{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:BUZEIDMR24MJWW53ZVZWC53A5B","short_pith_number":"pith:BUZEIDMR","canonical_record":{"source":{"id":"2606.29938","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-29T08:14:47Z","cross_cats_sorted":[],"title_canon_sha256":"465486504dcfb09fcb7395694373bd10ef0425ed8c425db23832d31834df5fe9","abstract_canon_sha256":"509b2d29c1592e21626ffb44c0a81d378b29be532c8f30aca3f12be4938d76e4"},"schema_version":"1.0"},"canonical_sha256":"0d32440d91d7189b5bbbcd73617760e84d841610b61a83ef1f664c0f5815a6fd","source":{"kind":"arxiv","id":"2606.29938","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.29938","created_at":"2026-06-30T02:17:41Z"},{"alias_kind":"arxiv_version","alias_value":"2606.29938v1","created_at":"2026-06-30T02:17:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.29938","created_at":"2026-06-30T02:17:41Z"},{"alias_kind":"pith_short_12","alias_value":"BUZEIDMR24MJ","created_at":"2026-06-30T02:17:41Z"},{"alias_kind":"pith_short_16","alias_value":"BUZEIDMR24MJWW53","created_at":"2026-06-30T02:17:41Z"},{"alias_kind":"pith_short_8","alias_value":"BUZEIDMR","created_at":"2026-06-30T02:17:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:BUZEIDMR24MJWW53ZVZWC53A5B","target":"record","payload":{"canonical_record":{"source":{"id":"2606.29938","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-29T08:14:47Z","cross_cats_sorted":[],"title_canon_sha256":"465486504dcfb09fcb7395694373bd10ef0425ed8c425db23832d31834df5fe9","abstract_canon_sha256":"509b2d29c1592e21626ffb44c0a81d378b29be532c8f30aca3f12be4938d76e4"},"schema_version":"1.0"},"canonical_sha256":"0d32440d91d7189b5bbbcd73617760e84d841610b61a83ef1f664c0f5815a6fd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T02:17:41.762691Z","signature_b64":"dZYAFh8lyPbuOZv4YSlBBJo5utkxnou4y4wlPQWNKY4/eIt+/7fdGZ95pYvk4hQSezQXlNBF9sbnARJXbnGJDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0d32440d91d7189b5bbbcd73617760e84d841610b61a83ef1f664c0f5815a6fd","last_reissued_at":"2026-06-30T02:17:41.762085Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T02:17:41.762085Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.29938","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":"2026-06-30T02:17:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XJcr+ASafZEcpiX9BJT+iMdK92A2ODhgSaVM1+wHCqVS5j99ugrf2OVLMhh2oZJgFmkSS9ZwYrKPtzZQ94zzAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-03T18:14:16.762752Z"},"content_sha256":"d879d8e0658381cd2823e73fcd2f6d46909f9a5f39eb187c233671488842a8d1","schema_version":"1.0","event_id":"sha256:d879d8e0658381cd2823e73fcd2f6d46909f9a5f39eb187c233671488842a8d1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:BUZEIDMR24MJWW53ZVZWC53A5B","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LatentRevise: Learning from Zero-Hit Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Guangtao Zhai, Jing Bai, Qi Jia, Xueting Han, Yiqiu Guo","submitted_at":"2026-06-29T08:14:47Z","abstract_excerpt":"Reinforcement learning with verifiable rewards (RLVR) is bottlenecked by hard prompts on which correct trajectories have low probability, so sampling misses them within a practical budget and leaves the policy update with little useful signal. We frame such zero-hit prompts as RLVR's sampling frontier, where new reasoning behavior is most valuable yet least likely to be sampled. Importantly, failed rollouts can be informative: they expose where the model's reasoning went wrong. We introduce LatentRevise, a first-order latent revision method that recovers training signal for this zero-hit regim"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.29938","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/2606.29938/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":"2026-06-30T02:17:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3eb/n2wfWzQZu2hdfnPdGuZSNCk5lgCoKvkywZ93j2d3iZu0ND76fWLzTwAViUyJVbrswqF3yqqdeTkIkyjMBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-03T18:14:16.763139Z"},"content_sha256":"898c79b6727d6676fee86b938633157e37af7c2ec6bcac80eb682f81ae2d07e7","schema_version":"1.0","event_id":"sha256:898c79b6727d6676fee86b938633157e37af7c2ec6bcac80eb682f81ae2d07e7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BUZEIDMR24MJWW53ZVZWC53A5B/bundle.json","state_url":"https://pith.science/pith/BUZEIDMR24MJWW53ZVZWC53A5B/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BUZEIDMR24MJWW53ZVZWC53A5B/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":"2026-07-03T18:14:16Z","links":{"resolver":"https://pith.science/pith/BUZEIDMR24MJWW53ZVZWC53A5B","bundle":"https://pith.science/pith/BUZEIDMR24MJWW53ZVZWC53A5B/bundle.json","state":"https://pith.science/pith/BUZEIDMR24MJWW53ZVZWC53A5B/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BUZEIDMR24MJWW53ZVZWC53A5B/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:BUZEIDMR24MJWW53ZVZWC53A5B","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":"509b2d29c1592e21626ffb44c0a81d378b29be532c8f30aca3f12be4938d76e4","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-29T08:14:47Z","title_canon_sha256":"465486504dcfb09fcb7395694373bd10ef0425ed8c425db23832d31834df5fe9"},"schema_version":"1.0","source":{"id":"2606.29938","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.29938","created_at":"2026-06-30T02:17:41Z"},{"alias_kind":"arxiv_version","alias_value":"2606.29938v1","created_at":"2026-06-30T02:17:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.29938","created_at":"2026-06-30T02:17:41Z"},{"alias_kind":"pith_short_12","alias_value":"BUZEIDMR24MJ","created_at":"2026-06-30T02:17:41Z"},{"alias_kind":"pith_short_16","alias_value":"BUZEIDMR24MJWW53","created_at":"2026-06-30T02:17:41Z"},{"alias_kind":"pith_short_8","alias_value":"BUZEIDMR","created_at":"2026-06-30T02:17:41Z"}],"graph_snapshots":[{"event_id":"sha256:898c79b6727d6676fee86b938633157e37af7c2ec6bcac80eb682f81ae2d07e7","target":"graph","created_at":"2026-06-30T02:17:41Z","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/2606.29938/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning with verifiable rewards (RLVR) is bottlenecked by hard prompts on which correct trajectories have low probability, so sampling misses them within a practical budget and leaves the policy update with little useful signal. We frame such zero-hit prompts as RLVR's sampling frontier, where new reasoning behavior is most valuable yet least likely to be sampled. Importantly, failed rollouts can be informative: they expose where the model's reasoning went wrong. We introduce LatentRevise, a first-order latent revision method that recovers training signal for this zero-hit regim","authors_text":"Guangtao Zhai, Jing Bai, Qi Jia, Xueting Han, Yiqiu Guo","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-29T08:14:47Z","title":"LatentRevise: Learning from Zero-Hit Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.29938","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:d879d8e0658381cd2823e73fcd2f6d46909f9a5f39eb187c233671488842a8d1","target":"record","created_at":"2026-06-30T02:17:41Z","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":"509b2d29c1592e21626ffb44c0a81d378b29be532c8f30aca3f12be4938d76e4","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-29T08:14:47Z","title_canon_sha256":"465486504dcfb09fcb7395694373bd10ef0425ed8c425db23832d31834df5fe9"},"schema_version":"1.0","source":{"id":"2606.29938","kind":"arxiv","version":1}},"canonical_sha256":"0d32440d91d7189b5bbbcd73617760e84d841610b61a83ef1f664c0f5815a6fd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0d32440d91d7189b5bbbcd73617760e84d841610b61a83ef1f664c0f5815a6fd","first_computed_at":"2026-06-30T02:17:41.762085Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-30T02:17:41.762085Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dZYAFh8lyPbuOZv4YSlBBJo5utkxnou4y4wlPQWNKY4/eIt+/7fdGZ95pYvk4hQSezQXlNBF9sbnARJXbnGJDQ==","signature_status":"signed_v1","signed_at":"2026-06-30T02:17:41.762691Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.29938","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d879d8e0658381cd2823e73fcd2f6d46909f9a5f39eb187c233671488842a8d1","sha256:898c79b6727d6676fee86b938633157e37af7c2ec6bcac80eb682f81ae2d07e7"],"state_sha256":"0b702c5a7499812378c40714c163d23e1f8eb1783c5ef778f25e610958cd1944"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nspD+/30E/B56aG94206ThHnxPg+px1lA9HBVCeTUxindVNPuyiCnAK6IpdjCG05djCBdyMhbu8Y9GWhs9bbCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-03T18:14:16.765176Z","bundle_sha256":"5ba56a22afb5ccf75c0a344c3fa104619834a1b1596d7e111c9a3c35faa91ec5"}}