{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:J7F5OYQ5MHRNOYL3PJQ762N64A","short_pith_number":"pith:J7F5OYQ5","canonical_record":{"source":{"id":"2602.11675","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-02-12T07:48:21Z","cross_cats_sorted":[],"title_canon_sha256":"dfc6af3c865c59bb01d646c95472519247da114c9af32f42b085052598999647","abstract_canon_sha256":"661e71f4dec067c55314260d28df484caf64fb8bcb838baa0a4d0a30ba4890d9"},"schema_version":"1.0"},"canonical_sha256":"4fcbd7621d61e2d7617b7a61ff69bee007aae4e7dab92eec6ed8ab2c6bf44248","source":{"kind":"arxiv","id":"2602.11675","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.11675","created_at":"2026-05-21T01:04:23Z"},{"alias_kind":"arxiv_version","alias_value":"2602.11675v4","created_at":"2026-05-21T01:04:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.11675","created_at":"2026-05-21T01:04:23Z"},{"alias_kind":"pith_short_12","alias_value":"J7F5OYQ5MHRN","created_at":"2026-05-21T01:04:23Z"},{"alias_kind":"pith_short_16","alias_value":"J7F5OYQ5MHRNOYL3","created_at":"2026-05-21T01:04:23Z"},{"alias_kind":"pith_short_8","alias_value":"J7F5OYQ5","created_at":"2026-05-21T01:04:23Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:J7F5OYQ5MHRNOYL3PJQ762N64A","target":"record","payload":{"canonical_record":{"source":{"id":"2602.11675","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-02-12T07:48:21Z","cross_cats_sorted":[],"title_canon_sha256":"dfc6af3c865c59bb01d646c95472519247da114c9af32f42b085052598999647","abstract_canon_sha256":"661e71f4dec067c55314260d28df484caf64fb8bcb838baa0a4d0a30ba4890d9"},"schema_version":"1.0"},"canonical_sha256":"4fcbd7621d61e2d7617b7a61ff69bee007aae4e7dab92eec6ed8ab2c6bf44248","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:04:23.152743Z","signature_b64":"PiPvBFQ/trn6i7mDjDCQAD4c91zpIuxm03YEQQYh3woXy8hi3He+10IvlHoSfqQ2zazq24BG9StgcYULujYgBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4fcbd7621d61e2d7617b7a61ff69bee007aae4e7dab92eec6ed8ab2c6bf44248","last_reissued_at":"2026-05-21T01:04:23.151766Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:04:23.151766Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2602.11675","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-21T01:04:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DN9VoMInuVj2kGIJJzKneVEeO3Ric3BjXdd6IoHagpSLRgWtoY0JkAGoPQiIi/WSdNnZfV5f9b2p0y/QG3jUAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T09:36:10.492943Z"},"content_sha256":"52e386b861293b61579e87a0d69a3a742f8b2418fd3b67d9716ac866258f3400","schema_version":"1.0","event_id":"sha256:52e386b861293b61579e87a0d69a3a742f8b2418fd3b67d9716ac866258f3400"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:J7F5OYQ5MHRNOYL3PJQ762N64A","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Epistemic Regret Minimization: Label-Free Causal Critique Beyond Outcome Reward","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Epistemic Regret Minimization identifies causal flaws in LLM reasoning traces without ground-truth labels","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Edward Y. Chang, Longling Geng","submitted_at":"2026-02-12T07:48:21Z","abstract_excerpt":"Large language models can answer causal questions correctly for the wrong reasons. Current RL methods reward \\emph{what} a model concludes but ignore \\emph{why}, reinforcing correlational shortcuts -- a failure we call \\emph{Reward Entrenchment}. We introduce \\emph{Epistemic Regret Minimization} (\\erm), a framework that critiques the causal \\emph{structure} of a model's reasoning trace rather than its answer. Applying established causal principles, \\erm flags unexamined confounders, correlation--intervention conflation, and unchecked back-door paths from exposed reasoning traces. The framework"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"A separation theorem proves outcome-only RL cannot distinguish correct from flawed causal models in confounded environments, and preliminary experiments show epistemic reward carries signal where outcome reward does not.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That causal flaws are reliably identifiable and correctable from reasoning traces alone without ground-truth labels or external verifiers, and that the observed corrections on CausalT5K and CLadder generalize beyond the tested models and scenarios.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Epistemic Regret Minimization identifies causal reasoning flaws in LLMs from traces alone, corrects stubborn models where outcome-only methods fail, and is supported by a separation theorem proving outcome-only RL cannot distinguish correct from flawed causal models in confounded settings.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Epistemic Regret Minimization identifies causal flaws in LLM reasoning traces without ground-truth labels","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"3b31d6e4d0acc81bc9f0dd8e9cf8877ca93e9d0881f07e5a7cd81c242658ada2"},"source":{"id":"2602.11675","kind":"arxiv","version":4},"verdict":{"id":"c94b2c6a-f016-4dca-a646-1cde4ebdba1b","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T05:15:57.563360Z","strongest_claim":"A separation theorem proves outcome-only RL cannot distinguish correct from flawed causal models in confounded environments, and preliminary experiments show epistemic reward carries signal where outcome reward does not.","one_line_summary":"Epistemic Regret Minimization identifies causal reasoning flaws in LLMs from traces alone, corrects stubborn models where outcome-only methods fail, and is supported by a separation theorem proving outcome-only RL cannot distinguish correct from flawed causal models in confounded settings.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That causal flaws are reliably identifiable and correctable from reasoning traces alone without ground-truth labels or external verifiers, and that the observed corrections on CausalT5K and CLadder generalize beyond the tested models and scenarios.","pith_extraction_headline":"Epistemic Regret Minimization identifies causal flaws in LLM reasoning traces without ground-truth labels"},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2602.11675/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":3,"snapshot_sha256":"60cc0fac08e1c6b18781b3250b295ff749ea533a854ae105f5fb76b8d331e0f0"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"c94b2c6a-f016-4dca-a646-1cde4ebdba1b"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-21T01:04:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JRu4/clU+dNrC80wWI6GQvX9UCKh11o0B9GkxMzlxnMibcQ4JtCpM7aoy0HLwDp9bGYyRO6ecOkLVhGOrm6YBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T09:36:10.493403Z"},"content_sha256":"67c49011d6882c798cee31849160557b864f0c6264c429b34eed6885ddf79a46","schema_version":"1.0","event_id":"sha256:67c49011d6882c798cee31849160557b864f0c6264c429b34eed6885ddf79a46"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/J7F5OYQ5MHRNOYL3PJQ762N64A/bundle.json","state_url":"https://pith.science/pith/J7F5OYQ5MHRNOYL3PJQ762N64A/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/J7F5OYQ5MHRNOYL3PJQ762N64A/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":"2026-06-30T09:36:10Z","links":{"resolver":"https://pith.science/pith/J7F5OYQ5MHRNOYL3PJQ762N64A","bundle":"https://pith.science/pith/J7F5OYQ5MHRNOYL3PJQ762N64A/bundle.json","state":"https://pith.science/pith/J7F5OYQ5MHRNOYL3PJQ762N64A/state.json","well_known_bundle":"https://pith.science/.well-known/pith/J7F5OYQ5MHRNOYL3PJQ762N64A/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:J7F5OYQ5MHRNOYL3PJQ762N64A","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":"661e71f4dec067c55314260d28df484caf64fb8bcb838baa0a4d0a30ba4890d9","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-02-12T07:48:21Z","title_canon_sha256":"dfc6af3c865c59bb01d646c95472519247da114c9af32f42b085052598999647"},"schema_version":"1.0","source":{"id":"2602.11675","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.11675","created_at":"2026-05-21T01:04:23Z"},{"alias_kind":"arxiv_version","alias_value":"2602.11675v4","created_at":"2026-05-21T01:04:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.11675","created_at":"2026-05-21T01:04:23Z"},{"alias_kind":"pith_short_12","alias_value":"J7F5OYQ5MHRN","created_at":"2026-05-21T01:04:23Z"},{"alias_kind":"pith_short_16","alias_value":"J7F5OYQ5MHRNOYL3","created_at":"2026-05-21T01:04:23Z"},{"alias_kind":"pith_short_8","alias_value":"J7F5OYQ5","created_at":"2026-05-21T01:04:23Z"}],"graph_snapshots":[{"event_id":"sha256:67c49011d6882c798cee31849160557b864f0c6264c429b34eed6885ddf79a46","target":"graph","created_at":"2026-05-21T01:04:23Z","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":"A separation theorem proves outcome-only RL cannot distinguish correct from flawed causal models in confounded environments, and preliminary experiments show epistemic reward carries signal where outcome reward does not."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That causal flaws are reliably identifiable and correctable from reasoning traces alone without ground-truth labels or external verifiers, and that the observed corrections on CausalT5K and CLadder generalize beyond the tested models and scenarios."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Epistemic Regret Minimization identifies causal reasoning flaws in LLMs from traces alone, corrects stubborn models where outcome-only methods fail, and is supported by a separation theorem proving outcome-only RL cannot distinguish correct from flawed causal models in confounded settings."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Epistemic Regret Minimization identifies causal flaws in LLM reasoning traces without ground-truth labels"}],"snapshot_sha256":"3b31d6e4d0acc81bc9f0dd8e9cf8877ca93e9d0881f07e5a7cd81c242658ada2"},"formal_canon":{"evidence_count":3,"snapshot_sha256":"60cc0fac08e1c6b18781b3250b295ff749ea533a854ae105f5fb76b8d331e0f0"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2602.11675/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models can answer causal questions correctly for the wrong reasons. Current RL methods reward \\emph{what} a model concludes but ignore \\emph{why}, reinforcing correlational shortcuts -- a failure we call \\emph{Reward Entrenchment}. We introduce \\emph{Epistemic Regret Minimization} (\\erm), a framework that critiques the causal \\emph{structure} of a model's reasoning trace rather than its answer. Applying established causal principles, \\erm flags unexamined confounders, correlation--intervention conflation, and unchecked back-door paths from exposed reasoning traces. The framework","authors_text":"Edward Y. Chang, Longling Geng","cross_cats":[],"headline":"Epistemic Regret Minimization identifies causal flaws in LLM reasoning traces without ground-truth labels","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-02-12T07:48:21Z","title":"Epistemic Regret Minimization: Label-Free Causal Critique Beyond Outcome Reward"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.11675","kind":"arxiv","version":4},"verdict":{"created_at":"2026-05-16T05:15:57.563360Z","id":"c94b2c6a-f016-4dca-a646-1cde4ebdba1b","model_set":{"reader":"grok-4.3"},"one_line_summary":"Epistemic Regret Minimization identifies causal reasoning flaws in LLMs from traces alone, corrects stubborn models where outcome-only methods fail, and is supported by a separation theorem proving outcome-only RL cannot distinguish correct from flawed causal models in confounded settings.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Epistemic Regret Minimization identifies causal flaws in LLM reasoning traces without ground-truth labels","strongest_claim":"A separation theorem proves outcome-only RL cannot distinguish correct from flawed causal models in confounded environments, and preliminary experiments show epistemic reward carries signal where outcome reward does not.","weakest_assumption":"That causal flaws are reliably identifiable and correctable from reasoning traces alone without ground-truth labels or external verifiers, and that the observed corrections on CausalT5K and CLadder generalize beyond the tested models and scenarios."}},"verdict_id":"c94b2c6a-f016-4dca-a646-1cde4ebdba1b"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:52e386b861293b61579e87a0d69a3a742f8b2418fd3b67d9716ac866258f3400","target":"record","created_at":"2026-05-21T01:04:23Z","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":"661e71f4dec067c55314260d28df484caf64fb8bcb838baa0a4d0a30ba4890d9","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-02-12T07:48:21Z","title_canon_sha256":"dfc6af3c865c59bb01d646c95472519247da114c9af32f42b085052598999647"},"schema_version":"1.0","source":{"id":"2602.11675","kind":"arxiv","version":4}},"canonical_sha256":"4fcbd7621d61e2d7617b7a61ff69bee007aae4e7dab92eec6ed8ab2c6bf44248","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4fcbd7621d61e2d7617b7a61ff69bee007aae4e7dab92eec6ed8ab2c6bf44248","first_computed_at":"2026-05-21T01:04:23.151766Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-21T01:04:23.151766Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PiPvBFQ/trn6i7mDjDCQAD4c91zpIuxm03YEQQYh3woXy8hi3He+10IvlHoSfqQ2zazq24BG9StgcYULujYgBg==","signature_status":"signed_v1","signed_at":"2026-05-21T01:04:23.152743Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.11675","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:52e386b861293b61579e87a0d69a3a742f8b2418fd3b67d9716ac866258f3400","sha256:67c49011d6882c798cee31849160557b864f0c6264c429b34eed6885ddf79a46"],"state_sha256":"ab9b58448f07b33db53861b464143de310a5d9315a32c77d3b9e569023346106"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2zsxjjp+IOxwnTRPuOUuoPjti79Zf+n8RiP+CHgDkgpkgZL7RiLG3TcUVwDQj/lW6KCfq/ZBe9CCvaFquldpAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-30T09:36:10.495521Z","bundle_sha256":"999366ead42a727d14d78d8890cddcf0a284c4a9ed31e7783d0f5778a2f9b898"}}