{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:WLW2XT2UJNY7WMIKOBWMAK3A4K","short_pith_number":"pith:WLW2XT2U","canonical_record":{"source":{"id":"2605.24547","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-23T12:28:14Z","cross_cats_sorted":[],"title_canon_sha256":"2b69cef081f93566af8f67d6d9208a8a6a9f6c9eb8c909dec677fc9a99290e8e","abstract_canon_sha256":"989ea236cc259111ce66523cdfcc05c24975c81677e8a42c031255a12f557f73"},"schema_version":"1.0"},"canonical_sha256":"b2edabcf544b71fb310a706cc02b60e281016faad3ea16b9c06bceada364dde2","source":{"kind":"arxiv","id":"2605.24547","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.24547","created_at":"2026-05-26T01:03:45Z"},{"alias_kind":"arxiv_version","alias_value":"2605.24547v1","created_at":"2026-05-26T01:03:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24547","created_at":"2026-05-26T01:03:45Z"},{"alias_kind":"pith_short_12","alias_value":"WLW2XT2UJNY7","created_at":"2026-05-26T01:03:45Z"},{"alias_kind":"pith_short_16","alias_value":"WLW2XT2UJNY7WMIK","created_at":"2026-05-26T01:03:45Z"},{"alias_kind":"pith_short_8","alias_value":"WLW2XT2U","created_at":"2026-05-26T01:03:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:WLW2XT2UJNY7WMIKOBWMAK3A4K","target":"record","payload":{"canonical_record":{"source":{"id":"2605.24547","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-23T12:28:14Z","cross_cats_sorted":[],"title_canon_sha256":"2b69cef081f93566af8f67d6d9208a8a6a9f6c9eb8c909dec677fc9a99290e8e","abstract_canon_sha256":"989ea236cc259111ce66523cdfcc05c24975c81677e8a42c031255a12f557f73"},"schema_version":"1.0"},"canonical_sha256":"b2edabcf544b71fb310a706cc02b60e281016faad3ea16b9c06bceada364dde2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:03:45.778327Z","signature_b64":"Bi+aErmJU/mcpLpcp3IUMWJ76sspgZ7snxdxRVKfhbA9oIpxf8sBIae+JSsvmyJornIWNKSa04Ckhzq0uahrBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b2edabcf544b71fb310a706cc02b60e281016faad3ea16b9c06bceada364dde2","last_reissued_at":"2026-05-26T01:03:45.777480Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:03:45.777480Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.24547","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":"2026-05-26T01:03:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PnxgHBlG/BYXBBPTrXJ4bESlihI9+e+5I1wrfKVlmgyVL2L4sTYtF4SQt5PbyBeQt9r5S/Wvk9X/ct265CYhBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T21:26:55.524082Z"},"content_sha256":"2e530ed2a6acb0d9789a2a654067be51aa326b6eeab378bd6644e8a32e799abd","schema_version":"1.0","event_id":"sha256:2e530ed2a6acb0d9789a2a654067be51aa326b6eeab378bd6644e8a32e799abd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:WLW2XT2UJNY7WMIKOBWMAK3A4K","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RL with Learnable Textual Feedback: A Bilevel Approach","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Amrit Singh Bedi, Sidhaarth Sredharan, Souradip Chakraborty, Utsav Singh","submitted_at":"2026-05-23T12:28:14Z","abstract_excerpt":"Reinforcement learning with verifiable rewards can improve LLM reasoning, but learning remains sample-inefficient when terminal rewards are sparse. This has motivated a growing line of work on RL with textual feedback, where a critic model generates natural language feedback to guide a reasoning model (the actor), augmenting scalar rewards with richer learning signals. However, existing methods typically treat feedback as fixed or auxiliary, which misses a key property: feedback should not merely be correct, but should improve the policy (actor model) when provided in context. This motivates a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24547","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/2605.24547/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":"2026-05-26T01:03:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"g8eR/NG2AzjUCgQ/wdsqrDGnuDbBm/p/WLPZggL7JkRUi2gkP3PfD3atKeg4Sbm9k0+libOYJhupsC2Z1esHCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T21:26:55.524467Z"},"content_sha256":"9883a636d07fa477802891f28be1899ac73b7bdc669a7a23737d10edd2ced5b0","schema_version":"1.0","event_id":"sha256:9883a636d07fa477802891f28be1899ac73b7bdc669a7a23737d10edd2ced5b0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WLW2XT2UJNY7WMIKOBWMAK3A4K/bundle.json","state_url":"https://pith.science/pith/WLW2XT2UJNY7WMIKOBWMAK3A4K/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WLW2XT2UJNY7WMIKOBWMAK3A4K/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":"2026-06-27T21:26:55Z","links":{"resolver":"https://pith.science/pith/WLW2XT2UJNY7WMIKOBWMAK3A4K","bundle":"https://pith.science/pith/WLW2XT2UJNY7WMIKOBWMAK3A4K/bundle.json","state":"https://pith.science/pith/WLW2XT2UJNY7WMIKOBWMAK3A4K/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WLW2XT2UJNY7WMIKOBWMAK3A4K/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:WLW2XT2UJNY7WMIKOBWMAK3A4K","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":"989ea236cc259111ce66523cdfcc05c24975c81677e8a42c031255a12f557f73","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-23T12:28:14Z","title_canon_sha256":"2b69cef081f93566af8f67d6d9208a8a6a9f6c9eb8c909dec677fc9a99290e8e"},"schema_version":"1.0","source":{"id":"2605.24547","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.24547","created_at":"2026-05-26T01:03:45Z"},{"alias_kind":"arxiv_version","alias_value":"2605.24547v1","created_at":"2026-05-26T01:03:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24547","created_at":"2026-05-26T01:03:45Z"},{"alias_kind":"pith_short_12","alias_value":"WLW2XT2UJNY7","created_at":"2026-05-26T01:03:45Z"},{"alias_kind":"pith_short_16","alias_value":"WLW2XT2UJNY7WMIK","created_at":"2026-05-26T01:03:45Z"},{"alias_kind":"pith_short_8","alias_value":"WLW2XT2U","created_at":"2026-05-26T01:03:45Z"}],"graph_snapshots":[{"event_id":"sha256:9883a636d07fa477802891f28be1899ac73b7bdc669a7a23737d10edd2ced5b0","target":"graph","created_at":"2026-05-26T01:03:45Z","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/2605.24547/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning with verifiable rewards can improve LLM reasoning, but learning remains sample-inefficient when terminal rewards are sparse. This has motivated a growing line of work on RL with textual feedback, where a critic model generates natural language feedback to guide a reasoning model (the actor), augmenting scalar rewards with richer learning signals. However, existing methods typically treat feedback as fixed or auxiliary, which misses a key property: feedback should not merely be correct, but should improve the policy (actor model) when provided in context. This motivates a","authors_text":"Amrit Singh Bedi, Sidhaarth Sredharan, Souradip Chakraborty, Utsav Singh","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-23T12:28:14Z","title":"RL with Learnable Textual Feedback: A Bilevel Approach"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24547","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":"sha256:2e530ed2a6acb0d9789a2a654067be51aa326b6eeab378bd6644e8a32e799abd","target":"record","created_at":"2026-05-26T01:03:45Z","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":"989ea236cc259111ce66523cdfcc05c24975c81677e8a42c031255a12f557f73","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-23T12:28:14Z","title_canon_sha256":"2b69cef081f93566af8f67d6d9208a8a6a9f6c9eb8c909dec677fc9a99290e8e"},"schema_version":"1.0","source":{"id":"2605.24547","kind":"arxiv","version":1}},"canonical_sha256":"b2edabcf544b71fb310a706cc02b60e281016faad3ea16b9c06bceada364dde2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b2edabcf544b71fb310a706cc02b60e281016faad3ea16b9c06bceada364dde2","first_computed_at":"2026-05-26T01:03:45.777480Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-26T01:03:45.777480Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Bi+aErmJU/mcpLpcp3IUMWJ76sspgZ7snxdxRVKfhbA9oIpxf8sBIae+JSsvmyJornIWNKSa04Ckhzq0uahrBg==","signature_status":"signed_v1","signed_at":"2026-05-26T01:03:45.778327Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.24547","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2e530ed2a6acb0d9789a2a654067be51aa326b6eeab378bd6644e8a32e799abd","sha256:9883a636d07fa477802891f28be1899ac73b7bdc669a7a23737d10edd2ced5b0"],"state_sha256":"f8ae5793fcafe8d71412000e2556569b2311decda7074aa8ac05b0594edf4d75"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KXT+U5JBlnFhvtexIOS4jLpyR0fYjNU0RRPiNzhACMd7Gd3z50WhcocMZ0v3fk0Dm31OHD1Kcw77w/j+0LnqAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-27T21:26:55.526453Z","bundle_sha256":"107cf3341d80d3a8f8ff147390a48a2344df05936d308ebf6c5f8b2cebb9e1ca"}}