{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:NMA5VANDOEABQKWDQEYLB725RZ","short_pith_number":"pith:NMA5VAND","canonical_record":{"source":{"id":"2502.15840","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-20T15:52:29Z","cross_cats_sorted":[],"title_canon_sha256":"adc7ea87fa723dc4d501543711fb2f7cc059a42d0cfddc0cf55ba531af81f115","abstract_canon_sha256":"ba0cc68b18019519ba1c3361d1774f9cbe00509504561baf211ea008e53b9c03"},"schema_version":"1.0"},"canonical_sha256":"6b01da81a37100182ac38130b0ff5d8e7bc3b8ef1b03e1792c0c1d0e926abfbd","source":{"kind":"arxiv","id":"2502.15840","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.15840","created_at":"2026-07-05T10:18:17Z"},{"alias_kind":"arxiv_version","alias_value":"2502.15840v1","created_at":"2026-07-05T10:18:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.15840","created_at":"2026-07-05T10:18:17Z"},{"alias_kind":"pith_short_12","alias_value":"NMA5VANDOEAB","created_at":"2026-07-05T10:18:17Z"},{"alias_kind":"pith_short_16","alias_value":"NMA5VANDOEABQKWD","created_at":"2026-07-05T10:18:17Z"},{"alias_kind":"pith_short_8","alias_value":"NMA5VAND","created_at":"2026-07-05T10:18:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:NMA5VANDOEABQKWDQEYLB725RZ","target":"record","payload":{"canonical_record":{"source":{"id":"2502.15840","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-20T15:52:29Z","cross_cats_sorted":[],"title_canon_sha256":"adc7ea87fa723dc4d501543711fb2f7cc059a42d0cfddc0cf55ba531af81f115","abstract_canon_sha256":"ba0cc68b18019519ba1c3361d1774f9cbe00509504561baf211ea008e53b9c03"},"schema_version":"1.0"},"canonical_sha256":"6b01da81a37100182ac38130b0ff5d8e7bc3b8ef1b03e1792c0c1d0e926abfbd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:18:17.123888Z","signature_b64":"zPqLJ6DBAVXYGR0/oVM3RAF8pcFdMvHCyNQSKpJKQYCa0ZyHLVs98a0NZJ759IfkAltduZcKgqgFPQoG9jz9Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6b01da81a37100182ac38130b0ff5d8e7bc3b8ef1b03e1792c0c1d0e926abfbd","last_reissued_at":"2026-07-05T10:18:17.123401Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:18:17.123401Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.15840","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":"2026-07-05T10:18:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"q8CQec5rrLEJwTpersn7F0dWsb5IssRmutijPC/oPNYC7rV5CjLR3keAPH40LqdMrTFul9qGbQrmbJnHYk+LAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-02T06:21:38.574086Z"},"content_sha256":"099c745d72072d010ee5c89c30d94d400d477c2553eb077c1caa35e2bc0b3286","schema_version":"1.0","event_id":"sha256:099c745d72072d010ee5c89c30d94d400d477c2553eb077c1caa35e2bc0b3286"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:NMA5VANDOEABQKWDQEYLB725RZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Vending-Bench: A Benchmark for Long-Term Coherence of Autonomous Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Axel Backlund, Lukas Petersson","submitted_at":"2025-02-20T15:52:29Z","abstract_excerpt":"While Large Language Models (LLMs) can exhibit impressive proficiency in isolated, short-term tasks, they often fail to maintain coherent performance over longer time horizons. In this paper, we present Vending-Bench, a simulated environment designed to specifically test an LLM-based agent's ability to manage a straightforward, long-running business scenario: operating a vending machine. Agents must balance inventories, place orders, set prices, and handle daily fees - tasks that are each simple but collectively, over long horizons (>20M tokens per run) stress an LLM's capacity for sustained, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.15840","kind":"arxiv","version":1},"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/2502.15840/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":"2026-07-05T10:18:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wOjxSm2y1UmFS5gCIAD5u313cTIH8Hl7u3UXh8pvtCLKaRwIt8ieMVLB06s6gxCjEaN981jtyFw0/PhGsevFBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-02T06:21:38.574596Z"},"content_sha256":"6327d74c3e51ba1aacfaa0a8efb23e39b8ab81b2d1757ae961c8f3bd5cc6cc58","schema_version":"1.0","event_id":"sha256:6327d74c3e51ba1aacfaa0a8efb23e39b8ab81b2d1757ae961c8f3bd5cc6cc58"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NMA5VANDOEABQKWDQEYLB725RZ/bundle.json","state_url":"https://pith.science/pith/NMA5VANDOEABQKWDQEYLB725RZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NMA5VANDOEABQKWDQEYLB725RZ/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":"2026-08-02T06:21:38Z","links":{"resolver":"https://pith.science/pith/NMA5VANDOEABQKWDQEYLB725RZ","bundle":"https://pith.science/pith/NMA5VANDOEABQKWDQEYLB725RZ/bundle.json","state":"https://pith.science/pith/NMA5VANDOEABQKWDQEYLB725RZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NMA5VANDOEABQKWDQEYLB725RZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:NMA5VANDOEABQKWDQEYLB725RZ","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":"ba0cc68b18019519ba1c3361d1774f9cbe00509504561baf211ea008e53b9c03","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-20T15:52:29Z","title_canon_sha256":"adc7ea87fa723dc4d501543711fb2f7cc059a42d0cfddc0cf55ba531af81f115"},"schema_version":"1.0","source":{"id":"2502.15840","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.15840","created_at":"2026-07-05T10:18:17Z"},{"alias_kind":"arxiv_version","alias_value":"2502.15840v1","created_at":"2026-07-05T10:18:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.15840","created_at":"2026-07-05T10:18:17Z"},{"alias_kind":"pith_short_12","alias_value":"NMA5VANDOEAB","created_at":"2026-07-05T10:18:17Z"},{"alias_kind":"pith_short_16","alias_value":"NMA5VANDOEABQKWD","created_at":"2026-07-05T10:18:17Z"},{"alias_kind":"pith_short_8","alias_value":"NMA5VAND","created_at":"2026-07-05T10:18:17Z"}],"graph_snapshots":[{"event_id":"sha256:6327d74c3e51ba1aacfaa0a8efb23e39b8ab81b2d1757ae961c8f3bd5cc6cc58","target":"graph","created_at":"2026-07-05T10:18:17Z","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/2502.15840/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While Large Language Models (LLMs) can exhibit impressive proficiency in isolated, short-term tasks, they often fail to maintain coherent performance over longer time horizons. In this paper, we present Vending-Bench, a simulated environment designed to specifically test an LLM-based agent's ability to manage a straightforward, long-running business scenario: operating a vending machine. Agents must balance inventories, place orders, set prices, and handle daily fees - tasks that are each simple but collectively, over long horizons (>20M tokens per run) stress an LLM's capacity for sustained, ","authors_text":"Axel Backlund, Lukas Petersson","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-20T15:52:29Z","title":"Vending-Bench: A Benchmark for Long-Term Coherence of Autonomous Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.15840","kind":"arxiv","version":1},"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":"sha256:099c745d72072d010ee5c89c30d94d400d477c2553eb077c1caa35e2bc0b3286","target":"record","created_at":"2026-07-05T10:18:17Z","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":"ba0cc68b18019519ba1c3361d1774f9cbe00509504561baf211ea008e53b9c03","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-20T15:52:29Z","title_canon_sha256":"adc7ea87fa723dc4d501543711fb2f7cc059a42d0cfddc0cf55ba531af81f115"},"schema_version":"1.0","source":{"id":"2502.15840","kind":"arxiv","version":1}},"canonical_sha256":"6b01da81a37100182ac38130b0ff5d8e7bc3b8ef1b03e1792c0c1d0e926abfbd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6b01da81a37100182ac38130b0ff5d8e7bc3b8ef1b03e1792c0c1d0e926abfbd","first_computed_at":"2026-07-05T10:18:17.123401Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:18:17.123401Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zPqLJ6DBAVXYGR0/oVM3RAF8pcFdMvHCyNQSKpJKQYCa0ZyHLVs98a0NZJ759IfkAltduZcKgqgFPQoG9jz9Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:18:17.123888Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.15840","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:099c745d72072d010ee5c89c30d94d400d477c2553eb077c1caa35e2bc0b3286","sha256:6327d74c3e51ba1aacfaa0a8efb23e39b8ab81b2d1757ae961c8f3bd5cc6cc58"],"state_sha256":"78da5e9507b474f6619db4b546deed083cf02b2cb09fcd632986d08f9231b5af"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VZVUcHwTQg1ZJAKheB4bf/vzRJjFYbKkiRpMF8Yw4QL/dJ21f2wFbwis6x7XCoze83M5Pt/j639s36eELH0/Dg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-02T06:21:38.578647Z","bundle_sha256":"fdfc4ecf8cdb0753257ee07c01762256d17435f4c945f884338c9753f799659c"}}