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KZ26O2L2Z2BJB26ZRAG4DJ4Z3L/bundle.json","state_url":"https://pith.science/pith/KZ26O2L2Z2BJB26ZRAG4DJ4Z3L/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KZ26O2L2Z2BJB26ZRAG4DJ4Z3L/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":"2026-08-08T14:23:19Z","links":{"resolver":"https://pith.science/pith/KZ26O2L2Z2BJB26ZRAG4DJ4Z3L","bundle":"https://pith.science/pith/KZ26O2L2Z2BJB26ZRAG4DJ4Z3L/bundle.json","state":"https://pith.science/pith/KZ26O2L2Z2BJB26ZRAG4DJ4Z3L/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KZ26O2L2Z2BJB26ZRAG4DJ4Z3L/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:KZ26O2L2Z2BJB26ZRAG4DJ4Z3L","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":"357a22316533092c38d1b3abfb0e8470d8ff5f6a4fc1b3bae4068f728846f0ec","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-25T02:30:46Z","title_canon_sha256":"26a6efc2146342efc57a5f35886204d89292ef5422c2cdbd416fd9d37406e06d"},"schema_version":"1.0","source":{"id":"2405.16012","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.16012","created_at":"2026-07-05T09:27:22Z"},{"alias_kind":"arxiv_version","alias_value":"2405.16012v3","created_at":"2026-07-05T09:27:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.16012","created_at":"2026-07-05T09:27:22Z"},{"alias_kind":"pith_short_12","alias_value":"KZ26O2L2Z2BJ","created_at":"2026-07-05T09:27:22Z"},{"alias_kind":"pith_short_16","alias_value":"KZ26O2L2Z2BJB26Z","created_at":"2026-07-05T09:27:22Z"},{"alias_kind":"pith_short_8","alias_value":"KZ26O2L2","created_at":"2026-07-05T09:27:22Z"}],"graph_snapshots":[{"event_id":"sha256:3a00d7a74fa52f8587520aaa0db3940d562d59008fc08457a4c49a049aaaf02b","target":"graph","created_at":"2026-07-05T09:27:22Z","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/2405.16012/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper studies Generative Flow Networks (GFlowNets), which learn to sample objects proportionally to a given reward function through the trajectory of state transitions. In this work, we observe that GFlowNets tend to under-exploit the high-reward objects due to training on insufficient number of trajectories, which may lead to a large gap between the estimated flow and the (known) reward value. In response to this challenge, we propose a pessimistic backward policy for GFlowNets (PBP-GFN), which maximizes the observed flow to align closely with the true reward for the object. We extensive","authors_text":"Hyosoon Jang, Jinkyoo Park, Minsu Kim, Sungsoo Ahn, Yunhui Jang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-25T02:30:46Z","title":"Pessimistic Backward Policy for GFlowNets"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.16012","kind":"arxiv","version":3},"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":"sha256:0f2dfdbd39c1649e9021e7c745f54d8b1e6ee42721b6de80ba4e3379f261c2a8","target":"record","created_at":"2026-07-05T09:27:22Z","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":"357a22316533092c38d1b3abfb0e8470d8ff5f6a4fc1b3bae4068f728846f0ec","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-25T02:30:46Z","title_canon_sha256":"26a6efc2146342efc57a5f35886204d89292ef5422c2cdbd416fd9d37406e06d"},"schema_version":"1.0","source":{"id":"2405.16012","kind":"arxiv","version":3}},"canonical_sha256":"5675e7697ace8290ebd9880dc1a799dac4f638ced8907c0507eecddc41f21106","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5675e7697ace8290ebd9880dc1a799dac4f638ced8907c0507eecddc41f21106","first_computed_at":"2026-07-05T09:27:22.303013Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:27:22.303013Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NrfsLUf/Ko9MbjQe5NEnmGOB3+hR8wAVgGEqxCLSyi0YmGcw3/IXit49E8rNEnBAmx5kDdCld+GI807iUzaoDw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:27:22.303794Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.16012","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0f2dfdbd39c1649e9021e7c745f54d8b1e6ee42721b6de80ba4e3379f261c2a8","sha256:3a00d7a74fa52f8587520aaa0db3940d562d59008fc08457a4c49a049aaaf02b"],"state_sha256":"df72eff2cca187117283edab7044f83c384036a04bd866a35c204ee89d4d9f2f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Hcqz/iHg4TERUqTipD/fXyEBJ/Pwc79UaRAtNjRDJBihDnm9iXaFWGZwSwvrWvOV+13fRqSoeZW4Kqegh3+LBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T14:23:19.122892Z","bundle_sha256":"90c3fd953f4450acfd77547f61d3a23c15807c0fba0d5a923da24ea6fd7de55d"}}