{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:PNHIVQOVDNSEPR6BRVX6FLBGK5","short_pith_number":"pith:PNHIVQOV","canonical_record":{"source":{"id":"2605.16837","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cond-mat.mtrl-sci","submitted_at":"2026-05-16T06:40:56Z","cross_cats_sorted":[],"title_canon_sha256":"5f7fece14ea4c39ddd7e84309082739f4da177c4a7ad1e4635b5ac94a6d1550e","abstract_canon_sha256":"ff0613b11f85a4d3ff40035a1fd5ab17566a443b870684e87d35fc89bc0c6712"},"schema_version":"1.0"},"canonical_sha256":"7b4e8ac1d51b6447c7c18d6fe2ac26574435619bfcf8df006265674a0a63fc5c","source":{"kind":"arxiv","id":"2605.16837","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.16837","created_at":"2026-05-20T00:03:25Z"},{"alias_kind":"arxiv_version","alias_value":"2605.16837v1","created_at":"2026-05-20T00:03:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16837","created_at":"2026-05-20T00:03:25Z"},{"alias_kind":"pith_short_12","alias_value":"PNHIVQOVDNSE","created_at":"2026-05-20T00:03:25Z"},{"alias_kind":"pith_short_16","alias_value":"PNHIVQOVDNSEPR6B","created_at":"2026-05-20T00:03:25Z"},{"alias_kind":"pith_short_8","alias_value":"PNHIVQOV","created_at":"2026-05-20T00:03:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:PNHIVQOVDNSEPR6BRVX6FLBGK5","target":"record","payload":{"canonical_record":{"source":{"id":"2605.16837","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cond-mat.mtrl-sci","submitted_at":"2026-05-16T06:40:56Z","cross_cats_sorted":[],"title_canon_sha256":"5f7fece14ea4c39ddd7e84309082739f4da177c4a7ad1e4635b5ac94a6d1550e","abstract_canon_sha256":"ff0613b11f85a4d3ff40035a1fd5ab17566a443b870684e87d35fc89bc0c6712"},"schema_version":"1.0"},"canonical_sha256":"7b4e8ac1d51b6447c7c18d6fe2ac26574435619bfcf8df006265674a0a63fc5c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:03:25.391575Z","signature_b64":"SGnL5em/hxPSZuN0Uq7aVgptccQW3/poYuzg05hqakRxDEXwpC0NonkI6pc5zURdxXMqqRUlGIZ4y6mgy1NeBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7b4e8ac1d51b6447c7c18d6fe2ac26574435619bfcf8df006265674a0a63fc5c","last_reissued_at":"2026-05-20T00:03:25.390730Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:03:25.390730Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.16837","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":"2026-05-20T00:03:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BwyGmmyHcXbuAWsi9ZRq4RGNe5Qw/gQkQinRrf4z3pbGHFjx21AfnPHL1QfXZGOxXdHGqLh9YZyVTSuSquoSAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T13:05:10.871616Z"},"content_sha256":"6adc73904a4c903e0260fcee5f75fdb69b89e62745925193130282f9d01d4bee","schema_version":"1.0","event_id":"sha256:6adc73904a4c903e0260fcee5f75fdb69b89e62745925193130282f9d01d4bee"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:PNHIVQOVDNSEPR6BRVX6FLBGK5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learning inelastic constitutive models from stress-strain data under hard thermodynamic constraints","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"Embedding non-equilibrium thermodynamics, objectivity and stability as hard constraints inside a neural network lets it learn consistent inelastic constitutive models from limited stress-strain data alone.","cross_cats":[],"primary_cat":"cond-mat.mtrl-sci","authors_text":"Filippo Masi","submitted_at":"2026-05-16T06:40:56Z","abstract_excerpt":"Machine learning approaches informed by physics have offered new insights into the discovery of constitutive models from data, helping overcome some limitations of traditional constitutive modelling while reducing the cost of otherwise computationally intensive simulations. Yet, most existing approaches only partially enforce the requirements of physics and thermodynamics, leaving open questions about their consistency across a broad range of material behaviours and their ability to generalise robustly to unseen loading paths when only limited measurements are available.\n  This work establishe"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The thermodynamics-constrained learning framework learns thermodynamically consistent and robust constitutive models for a range of inelastic materials of increasing complexity that generalise to more demanding, unobserved paths and can autonomously recover interpretable internal variables.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That embedding the principles of non-equilibrium thermodynamics, objectivity and stability as hard, scalable constraints in the neural architecture is sufficient to guarantee consistency and robust generalization when trained only on limited macroscopic stress-strain measurements for heterogeneous, history-dependent materials.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A thermodynamics-constrained ML framework learns robust, consistent constitutive models for inelastic materials from macroscopic stress-strain data and generalizes to unseen paths.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Embedding non-equilibrium thermodynamics, objectivity and stability as hard constraints inside a neural network lets it learn consistent inelastic constitutive models from limited stress-strain data alone.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"61d2511af1a27ab7e36be59500228016ec532e5167fa109b76ed4391463cb6ac"},"source":{"id":"2605.16837","kind":"arxiv","version":1},"verdict":{"id":"aba9be41-98b6-481a-a560-74f4d9438534","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T21:09:16.863473Z","strongest_claim":"The thermodynamics-constrained learning framework learns thermodynamically consistent and robust constitutive models for a range of inelastic materials of increasing complexity that generalise to more demanding, unobserved paths and can autonomously recover interpretable internal variables.","one_line_summary":"A thermodynamics-constrained ML framework learns robust, consistent constitutive models for inelastic materials from macroscopic stress-strain data and generalizes to unseen paths.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That embedding the principles of non-equilibrium thermodynamics, objectivity and stability as hard, scalable constraints in the neural architecture is sufficient to guarantee consistency and robust generalization when trained only on limited macroscopic stress-strain measurements for heterogeneous, history-dependent materials.","pith_extraction_headline":"Embedding non-equilibrium thermodynamics, objectivity and stability as hard constraints inside a neural network lets it learn consistent inelastic constitutive models from limited stress-strain data alone."},"integrity":{"clean":false,"summary":{"advisory":0,"critical":1,"by_detector":{"doi_compliance":{"total":1,"advisory":0,"critical":1,"informational":0}},"informational":0},"endpoint":"/pith/2605.16837/integrity.json","findings":[{"note":"Identifier '10.1016/0020-7683(94' is syntactically valid but the DOI registry (doi.org) returned 404, and Crossref / OpenAlex / internal corpus also have no record. The cited work could not be located through any authoritative source.","detector":"doi_compliance","severity":"critical","ref_index":74,"audited_at":"2026-05-19T21:21:17.159186Z","detected_doi":"10.1016/0020-7683(94","finding_type":"unresolvable_identifier","verdict_class":"cross_source","detected_arxiv_id":null}],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T21:31:19.242578Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T21:21:17.159186Z","status":"completed","version":"1.0.0","findings_count":1},{"name":"claim_evidence","ran_at":"2026-05-19T19:01:56.248005Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.397063Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"9fc6bb78bb1aee1405b7c5bf986e2febb28855b371fa8f67900ea99887e6b8e9"},"references":{"count":85,"sample":[{"doi":"","year":2017,"title":"Inputconvexneuralnetworks","work_id":"e50224ed-f5c3-4e88-87b0-e6a3dfca45a1","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1007/bf00279992","year":1976,"title":"Archive for Rational Mechanics and Analysis63(4), 337–403 (1976) https://doi","work_id":"c4aa0c13-2138-4abc-a8b2-f7d54a3bd2a1","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1016/j.mechrescom.2017.07.011","year":2018,"title":"A. Berezovski. Internal variables associated with microstructures in solids.Mechanics Research Communications, 93:30–34, 2018. doi: https://doi.org/10.1016/j.mechrescom.2017.07.011. 28 (a) (c)(b) hard","work_id":"8d7b7161-3d7b-4bdc-b31a-572f47a94010","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1007/s00161-015-0453-2","year":2016,"title":"A. Berezovski and P. Ván. Microinertia and internal variables.Continuum Mechanics and Thermodynamics, 28(4): 1027–1037, 2016. doi: https://doi.org/10.1007/s00161-015-0453-2","work_id":"0bb73e62-5fd0-42b5-9a5a-cb807556b6db","ref_index":6,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1016/j.cma.2025.118145","year":2025,"title":"J. Bleyer. Learning elastoplasticity with implicit layers.Computer Methods in Applied Mechanics and Engineering, 444:118145, 2025. doi: https://doi.org/10.1016/j.cma.2025.118145","work_id":"4234a6c5-9a66-42ea-b526-c0feb44be60c","ref_index":7,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":85,"snapshot_sha256":"94a347ebd359f74baf19af74582015d9e6ca09628dfa8a3ef6f5fc22911900a2","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"291f8cc5ebe89ce59092ac76ca6d7fba74fdb1306c63c847aa353ab9fb099585"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"aba9be41-98b6-481a-a560-74f4d9438534"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:03:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cxWx06Fa81ANGlaGg5tO9xyNZQkLvBBZk66vODKpau6xzObYNyMTD4omKVk9oePB3yO5eC+BB9Rgg2KpCUvkBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T13:05:10.872400Z"},"content_sha256":"3c1bfe7ae3afeec34ec705f0ae9810289f81b4399b17cc2d25b4f78ccf6baab7","schema_version":"1.0","event_id":"sha256:3c1bfe7ae3afeec34ec705f0ae9810289f81b4399b17cc2d25b4f78ccf6baab7"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:PNHIVQOVDNSEPR6BRVX6FLBGK5","target":"integrity","payload":{"note":"Identifier '10.1016/0020-7683(94' is syntactically valid but the DOI registry (doi.org) returned 404, and Crossref / OpenAlex / internal corpus also have no record. The cited work could not be located through any authoritative source.","snippet":"M. Rubin. Plasticity theory formulated in terms of physically based microstructural variables—Part I. Theory. International journal of solids and structures,31(19):2615–2634,1994. doi: https://doi.org/10.1016/0020-7683(94) 90222-4. 32","arxiv_id":"2605.16837","detector":"doi_compliance","evidence":{"doi":"10.1016/0020-7683(94","arxiv_id":null,"ref_index":74,"raw_excerpt":"M. Rubin. Plasticity theory formulated in terms of physically based microstructural variables—Part I. Theory. International journal of solids and structures,31(19):2615–2634,1994. doi: https://doi.org/10.1016/0020-7683(94) 90222-4. 32","verdict_class":"cross_source","checked_sources":["crossref_by_doi","openalex_by_doi","doi_org_head"]},"severity":"critical","ref_index":74,"audited_at":"2026-05-19T21:21:17.159186Z","event_type":"pith.integrity.v1","detected_doi":"10.1016/0020-7683(94","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"unresolvable_identifier","evidence_hash":"1f35eaaaa94ac3c438fd0972d01b212fc984803135a0a81fe9244f2d97d4e24c","paper_version":1,"verdict_class":"cross_source","resolved_title":null,"detector_version":"1.0.0","detected_arxiv_id":null,"integrity_event_id":3149,"payload_sha256":"a0ab7594105410292f36538b3481c452817bf74557b82b247bf1921de45b88a5","signature_b64":"467e5fJfefpxw5SxXj4vhsAm9DtO83FSXIDS5USOQSmLqgKvZcI+G/z5rV4ksGNAtVadOZfM8zu6lQZ+a7adAg==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-19T21:22:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"crwyaokGn8bQO0rkku8Ee0Kqh2IfhU3JMkrC1AZO8Z3M7aSpSKl1NbXyxSsHSTlMW8DGj81jXp3zRORTfhlgCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T13:05:10.873515Z"},"content_sha256":"e1bb7dc00da9f64d23da4dd29575849842e07a81e8b05087b1e10e7d84709573","schema_version":"1.0","event_id":"sha256:e1bb7dc00da9f64d23da4dd29575849842e07a81e8b05087b1e10e7d84709573"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PNHIVQOVDNSEPR6BRVX6FLBGK5/bundle.json","state_url":"https://pith.science/pith/PNHIVQOVDNSEPR6BRVX6FLBGK5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PNHIVQOVDNSEPR6BRVX6FLBGK5/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":"2026-05-30T13:05:10Z","links":{"resolver":"https://pith.science/pith/PNHIVQOVDNSEPR6BRVX6FLBGK5","bundle":"https://pith.science/pith/PNHIVQOVDNSEPR6BRVX6FLBGK5/bundle.json","state":"https://pith.science/pith/PNHIVQOVDNSEPR6BRVX6FLBGK5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PNHIVQOVDNSEPR6BRVX6FLBGK5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:PNHIVQOVDNSEPR6BRVX6FLBGK5","merge_version":"pith-open-graph-merge-v1","event_count":3,"valid_event_count":3,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"ff0613b11f85a4d3ff40035a1fd5ab17566a443b870684e87d35fc89bc0c6712","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cond-mat.mtrl-sci","submitted_at":"2026-05-16T06:40:56Z","title_canon_sha256":"5f7fece14ea4c39ddd7e84309082739f4da177c4a7ad1e4635b5ac94a6d1550e"},"schema_version":"1.0","source":{"id":"2605.16837","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.16837","created_at":"2026-05-20T00:03:25Z"},{"alias_kind":"arxiv_version","alias_value":"2605.16837v1","created_at":"2026-05-20T00:03:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16837","created_at":"2026-05-20T00:03:25Z"},{"alias_kind":"pith_short_12","alias_value":"PNHIVQOVDNSE","created_at":"2026-05-20T00:03:25Z"},{"alias_kind":"pith_short_16","alias_value":"PNHIVQOVDNSEPR6B","created_at":"2026-05-20T00:03:25Z"},{"alias_kind":"pith_short_8","alias_value":"PNHIVQOV","created_at":"2026-05-20T00:03:25Z"}],"graph_snapshots":[{"event_id":"sha256:3c1bfe7ae3afeec34ec705f0ae9810289f81b4399b17cc2d25b4f78ccf6baab7","target":"graph","created_at":"2026-05-20T00:03:25Z","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":"The thermodynamics-constrained learning framework learns thermodynamically consistent and robust constitutive models for a range of inelastic materials of increasing complexity that generalise to more demanding, unobserved paths and can autonomously recover interpretable internal variables."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That embedding the principles of non-equilibrium thermodynamics, objectivity and stability as hard, scalable constraints in the neural architecture is sufficient to guarantee consistency and robust generalization when trained only on limited macroscopic stress-strain measurements for heterogeneous, history-dependent materials."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A thermodynamics-constrained ML framework learns robust, consistent constitutive models for inelastic materials from macroscopic stress-strain data and generalizes to unseen paths."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Embedding non-equilibrium thermodynamics, objectivity and stability as hard constraints inside a neural network lets it learn consistent inelastic constitutive models from limited stress-strain data alone."}],"snapshot_sha256":"61d2511af1a27ab7e36be59500228016ec532e5167fa109b76ed4391463cb6ac"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"291f8cc5ebe89ce59092ac76ca6d7fba74fdb1306c63c847aa353ab9fb099585"},"integrity":{"available":true,"clean":false,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T21:31:19.242578Z","status":"completed","version":"1.0.0"},{"findings_count":1,"name":"doi_compliance","ran_at":"2026-05-19T21:21:17.159186Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T19:01:56.248005Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.397063Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.16837/integrity.json","findings":[{"audited_at":"2026-05-19T21:21:17.159186Z","detected_arxiv_id":null,"detected_doi":"10.1016/0020-7683(94","detector":"doi_compliance","finding_type":"unresolvable_identifier","note":"Identifier '10.1016/0020-7683(94' is syntactically valid but the DOI registry (doi.org) returned 404, and Crossref / OpenAlex / internal corpus also have no record. The cited work could not be located through any authoritative source.","ref_index":74,"severity":"critical","verdict_class":"cross_source"}],"snapshot_sha256":"9fc6bb78bb1aee1405b7c5bf986e2febb28855b371fa8f67900ea99887e6b8e9","summary":{"advisory":0,"by_detector":{"doi_compliance":{"advisory":0,"critical":1,"informational":0,"total":1}},"critical":1,"informational":0}},"paper":{"abstract_excerpt":"Machine learning approaches informed by physics have offered new insights into the discovery of constitutive models from data, helping overcome some limitations of traditional constitutive modelling while reducing the cost of otherwise computationally intensive simulations. Yet, most existing approaches only partially enforce the requirements of physics and thermodynamics, leaving open questions about their consistency across a broad range of material behaviours and their ability to generalise robustly to unseen loading paths when only limited measurements are available.\n  This work establishe","authors_text":"Filippo Masi","cross_cats":[],"headline":"Embedding non-equilibrium thermodynamics, objectivity and stability as hard constraints inside a neural network lets it learn consistent inelastic constitutive models from limited stress-strain data alone.","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cond-mat.mtrl-sci","submitted_at":"2026-05-16T06:40:56Z","title":"Learning inelastic constitutive models from stress-strain data under hard thermodynamic constraints"},"references":{"count":85,"internal_anchors":0,"resolved_work":85,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Inputconvexneuralnetworks","work_id":"e50224ed-f5c3-4e88-87b0-e6a3dfca45a1","year":2017},{"cited_arxiv_id":"","doi":"10.1007/bf00279992","is_internal_anchor":false,"ref_index":4,"title":"Archive for Rational Mechanics and Analysis63(4), 337–403 (1976) https://doi","work_id":"c4aa0c13-2138-4abc-a8b2-f7d54a3bd2a1","year":1976},{"cited_arxiv_id":"","doi":"10.1016/j.mechrescom.2017.07.011","is_internal_anchor":false,"ref_index":5,"title":"A. Berezovski. Internal variables associated with microstructures in solids.Mechanics Research Communications, 93:30–34, 2018. doi: https://doi.org/10.1016/j.mechrescom.2017.07.011. 28 (a) (c)(b) hard","work_id":"8d7b7161-3d7b-4bdc-b31a-572f47a94010","year":2018},{"cited_arxiv_id":"","doi":"10.1007/s00161-015-0453-2","is_internal_anchor":false,"ref_index":6,"title":"A. Berezovski and P. Ván. Microinertia and internal variables.Continuum Mechanics and Thermodynamics, 28(4): 1027–1037, 2016. doi: https://doi.org/10.1007/s00161-015-0453-2","work_id":"0bb73e62-5fd0-42b5-9a5a-cb807556b6db","year":2016},{"cited_arxiv_id":"","doi":"10.1016/j.cma.2025.118145","is_internal_anchor":false,"ref_index":7,"title":"J. Bleyer. Learning elastoplasticity with implicit layers.Computer Methods in Applied Mechanics and Engineering, 444:118145, 2025. doi: https://doi.org/10.1016/j.cma.2025.118145","work_id":"4234a6c5-9a66-42ea-b526-c0feb44be60c","year":2025}],"snapshot_sha256":"94a347ebd359f74baf19af74582015d9e6ca09628dfa8a3ef6f5fc22911900a2"},"source":{"id":"2605.16837","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T21:09:16.863473Z","id":"aba9be41-98b6-481a-a560-74f4d9438534","model_set":{"reader":"grok-4.3"},"one_line_summary":"A thermodynamics-constrained ML framework learns robust, consistent constitutive models for inelastic materials from macroscopic stress-strain data and generalizes to unseen paths.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Embedding non-equilibrium thermodynamics, objectivity and stability as hard constraints inside a neural network lets it learn consistent inelastic constitutive models from limited stress-strain data alone.","strongest_claim":"The thermodynamics-constrained learning framework learns thermodynamically consistent and robust constitutive models for a range of inelastic materials of increasing complexity that generalise to more demanding, unobserved paths and can autonomously recover interpretable internal variables.","weakest_assumption":"That embedding the principles of non-equilibrium thermodynamics, objectivity and stability as hard, scalable constraints in the neural architecture is sufficient to guarantee consistency and robust generalization when trained only on limited macroscopic stress-strain measurements for heterogeneous, history-dependent materials."}},"verdict_id":"aba9be41-98b6-481a-a560-74f4d9438534"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6adc73904a4c903e0260fcee5f75fdb69b89e62745925193130282f9d01d4bee","target":"record","created_at":"2026-05-20T00:03:25Z","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":"ff0613b11f85a4d3ff40035a1fd5ab17566a443b870684e87d35fc89bc0c6712","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cond-mat.mtrl-sci","submitted_at":"2026-05-16T06:40:56Z","title_canon_sha256":"5f7fece14ea4c39ddd7e84309082739f4da177c4a7ad1e4635b5ac94a6d1550e"},"schema_version":"1.0","source":{"id":"2605.16837","kind":"arxiv","version":1}},"canonical_sha256":"7b4e8ac1d51b6447c7c18d6fe2ac26574435619bfcf8df006265674a0a63fc5c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7b4e8ac1d51b6447c7c18d6fe2ac26574435619bfcf8df006265674a0a63fc5c","first_computed_at":"2026-05-20T00:03:25.390730Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:03:25.390730Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SGnL5em/hxPSZuN0Uq7aVgptccQW3/poYuzg05hqakRxDEXwpC0NonkI6pc5zURdxXMqqRUlGIZ4y6mgy1NeBw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:03:25.391575Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.16837","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e1bb7dc00da9f64d23da4dd29575849842e07a81e8b05087b1e10e7d84709573","sha256:6adc73904a4c903e0260fcee5f75fdb69b89e62745925193130282f9d01d4bee","sha256:3c1bfe7ae3afeec34ec705f0ae9810289f81b4399b17cc2d25b4f78ccf6baab7"],"state_sha256":"a9d79f2ed8240cd5dd3abcce9363ed29d43fdef83e780a2d1cd69b9df93d3bbf"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RFv6rvZwuNUe1CcIAoxsV8pyyUbsH3MqwTQ/snFB1UxWfurPODQFe9YIMqH4OZ+QsmwLmZT6ZH4P37ioB/BxAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-30T13:05:10.876432Z","bundle_sha256":"a53ac14921093734d61f6145da941e58a66b732ed33d1315123e719f01ddfae3"}}