{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:L6JCEDKVGC6MDY3ORDADQQPBFP","short_pith_number":"pith:L6JCEDKV","schema_version":"1.0","canonical_sha256":"5f92220d5530bcc1e36e88c03841e12bd473366a0f4d746d9182bbdc7c017a28","source":{"kind":"arxiv","id":"2108.12463","version":2},"attestation_state":"computed","paper":{"title":"Automatic Text Evaluation through the Lens of Wasserstein Barycenters","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chloe Clavel, Guillaume Staerman, Pablo Piantanida, Pierre Colombo","submitted_at":"2021-08-27T19:08:52Z","abstract_excerpt":"A new metric \\texttt{BaryScore} to evaluate text generation based on deep contextualized embeddings e.g., BERT, Roberta, ELMo) is introduced. This metric is motivated by a new framework relying on optimal transport tools, i.e., Wasserstein distance and barycenter. By modelling the layer output of deep contextualized embeddings as a probability distribution rather than by a vector embedding; this framework provides a natural way to aggregate the different outputs through the Wasserstein space topology. In addition, it provides theoretical grounds to our metric and offers an alternative to avail"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2108.12463","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-08-27T19:08:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3097a55495b1926fb11c104219fd65a16d6387036ebebef6ae776160e3303556","abstract_canon_sha256":"473a15fc2e3339eff010c26bb3a78c4d64be53e2bd349ef7f08a2a54081f2150"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:12:50.352873Z","signature_b64":"T46mlmcXCwGt04Uk28TDCM/HyuQQFWMx1O9QWCxc15FU0DWNzd6FUOupv37x3AOECUfAAHxl7C6otzSnQzWECA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f92220d5530bcc1e36e88c03841e12bd473366a0f4d746d9182bbdc7c017a28","last_reissued_at":"2026-07-05T03:12:50.352472Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:12:50.352472Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Automatic Text Evaluation through the Lens of Wasserstein Barycenters","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chloe Clavel, Guillaume Staerman, Pablo Piantanida, Pierre Colombo","submitted_at":"2021-08-27T19:08:52Z","abstract_excerpt":"A new metric \\texttt{BaryScore} to evaluate text generation based on deep contextualized embeddings e.g., BERT, Roberta, ELMo) is introduced. This metric is motivated by a new framework relying on optimal transport tools, i.e., Wasserstein distance and barycenter. By modelling the layer output of deep contextualized embeddings as a probability distribution rather than by a vector embedding; this framework provides a natural way to aggregate the different outputs through the Wasserstein space topology. In addition, it provides theoretical grounds to our metric and offers an alternative to avail"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.12463","kind":"arxiv","version":2},"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/2108.12463/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2108.12463","created_at":"2026-07-05T03:12:50.352534+00:00"},{"alias_kind":"arxiv_version","alias_value":"2108.12463v2","created_at":"2026-07-05T03:12:50.352534+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.12463","created_at":"2026-07-05T03:12:50.352534+00:00"},{"alias_kind":"pith_short_12","alias_value":"L6JCEDKVGC6M","created_at":"2026-07-05T03:12:50.352534+00:00"},{"alias_kind":"pith_short_16","alias_value":"L6JCEDKVGC6MDY3O","created_at":"2026-07-05T03:12:50.352534+00:00"},{"alias_kind":"pith_short_8","alias_value":"L6JCEDKV","created_at":"2026-07-05T03:12:50.352534+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.03626","citing_title":"Explainable Knowledge Graph Retrieval-Augmented Generation (KG-RAG) with KG-SMILE","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L6JCEDKVGC6MDY3ORDADQQPBFP","json":"https://pith.science/pith/L6JCEDKVGC6MDY3ORDADQQPBFP.json","graph_json":"https://pith.science/api/pith-number/L6JCEDKVGC6MDY3ORDADQQPBFP/graph.json","events_json":"https://pith.science/api/pith-number/L6JCEDKVGC6MDY3ORDADQQPBFP/events.json","paper":"https://pith.science/paper/L6JCEDKV"},"agent_actions":{"view_html":"https://pith.science/pith/L6JCEDKVGC6MDY3ORDADQQPBFP","download_json":"https://pith.science/pith/L6JCEDKVGC6MDY3ORDADQQPBFP.json","view_paper":"https://pith.science/paper/L6JCEDKV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2108.12463&json=true","fetch_graph":"https://pith.science/api/pith-number/L6JCEDKVGC6MDY3ORDADQQPBFP/graph.json","fetch_events":"https://pith.science/api/pith-number/L6JCEDKVGC6MDY3ORDADQQPBFP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L6JCEDKVGC6MDY3ORDADQQPBFP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L6JCEDKVGC6MDY3ORDADQQPBFP/action/storage_attestation","attest_author":"https://pith.science/pith/L6JCEDKVGC6MDY3ORDADQQPBFP/action/author_attestation","sign_citation":"https://pith.science/pith/L6JCEDKVGC6MDY3ORDADQQPBFP/action/citation_signature","submit_replication":"https://pith.science/pith/L6JCEDKVGC6MDY3ORDADQQPBFP/action/replication_record"}},"created_at":"2026-07-05T03:12:50.352534+00:00","updated_at":"2026-07-05T03:12:50.352534+00:00"}