{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:3ZHM62E7UQUWQAOP2UTMJHVW5D","short_pith_number":"pith:3ZHM62E7","canonical_record":{"source":{"id":"2603.16453","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.AI","submitted_at":"2026-03-17T12:35:52Z","cross_cats_sorted":[],"title_canon_sha256":"110429c64813f96d3ed99e20596684dfe6eb245c457e26969e15bc9be58968b1","abstract_canon_sha256":"280ee425504098bfc5b57948c62d8130bfe0e4388e7810ce96158454558f9366"},"schema_version":"1.0"},"canonical_sha256":"de4ecf689fa4296801cfd526c49eb6e8e5e359d21859c0000e29e151f373f14b","source":{"kind":"arxiv","id":"2603.16453","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.16453","created_at":"2026-06-23T00:11:49Z"},{"alias_kind":"arxiv_version","alias_value":"2603.16453v2","created_at":"2026-06-23T00:11:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.16453","created_at":"2026-06-23T00:11:49Z"},{"alias_kind":"pith_short_12","alias_value":"3ZHM62E7UQUW","created_at":"2026-06-23T00:11:49Z"},{"alias_kind":"pith_short_16","alias_value":"3ZHM62E7UQUWQAOP","created_at":"2026-06-23T00:11:49Z"},{"alias_kind":"pith_short_8","alias_value":"3ZHM62E7","created_at":"2026-06-23T00:11:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:3ZHM62E7UQUWQAOP2UTMJHVW5D","target":"record","payload":{"canonical_record":{"source":{"id":"2603.16453","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.AI","submitted_at":"2026-03-17T12:35:52Z","cross_cats_sorted":[],"title_canon_sha256":"110429c64813f96d3ed99e20596684dfe6eb245c457e26969e15bc9be58968b1","abstract_canon_sha256":"280ee425504098bfc5b57948c62d8130bfe0e4388e7810ce96158454558f9366"},"schema_version":"1.0"},"canonical_sha256":"de4ecf689fa4296801cfd526c49eb6e8e5e359d21859c0000e29e151f373f14b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T00:11:49.312106Z","signature_b64":"ZzAdffYZef8sPHGMKM5BjDh3je6ASxxVYyiO/UsfGT5pnI8pjucVIPsfJpclCzbelfOl69TO1GAaI/+PiqNKAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"de4ecf689fa4296801cfd526c49eb6e8e5e359d21859c0000e29e151f373f14b","last_reissued_at":"2026-06-23T00:11:49.311627Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T00:11:49.311627Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2603.16453","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-23T00:11:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jLCBXpY34dNLYAqxEGxRaLavYM2bScjV5U//BUevrf7BHUUAT5/KbedhdwByhaqmGql9L9a7JaGz/g4RzzJ6BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T00:16:29.280855Z"},"content_sha256":"2397a363a8166f190f2523479edccaa9e16651d4e8a6cc4ded209abe1f91bfaa","schema_version":"1.0","event_id":"sha256:2397a363a8166f190f2523479edccaa9e16651d4e8a6cc4ded209abe1f91bfaa"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:3ZHM62E7UQUWQAOP2UTMJHVW5D","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RetailBench: Evaluating Long-Horizon Autonomous Decision-Making and Strategy Stability of LLM Agents in Realistic Retail Environments","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jingtong Wu, Jun Wang, Linghua Zhang, Zhisong Zhang","submitted_at":"2026-03-17T12:35:52Z","abstract_excerpt":"Large language model (LLM) agents have made rapid progress on short-horizon, well-scoped tasks, yet their ability to sustain coherent decisions in dynamic long-horizon environments remains uncertain. We introduce RetailBench, a data-grounded simulation benchmark for evaluating tool-using LLM agents in single-store supermarket operation. RetailBench models retail management as a partially observable decision process and is designed to support thousand-day-scale simulations. In this environment, agents must manage pricing, replenishment, supplier selection, shelf assortment, inventory aging, cus"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.16453","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/2603.16453/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":"2026-06-23T00:11:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RH25HMdxIM37HvGv9jaCSQmj4QRqmOSJqzvWXOInXsHTAn96bAmXL6U5OIvslUOmw5b518STB5I0JYCH1zkMAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T00:16:29.281220Z"},"content_sha256":"e565f75bd870b7e4b1d786e1e9c50c301f7ce23e88e68b7313ceab081dc9a09d","schema_version":"1.0","event_id":"sha256:e565f75bd870b7e4b1d786e1e9c50c301f7ce23e88e68b7313ceab081dc9a09d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3ZHM62E7UQUWQAOP2UTMJHVW5D/bundle.json","state_url":"https://pith.science/pith/3ZHM62E7UQUWQAOP2UTMJHVW5D/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3ZHM62E7UQUWQAOP2UTMJHVW5D/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":"2026-06-27T00:16:29Z","links":{"resolver":"https://pith.science/pith/3ZHM62E7UQUWQAOP2UTMJHVW5D","bundle":"https://pith.science/pith/3ZHM62E7UQUWQAOP2UTMJHVW5D/bundle.json","state":"https://pith.science/pith/3ZHM62E7UQUWQAOP2UTMJHVW5D/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3ZHM62E7UQUWQAOP2UTMJHVW5D/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:3ZHM62E7UQUWQAOP2UTMJHVW5D","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":"280ee425504098bfc5b57948c62d8130bfe0e4388e7810ce96158454558f9366","cross_cats_sorted":[],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.AI","submitted_at":"2026-03-17T12:35:52Z","title_canon_sha256":"110429c64813f96d3ed99e20596684dfe6eb245c457e26969e15bc9be58968b1"},"schema_version":"1.0","source":{"id":"2603.16453","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.16453","created_at":"2026-06-23T00:11:49Z"},{"alias_kind":"arxiv_version","alias_value":"2603.16453v2","created_at":"2026-06-23T00:11:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.16453","created_at":"2026-06-23T00:11:49Z"},{"alias_kind":"pith_short_12","alias_value":"3ZHM62E7UQUW","created_at":"2026-06-23T00:11:49Z"},{"alias_kind":"pith_short_16","alias_value":"3ZHM62E7UQUWQAOP","created_at":"2026-06-23T00:11:49Z"},{"alias_kind":"pith_short_8","alias_value":"3ZHM62E7","created_at":"2026-06-23T00:11:49Z"}],"graph_snapshots":[{"event_id":"sha256:e565f75bd870b7e4b1d786e1e9c50c301f7ce23e88e68b7313ceab081dc9a09d","target":"graph","created_at":"2026-06-23T00:11:49Z","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/2603.16453/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language model (LLM) agents have made rapid progress on short-horizon, well-scoped tasks, yet their ability to sustain coherent decisions in dynamic long-horizon environments remains uncertain. We introduce RetailBench, a data-grounded simulation benchmark for evaluating tool-using LLM agents in single-store supermarket operation. RetailBench models retail management as a partially observable decision process and is designed to support thousand-day-scale simulations. In this environment, agents must manage pricing, replenishment, supplier selection, shelf assortment, inventory aging, cus","authors_text":"Jingtong Wu, Jun Wang, Linghua Zhang, Zhisong Zhang","cross_cats":[],"headline":"","license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.AI","submitted_at":"2026-03-17T12:35:52Z","title":"RetailBench: Evaluating Long-Horizon Autonomous Decision-Making and Strategy Stability of LLM Agents in Realistic Retail Environments"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.16453","kind":"arxiv","version":2},"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":"sha256:2397a363a8166f190f2523479edccaa9e16651d4e8a6cc4ded209abe1f91bfaa","target":"record","created_at":"2026-06-23T00:11:49Z","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":"280ee425504098bfc5b57948c62d8130bfe0e4388e7810ce96158454558f9366","cross_cats_sorted":[],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.AI","submitted_at":"2026-03-17T12:35:52Z","title_canon_sha256":"110429c64813f96d3ed99e20596684dfe6eb245c457e26969e15bc9be58968b1"},"schema_version":"1.0","source":{"id":"2603.16453","kind":"arxiv","version":2}},"canonical_sha256":"de4ecf689fa4296801cfd526c49eb6e8e5e359d21859c0000e29e151f373f14b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"de4ecf689fa4296801cfd526c49eb6e8e5e359d21859c0000e29e151f373f14b","first_computed_at":"2026-06-23T00:11:49.311627Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-23T00:11:49.311627Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZzAdffYZef8sPHGMKM5BjDh3je6ASxxVYyiO/UsfGT5pnI8pjucVIPsfJpclCzbelfOl69TO1GAaI/+PiqNKAA==","signature_status":"signed_v1","signed_at":"2026-06-23T00:11:49.312106Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.16453","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2397a363a8166f190f2523479edccaa9e16651d4e8a6cc4ded209abe1f91bfaa","sha256:e565f75bd870b7e4b1d786e1e9c50c301f7ce23e88e68b7313ceab081dc9a09d"],"state_sha256":"a9dbf0cf3b6c27ce40aae531270534c200efde5be29ff559269137e95ae0ac11"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mCQmf5VM4Pd9PXCWu+hepIXf0NiTcbirtRoaE11E+UV9JNGNz79znImtAfcQeJrw2+y2rro8zHyg3ZPeS15hDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-27T00:16:29.283098Z","bundle_sha256":"317fc32e61ff914f61b03695ea70ba3a692d913300489a4c7540431564fbff04"}}