{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:CQGIDWMLLCYQAN5KWACLG5YFZL","short_pith_number":"pith:CQGIDWML","canonical_record":{"source":{"id":"2605.14632","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T09:44:11Z","cross_cats_sorted":["stat.AP"],"title_canon_sha256":"063a9cba2f0771048e5352c4994b9d415a5e094ac5b56ca8108d0fb26480ca65","abstract_canon_sha256":"2c20719c1fca4d24a33a9e717ed041ab2b67b702c0a848b4b1c4b0a93d0f5351"},"schema_version":"1.0"},"canonical_sha256":"140c81d98b58b10037aab004b37705cae363a84c806c6f68c141ae957fbbb686","source":{"kind":"arxiv","id":"2605.14632","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14632","created_at":"2026-05-17T23:39:03Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14632v1","created_at":"2026-05-17T23:39:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14632","created_at":"2026-05-17T23:39:03Z"},{"alias_kind":"pith_short_12","alias_value":"CQGIDWMLLCYQ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"CQGIDWMLLCYQAN5K","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"CQGIDWML","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:CQGIDWMLLCYQAN5KWACLG5YFZL","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14632","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T09:44:11Z","cross_cats_sorted":["stat.AP"],"title_canon_sha256":"063a9cba2f0771048e5352c4994b9d415a5e094ac5b56ca8108d0fb26480ca65","abstract_canon_sha256":"2c20719c1fca4d24a33a9e717ed041ab2b67b702c0a848b4b1c4b0a93d0f5351"},"schema_version":"1.0"},"canonical_sha256":"140c81d98b58b10037aab004b37705cae363a84c806c6f68c141ae957fbbb686","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:03.961431Z","signature_b64":"2l+Io/yGkF28Vixf6yFoHFR33e6zcr9e1AHhoglOjbdSXYkR47bZkyzREPpzPOpI39c3Kw4CRcD1tDG4ytc+Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"140c81d98b58b10037aab004b37705cae363a84c806c6f68c141ae957fbbb686","last_reissued_at":"2026-05-17T23:39:03.960812Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:03.960812Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14632","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":"2026-05-17T23:39:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OvPsz8f6ZXETLay5IK9tsN3lAP7GabVnoRpjYFdP/sbgfISMYnDWQGyiVmayctONFDMbharsh4/lbLw+r5NfBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T19:27:32.806442Z"},"content_sha256":"041068ae902e2a591975fecb5c2e40f6e4d0473d96d606a3fea6335441fe36ec","schema_version":"1.0","event_id":"sha256:041068ae902e2a591975fecb5c2e40f6e4d0473d96d606a3fea6335441fe36ec"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:CQGIDWMLLCYQAN5KWACLG5YFZL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DRL-STAF: A Deep Reinforcement Learning Framework for State-Aware Forecasting of Complex Multivariate Hidden Markov Processes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"DRL-STAF jointly forecasts observations and estimates discrete hidden states in complex multivariate hidden Markov processes by combining deep neural networks with reinforcement learning.","cross_cats":["stat.AP"],"primary_cat":"cs.LG","authors_text":"Chen Zhang, Jingru Huang, Manrui Jiang, Yong Chen","submitted_at":"2026-05-14T09:44:11Z","abstract_excerpt":"Forecasting multivariate hidden Markov processes is challenging due to nonlinear and nonstationary observations, latent state transitions, and cross-sequence dependencies. While deep learning methods achieve strong predictive accuracy, they typically lack explicit state modeling, whereas Hidden Markov Models (HMMs) provide interpretable latent states but struggle with complex nonlinear emissions and scalability. To address these limitations, we propose DRL-STAF, a Deep Reinforcement Learning based STate-Aware Forecasting framework that jointly predicts next-step observations and estimates the "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"DRL-STAF outperforms HMM variants, standalone deep learning models, and existing DL-HMM hybrids in most cases, while also providing reliable hidden-state estimates.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"Reinforcement learning can effectively estimate discrete hidden states and learn flexible transition dynamics from data without relying on predefined structures, assuming the RL formulation captures the underlying process accurately.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"DRL-STAF uses deep RL to predict observations and estimate discrete hidden states for multivariate hidden Markov processes, outperforming HMMs, deep learning models, and hybrids in experiments.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"DRL-STAF jointly forecasts observations and estimates discrete hidden states in complex multivariate hidden Markov processes by combining deep neural networks with reinforcement learning.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"aa9fce3a70cdf4d4da870233bdba12af6953acd0764d472920cd6659a7cb0578"},"source":{"id":"2605.14632","kind":"arxiv","version":1},"verdict":{"id":"ca3efe8e-d9bd-4bc6-aa75-b0e087dde0de","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T04:36:18.431526Z","strongest_claim":"DRL-STAF outperforms HMM variants, standalone deep learning models, and existing DL-HMM hybrids in most cases, while also providing reliable hidden-state estimates.","one_line_summary":"DRL-STAF uses deep RL to predict observations and estimate discrete hidden states for multivariate hidden Markov processes, outperforming HMMs, deep learning models, and hybrids in experiments.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"Reinforcement learning can effectively estimate discrete hidden states and learn flexible transition dynamics from data without relying on predefined structures, assuming the RL formulation captures the underlying process accurately.","pith_extraction_headline":"DRL-STAF jointly forecasts observations and estimates discrete hidden states in complex multivariate hidden Markov processes by combining deep neural networks with reinforcement learning."},"references":{"count":64,"sample":[{"doi":"","year":null,"title":"ACM SIGIR , pages =","work_id":"374aaee2-8c83-457e-bee9-b7f01e4cec34","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"NeurIPS , pages =","work_id":"0e5a43b0-5d28-4d7e-a302-4e9a93ad7fa2","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"ICLR , year =","work_id":"88639ea1-4dc1-4aa7-a746-b2a2f9f2d51c","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"ICLR , year =","work_id":"cda52447-4411-47ff-8deb-79caa59bada2","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2012,"title":"IEEE Transactions on Audio, Speech, and Language Processing , author =","work_id":"1bc49324-c086-4ea8-b88a-5167fce36dce","ref_index":6,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":64,"snapshot_sha256":"232861c1d95da2c754019a554ae17d0c664d5b9c280b721c00cca880e0435282","internal_anchors":1},"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":"ca3efe8e-d9bd-4bc6-aa75-b0e087dde0de"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"K4wW4KZ6XgTQ7JfoPQAZmw5CJ9kzDmh3IzctniLpnVGLVNLO0VPp/1Iah+WPf979HYH7DuMQ1L1ORhKLYyLeDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T19:27:32.807521Z"},"content_sha256":"6d5fcd9e50f4b9b76158690a53c38db18650f91c2f90df1c30086d3a9c622c38","schema_version":"1.0","event_id":"sha256:6d5fcd9e50f4b9b76158690a53c38db18650f91c2f90df1c30086d3a9c622c38"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CQGIDWMLLCYQAN5KWACLG5YFZL/bundle.json","state_url":"https://pith.science/pith/CQGIDWMLLCYQAN5KWACLG5YFZL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CQGIDWMLLCYQAN5KWACLG5YFZL/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":"2026-05-23T19:27:32Z","links":{"resolver":"https://pith.science/pith/CQGIDWMLLCYQAN5KWACLG5YFZL","bundle":"https://pith.science/pith/CQGIDWMLLCYQAN5KWACLG5YFZL/bundle.json","state":"https://pith.science/pith/CQGIDWMLLCYQAN5KWACLG5YFZL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CQGIDWMLLCYQAN5KWACLG5YFZL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:CQGIDWMLLCYQAN5KWACLG5YFZL","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":"2c20719c1fca4d24a33a9e717ed041ab2b67b702c0a848b4b1c4b0a93d0f5351","cross_cats_sorted":["stat.AP"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T09:44:11Z","title_canon_sha256":"063a9cba2f0771048e5352c4994b9d415a5e094ac5b56ca8108d0fb26480ca65"},"schema_version":"1.0","source":{"id":"2605.14632","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14632","created_at":"2026-05-17T23:39:03Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14632v1","created_at":"2026-05-17T23:39:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14632","created_at":"2026-05-17T23:39:03Z"},{"alias_kind":"pith_short_12","alias_value":"CQGIDWMLLCYQ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"CQGIDWMLLCYQAN5K","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"CQGIDWML","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:6d5fcd9e50f4b9b76158690a53c38db18650f91c2f90df1c30086d3a9c622c38","target":"graph","created_at":"2026-05-17T23:39:03Z","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":"DRL-STAF outperforms HMM variants, standalone deep learning models, and existing DL-HMM hybrids in most cases, while also providing reliable hidden-state estimates."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"Reinforcement learning can effectively estimate discrete hidden states and learn flexible transition dynamics from data without relying on predefined structures, assuming the RL formulation captures the underlying process accurately."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"DRL-STAF uses deep RL to predict observations and estimate discrete hidden states for multivariate hidden Markov processes, outperforming HMMs, deep learning models, and hybrids in experiments."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"DRL-STAF jointly forecasts observations and estimates discrete hidden states in complex multivariate hidden Markov processes by combining deep neural networks with reinforcement learning."}],"snapshot_sha256":"aa9fce3a70cdf4d4da870233bdba12af6953acd0764d472920cd6659a7cb0578"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Forecasting multivariate hidden Markov processes is challenging due to nonlinear and nonstationary observations, latent state transitions, and cross-sequence dependencies. While deep learning methods achieve strong predictive accuracy, they typically lack explicit state modeling, whereas Hidden Markov Models (HMMs) provide interpretable latent states but struggle with complex nonlinear emissions and scalability. To address these limitations, we propose DRL-STAF, a Deep Reinforcement Learning based STate-Aware Forecasting framework that jointly predicts next-step observations and estimates the ","authors_text":"Chen Zhang, Jingru Huang, Manrui Jiang, Yong Chen","cross_cats":["stat.AP"],"headline":"DRL-STAF jointly forecasts observations and estimates discrete hidden states in complex multivariate hidden Markov processes by combining deep neural networks with reinforcement learning.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T09:44:11Z","title":"DRL-STAF: A Deep Reinforcement Learning Framework for State-Aware Forecasting of Complex Multivariate Hidden Markov Processes"},"references":{"count":64,"internal_anchors":1,"resolved_work":64,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"ACM SIGIR , pages =","work_id":"374aaee2-8c83-457e-bee9-b7f01e4cec34","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"NeurIPS , pages =","work_id":"0e5a43b0-5d28-4d7e-a302-4e9a93ad7fa2","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"ICLR , year =","work_id":"88639ea1-4dc1-4aa7-a746-b2a2f9f2d51c","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"ICLR , year =","work_id":"cda52447-4411-47ff-8deb-79caa59bada2","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":6,"title":"IEEE Transactions on Audio, Speech, and Language Processing , author =","work_id":"1bc49324-c086-4ea8-b88a-5167fce36dce","year":2012}],"snapshot_sha256":"232861c1d95da2c754019a554ae17d0c664d5b9c280b721c00cca880e0435282"},"source":{"id":"2605.14632","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T04:36:18.431526Z","id":"ca3efe8e-d9bd-4bc6-aa75-b0e087dde0de","model_set":{"reader":"grok-4.3"},"one_line_summary":"DRL-STAF uses deep RL to predict observations and estimate discrete hidden states for multivariate hidden Markov processes, outperforming HMMs, deep learning models, and hybrids in experiments.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"DRL-STAF jointly forecasts observations and estimates discrete hidden states in complex multivariate hidden Markov processes by combining deep neural networks with reinforcement learning.","strongest_claim":"DRL-STAF outperforms HMM variants, standalone deep learning models, and existing DL-HMM hybrids in most cases, while also providing reliable hidden-state estimates.","weakest_assumption":"Reinforcement learning can effectively estimate discrete hidden states and learn flexible transition dynamics from data without relying on predefined structures, assuming the RL formulation captures the underlying process accurately."}},"verdict_id":"ca3efe8e-d9bd-4bc6-aa75-b0e087dde0de"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:041068ae902e2a591975fecb5c2e40f6e4d0473d96d606a3fea6335441fe36ec","target":"record","created_at":"2026-05-17T23:39:03Z","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":"2c20719c1fca4d24a33a9e717ed041ab2b67b702c0a848b4b1c4b0a93d0f5351","cross_cats_sorted":["stat.AP"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T09:44:11Z","title_canon_sha256":"063a9cba2f0771048e5352c4994b9d415a5e094ac5b56ca8108d0fb26480ca65"},"schema_version":"1.0","source":{"id":"2605.14632","kind":"arxiv","version":1}},"canonical_sha256":"140c81d98b58b10037aab004b37705cae363a84c806c6f68c141ae957fbbb686","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"140c81d98b58b10037aab004b37705cae363a84c806c6f68c141ae957fbbb686","first_computed_at":"2026-05-17T23:39:03.960812Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:03.960812Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2l+Io/yGkF28Vixf6yFoHFR33e6zcr9e1AHhoglOjbdSXYkR47bZkyzREPpzPOpI39c3Kw4CRcD1tDG4ytc+Dg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:03.961431Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14632","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:041068ae902e2a591975fecb5c2e40f6e4d0473d96d606a3fea6335441fe36ec","sha256:6d5fcd9e50f4b9b76158690a53c38db18650f91c2f90df1c30086d3a9c622c38"],"state_sha256":"b35bca06847aa049c3441c41c0ee82d760ac0d95b291e30bfeeb41061b0efbc0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UVfVVur1nL7F3hFUdoIbCZKuKEMRTVs1KrYDqaUvo+7H7bYlcMcF/TgQS8KJMRFploDo+IpkJfwU1aEqyeoRDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-23T19:27:32.812122Z","bundle_sha256":"d2ef9b01914894279099c5a5fb02426b283d0f30b4d1b4e9ac9c3eb4e6dd5355"}}