{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:BVVYDGRBD5YHRLLHP62ENC2BSQ","short_pith_number":"pith:BVVYDGRB","schema_version":"1.0","canonical_sha256":"0d6b819a211f7078ad677fb4468b41942d36c5b1a34910fb048aea07e87fb86b","source":{"kind":"arxiv","id":"2508.07743","version":2},"attestation_state":"computed","paper":{"title":"Symmetry-Aware Transformer Training for Automated Planning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Elliot Gestrin, Jendrik Seipp, Markus Fritzsche","submitted_at":"2025-08-11T08:23:34Z","abstract_excerpt":"While transformers excel in many settings, their application in the field of automated planning is limited. Prior work like PlanGPT, a state-of-the-art decoder-only transformer, struggles with extrapolation from easy to hard planning problems. This in turn stems from problem symmetries: planning tasks can be represented with arbitrary variable names that carry no meaning beyond being identifiers. This causes a combinatorial explosion of equivalent representations that pure transformers cannot efficiently learn from. We propose a novel contrastive learning objective to make transformers symmetr"},"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":"2508.07743","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-08-11T08:23:34Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"469248b1b1e09262700085b9ae8499b9457499bcbf772a9ac2cab0f32cd50cc2","abstract_canon_sha256":"552aa8489faf964b5d7845f12361f87390e914b85e71c7cc7d85169592898013"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-29T01:15:02.922202Z","signature_b64":"H3TLws5Wriwrwk/G7kw2gQWjwp6WFJZuaD7RYFb6sfS3ngVOiRPGmcsQwSSoQncE7F98ioo3h9r/e2xVGmlHAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0d6b819a211f7078ad677fb4468b41942d36c5b1a34910fb048aea07e87fb86b","last_reissued_at":"2026-06-29T01:15:02.921709Z","signature_status":"signed_v1","first_computed_at":"2026-06-29T01:15:02.921709Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Symmetry-Aware Transformer Training for Automated Planning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Elliot Gestrin, Jendrik Seipp, Markus Fritzsche","submitted_at":"2025-08-11T08:23:34Z","abstract_excerpt":"While transformers excel in many settings, their application in the field of automated planning is limited. Prior work like PlanGPT, a state-of-the-art decoder-only transformer, struggles with extrapolation from easy to hard planning problems. This in turn stems from problem symmetries: planning tasks can be represented with arbitrary variable names that carry no meaning beyond being identifiers. This causes a combinatorial explosion of equivalent representations that pure transformers cannot efficiently learn from. We propose a novel contrastive learning objective to make transformers symmetr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.07743","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/2508.07743/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":"2508.07743","created_at":"2026-06-29T01:15:02.921768+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.07743v2","created_at":"2026-06-29T01:15:02.921768+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07743","created_at":"2026-06-29T01:15:02.921768+00:00"},{"alias_kind":"pith_short_12","alias_value":"BVVYDGRBD5YH","created_at":"2026-06-29T01:15:02.921768+00:00"},{"alias_kind":"pith_short_16","alias_value":"BVVYDGRBD5YHRLLH","created_at":"2026-06-29T01:15:02.921768+00:00"},{"alias_kind":"pith_short_8","alias_value":"BVVYDGRB","created_at":"2026-06-29T01:15:02.921768+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2606.00618","citing_title":"Efficient Test-time Inference for Generative Planning Models with OCL Search","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2605.03625","citing_title":"Self-Improvement for Fast, High-Quality Plan Generation","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BVVYDGRBD5YHRLLHP62ENC2BSQ","json":"https://pith.science/pith/BVVYDGRBD5YHRLLHP62ENC2BSQ.json","graph_json":"https://pith.science/api/pith-number/BVVYDGRBD5YHRLLHP62ENC2BSQ/graph.json","events_json":"https://pith.science/api/pith-number/BVVYDGRBD5YHRLLHP62ENC2BSQ/events.json","paper":"https://pith.science/paper/BVVYDGRB"},"agent_actions":{"view_html":"https://pith.science/pith/BVVYDGRBD5YHRLLHP62ENC2BSQ","download_json":"https://pith.science/pith/BVVYDGRBD5YHRLLHP62ENC2BSQ.json","view_paper":"https://pith.science/paper/BVVYDGRB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.07743&json=true","fetch_graph":"https://pith.science/api/pith-number/BVVYDGRBD5YHRLLHP62ENC2BSQ/graph.json","fetch_events":"https://pith.science/api/pith-number/BVVYDGRBD5YHRLLHP62ENC2BSQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BVVYDGRBD5YHRLLHP62ENC2BSQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BVVYDGRBD5YHRLLHP62ENC2BSQ/action/storage_attestation","attest_author":"https://pith.science/pith/BVVYDGRBD5YHRLLHP62ENC2BSQ/action/author_attestation","sign_citation":"https://pith.science/pith/BVVYDGRBD5YHRLLHP62ENC2BSQ/action/citation_signature","submit_replication":"https://pith.science/pith/BVVYDGRBD5YHRLLHP62ENC2BSQ/action/replication_record"}},"created_at":"2026-06-29T01:15:02.921768+00:00","updated_at":"2026-06-29T01:15:02.921768+00:00"}