{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:D6ND3DMTXEAAITDONZMOY46VTD","short_pith_number":"pith:D6ND3DMT","canonical_record":{"source":{"id":"2403.12017","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-18T17:52:57Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"9d42715de4c1f50eca414aa0a43f6bb218e1c1e8f276be7a754349ca0cf9b61c","abstract_canon_sha256":"792b9f23718a7254e38d0d4014e69f8a2f1a903e1e5e94e310ce18c173c11602"},"schema_version":"1.0"},"canonical_sha256":"1f9a3d8d93b900044c6e6e58ec73d598f82a9048f4eacbc820ae0fc1a3ed1f7b","source":{"kind":"arxiv","id":"2403.12017","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.12017","created_at":"2026-07-05T07:57:35Z"},{"alias_kind":"arxiv_version","alias_value":"2403.12017v1","created_at":"2026-07-05T07:57:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.12017","created_at":"2026-07-05T07:57:35Z"},{"alias_kind":"pith_short_12","alias_value":"D6ND3DMTXEAA","created_at":"2026-07-05T07:57:35Z"},{"alias_kind":"pith_short_16","alias_value":"D6ND3DMTXEAAITDO","created_at":"2026-07-05T07:57:35Z"},{"alias_kind":"pith_short_8","alias_value":"D6ND3DMT","created_at":"2026-07-05T07:57:35Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:D6ND3DMTXEAAITDONZMOY46VTD","target":"record","payload":{"canonical_record":{"source":{"id":"2403.12017","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-18T17:52:57Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"9d42715de4c1f50eca414aa0a43f6bb218e1c1e8f276be7a754349ca0cf9b61c","abstract_canon_sha256":"792b9f23718a7254e38d0d4014e69f8a2f1a903e1e5e94e310ce18c173c11602"},"schema_version":"1.0"},"canonical_sha256":"1f9a3d8d93b900044c6e6e58ec73d598f82a9048f4eacbc820ae0fc1a3ed1f7b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:57:35.216591Z","signature_b64":"cu3FVFxATH6s3qnnIExBSf8G0AoK0KzaZdvGv01mnf/X9QCdpxcbW4645v62N/XrByvl2MI2EpWqqkhUJE0aDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1f9a3d8d93b900044c6e6e58ec73d598f82a9048f4eacbc820ae0fc1a3ed1f7b","last_reissued_at":"2026-07-05T07:57:35.216081Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:57:35.216081Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.12017","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":"2026-07-05T07:57:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jhpZ/DHCEndjwQKb7Yx0CSzRFm6gg/l0QSIhdAGDALt6riGmwWzbCGvlHDxOlpcRbiRMJ89eFoQSIEmNcuXkDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T23:19:24.595701Z"},"content_sha256":"c5805b1f135357def9c47cbb991a52bb15a519f107712ec9498de916631f4e62","schema_version":"1.0","event_id":"sha256:c5805b1f135357def9c47cbb991a52bb15a519f107712ec9498de916631f4e62"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:D6ND3DMTXEAAITDONZMOY46VTD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Supervised Fine-Tuning as Inverse Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Hao Sun","submitted_at":"2024-03-18T17:52:57Z","abstract_excerpt":"The prevailing approach to aligning Large Language Models (LLMs) typically relies on human or AI feedback and assumes access to specific types of preference datasets. In our work, we question the efficacy of such datasets and explore various scenarios where alignment with expert demonstrations proves more realistic. We build a sequential decision-making framework to formulate the problem of aligning LLMs using demonstration datasets. Drawing insights from inverse reinforcement learning and imitation learning, we introduce various approaches for divergence minimization in the LLM alignment task"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.12017","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/2403.12017/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":"2026-07-05T07:57:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oXBRYNnw8dCSA2jhZuBv1j6cwb12N2EYKZNA1rNrOgPFoKZNmlv2R/Lq1jAybxzd7MCM0CTTenWWd5ecX8YmCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T23:19:24.596646Z"},"content_sha256":"9d3aba8ed062699c8c191af28c1fe0708f0ad1619ad0d892b5cd42878aab5232","schema_version":"1.0","event_id":"sha256:9d3aba8ed062699c8c191af28c1fe0708f0ad1619ad0d892b5cd42878aab5232"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/D6ND3DMTXEAAITDONZMOY46VTD/bundle.json","state_url":"https://pith.science/pith/D6ND3DMTXEAAITDONZMOY46VTD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/D6ND3DMTXEAAITDONZMOY46VTD/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":"2026-08-09T23:19:24Z","links":{"resolver":"https://pith.science/pith/D6ND3DMTXEAAITDONZMOY46VTD","bundle":"https://pith.science/pith/D6ND3DMTXEAAITDONZMOY46VTD/bundle.json","state":"https://pith.science/pith/D6ND3DMTXEAAITDONZMOY46VTD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/D6ND3DMTXEAAITDONZMOY46VTD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:D6ND3DMTXEAAITDONZMOY46VTD","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":"792b9f23718a7254e38d0d4014e69f8a2f1a903e1e5e94e310ce18c173c11602","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-18T17:52:57Z","title_canon_sha256":"9d42715de4c1f50eca414aa0a43f6bb218e1c1e8f276be7a754349ca0cf9b61c"},"schema_version":"1.0","source":{"id":"2403.12017","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.12017","created_at":"2026-07-05T07:57:35Z"},{"alias_kind":"arxiv_version","alias_value":"2403.12017v1","created_at":"2026-07-05T07:57:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.12017","created_at":"2026-07-05T07:57:35Z"},{"alias_kind":"pith_short_12","alias_value":"D6ND3DMTXEAA","created_at":"2026-07-05T07:57:35Z"},{"alias_kind":"pith_short_16","alias_value":"D6ND3DMTXEAAITDO","created_at":"2026-07-05T07:57:35Z"},{"alias_kind":"pith_short_8","alias_value":"D6ND3DMT","created_at":"2026-07-05T07:57:35Z"}],"graph_snapshots":[{"event_id":"sha256:9d3aba8ed062699c8c191af28c1fe0708f0ad1619ad0d892b5cd42878aab5232","target":"graph","created_at":"2026-07-05T07:57:35Z","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/2403.12017/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The prevailing approach to aligning Large Language Models (LLMs) typically relies on human or AI feedback and assumes access to specific types of preference datasets. In our work, we question the efficacy of such datasets and explore various scenarios where alignment with expert demonstrations proves more realistic. We build a sequential decision-making framework to formulate the problem of aligning LLMs using demonstration datasets. Drawing insights from inverse reinforcement learning and imitation learning, we introduce various approaches for divergence minimization in the LLM alignment task","authors_text":"Hao Sun","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-18T17:52:57Z","title":"Supervised Fine-Tuning as Inverse Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.12017","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:c5805b1f135357def9c47cbb991a52bb15a519f107712ec9498de916631f4e62","target":"record","created_at":"2026-07-05T07:57:35Z","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":"792b9f23718a7254e38d0d4014e69f8a2f1a903e1e5e94e310ce18c173c11602","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-18T17:52:57Z","title_canon_sha256":"9d42715de4c1f50eca414aa0a43f6bb218e1c1e8f276be7a754349ca0cf9b61c"},"schema_version":"1.0","source":{"id":"2403.12017","kind":"arxiv","version":1}},"canonical_sha256":"1f9a3d8d93b900044c6e6e58ec73d598f82a9048f4eacbc820ae0fc1a3ed1f7b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1f9a3d8d93b900044c6e6e58ec73d598f82a9048f4eacbc820ae0fc1a3ed1f7b","first_computed_at":"2026-07-05T07:57:35.216081Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:57:35.216081Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cu3FVFxATH6s3qnnIExBSf8G0AoK0KzaZdvGv01mnf/X9QCdpxcbW4645v62N/XrByvl2MI2EpWqqkhUJE0aDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:57:35.216591Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.12017","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c5805b1f135357def9c47cbb991a52bb15a519f107712ec9498de916631f4e62","sha256:9d3aba8ed062699c8c191af28c1fe0708f0ad1619ad0d892b5cd42878aab5232"],"state_sha256":"7b673db184e12b867e2b50a348ebd2d80af2750c415d16091ddc4b9e663cc2aa"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1Btp750vLi3EFMJVna+dPZLgD74vFDYAY1IkVtwaDs4rlw32ru/u8L5RAl4ZEYsFWzkm/IJ/4REe73uoyICXCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T23:19:24.602853Z","bundle_sha256":"af8d20c8db6d3851b74282d25b4baa5795717d6bf641f93126b6487e7284a902"}}