{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:WC2YVJNSA7IF42VOUWDSSF42GJ","short_pith_number":"pith:WC2YVJNS","canonical_record":{"source":{"id":"2406.08747","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-13T02:08:28Z","cross_cats_sorted":[],"title_canon_sha256":"ba86c64a394c5a8a602e4bbf94b3f0ef59ee833003138f9713f58f08331a7eca","abstract_canon_sha256":"8dc305769581aba0680a688c583317df7dd94f51b75d659c5f9c7cb36bf6b9b1"},"schema_version":"1.0"},"canonical_sha256":"b0b58aa5b207d05e6aaea58729179a32772e9b92002eb4a198502ab2e98b2c1b","source":{"kind":"arxiv","id":"2406.08747","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.08747","created_at":"2026-07-05T09:28:51Z"},{"alias_kind":"arxiv_version","alias_value":"2406.08747v2","created_at":"2026-07-05T09:28:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.08747","created_at":"2026-07-05T09:28:51Z"},{"alias_kind":"pith_short_12","alias_value":"WC2YVJNSA7IF","created_at":"2026-07-05T09:28:51Z"},{"alias_kind":"pith_short_16","alias_value":"WC2YVJNSA7IF42VO","created_at":"2026-07-05T09:28:51Z"},{"alias_kind":"pith_short_8","alias_value":"WC2YVJNS","created_at":"2026-07-05T09:28:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:WC2YVJNSA7IF42VOUWDSSF42GJ","target":"record","payload":{"canonical_record":{"source":{"id":"2406.08747","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-13T02:08:28Z","cross_cats_sorted":[],"title_canon_sha256":"ba86c64a394c5a8a602e4bbf94b3f0ef59ee833003138f9713f58f08331a7eca","abstract_canon_sha256":"8dc305769581aba0680a688c583317df7dd94f51b75d659c5f9c7cb36bf6b9b1"},"schema_version":"1.0"},"canonical_sha256":"b0b58aa5b207d05e6aaea58729179a32772e9b92002eb4a198502ab2e98b2c1b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:28:51.260474Z","signature_b64":"XBkinI4ygBqkAijUqmLN0PVcCL25yU505/E7althDjkHNR8rtwpqR1ytnypBsGkml76AcNLkFvU+yCqeGJ4TAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b0b58aa5b207d05e6aaea58729179a32772e9b92002eb4a198502ab2e98b2c1b","last_reissued_at":"2026-07-05T09:28:51.260000Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:28:51.260000Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.08747","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":"2026-07-05T09:28:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sTg7bsp0MW+LgyxBDQjkYzYOAQyHjzwYelWdeelOa12dONwftNLbH7dorUHi6QvEsahs2I+9y4gONpNxURDNDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T16:13:24.128244Z"},"content_sha256":"b0c8a0c9416000119977bdea42b3e888baf1ccf05675bacb243682576832463b","schema_version":"1.0","event_id":"sha256:b0c8a0c9416000119977bdea42b3e888baf1ccf05675bacb243682576832463b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:WC2YVJNSA7IF42VOUWDSSF42GJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"StreamBench: Towards Benchmarking Continuous Improvement of Language Agents","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cheng-Kuang Wu, Chieh-Yen Lin, Hung-yi Lee, Yun-Nung Chen, Zhi Rui Tam","submitted_at":"2024-06-13T02:08:28Z","abstract_excerpt":"Recent works have shown that large language model (LLM) agents are able to improve themselves from experience, which is an important ability for continuous enhancement post-deployment. However, existing benchmarks primarily evaluate their innate capabilities and do not assess their ability to improve over time. To address this gap, we introduce StreamBench, a pioneering benchmark designed to evaluate the continuous improvement of LLM agents over an input-feedback sequence. StreamBench simulates an online learning environment where LLMs receive a continuous flow of feedback stream and iterative"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.08747","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/2406.08747/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":"2026-07-05T09:28:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"49wU8NBI90mePP/TTehBLJTVtjqNwhdr+ckqSmKdOVaIQhtBNiptl7mL5N9ERUIJeuhnaxRCmnU/s8zaF3ftAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T16:13:24.128799Z"},"content_sha256":"ed96615921978cd6d35086b076e7d457a18cc07001f500c384d1ad9792859e34","schema_version":"1.0","event_id":"sha256:ed96615921978cd6d35086b076e7d457a18cc07001f500c384d1ad9792859e34"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WC2YVJNSA7IF42VOUWDSSF42GJ/bundle.json","state_url":"https://pith.science/pith/WC2YVJNSA7IF42VOUWDSSF42GJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WC2YVJNSA7IF42VOUWDSSF42GJ/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":"2026-08-08T16:13:24Z","links":{"resolver":"https://pith.science/pith/WC2YVJNSA7IF42VOUWDSSF42GJ","bundle":"https://pith.science/pith/WC2YVJNSA7IF42VOUWDSSF42GJ/bundle.json","state":"https://pith.science/pith/WC2YVJNSA7IF42VOUWDSSF42GJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WC2YVJNSA7IF42VOUWDSSF42GJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:WC2YVJNSA7IF42VOUWDSSF42GJ","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":"8dc305769581aba0680a688c583317df7dd94f51b75d659c5f9c7cb36bf6b9b1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-13T02:08:28Z","title_canon_sha256":"ba86c64a394c5a8a602e4bbf94b3f0ef59ee833003138f9713f58f08331a7eca"},"schema_version":"1.0","source":{"id":"2406.08747","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.08747","created_at":"2026-07-05T09:28:51Z"},{"alias_kind":"arxiv_version","alias_value":"2406.08747v2","created_at":"2026-07-05T09:28:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.08747","created_at":"2026-07-05T09:28:51Z"},{"alias_kind":"pith_short_12","alias_value":"WC2YVJNSA7IF","created_at":"2026-07-05T09:28:51Z"},{"alias_kind":"pith_short_16","alias_value":"WC2YVJNSA7IF42VO","created_at":"2026-07-05T09:28:51Z"},{"alias_kind":"pith_short_8","alias_value":"WC2YVJNS","created_at":"2026-07-05T09:28:51Z"}],"graph_snapshots":[{"event_id":"sha256:ed96615921978cd6d35086b076e7d457a18cc07001f500c384d1ad9792859e34","target":"graph","created_at":"2026-07-05T09:28:51Z","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/2406.08747/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent works have shown that large language model (LLM) agents are able to improve themselves from experience, which is an important ability for continuous enhancement post-deployment. However, existing benchmarks primarily evaluate their innate capabilities and do not assess their ability to improve over time. To address this gap, we introduce StreamBench, a pioneering benchmark designed to evaluate the continuous improvement of LLM agents over an input-feedback sequence. StreamBench simulates an online learning environment where LLMs receive a continuous flow of feedback stream and iterative","authors_text":"Cheng-Kuang Wu, Chieh-Yen Lin, Hung-yi Lee, Yun-Nung Chen, Zhi Rui Tam","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-13T02:08:28Z","title":"StreamBench: Towards Benchmarking Continuous Improvement of Language Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.08747","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:b0c8a0c9416000119977bdea42b3e888baf1ccf05675bacb243682576832463b","target":"record","created_at":"2026-07-05T09:28:51Z","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":"8dc305769581aba0680a688c583317df7dd94f51b75d659c5f9c7cb36bf6b9b1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-13T02:08:28Z","title_canon_sha256":"ba86c64a394c5a8a602e4bbf94b3f0ef59ee833003138f9713f58f08331a7eca"},"schema_version":"1.0","source":{"id":"2406.08747","kind":"arxiv","version":2}},"canonical_sha256":"b0b58aa5b207d05e6aaea58729179a32772e9b92002eb4a198502ab2e98b2c1b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b0b58aa5b207d05e6aaea58729179a32772e9b92002eb4a198502ab2e98b2c1b","first_computed_at":"2026-07-05T09:28:51.260000Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:28:51.260000Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XBkinI4ygBqkAijUqmLN0PVcCL25yU505/E7althDjkHNR8rtwpqR1ytnypBsGkml76AcNLkFvU+yCqeGJ4TAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:28:51.260474Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.08747","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b0c8a0c9416000119977bdea42b3e888baf1ccf05675bacb243682576832463b","sha256:ed96615921978cd6d35086b076e7d457a18cc07001f500c384d1ad9792859e34"],"state_sha256":"6f7dff9fbbb5d891f761a50cda6791bf0805b759d9d2b5091b857521756f72d3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L609Hr5DZqsocgx8gNaKm9RJq8mzbYKi0F70ooGnWuN5GuIriQb6HsAx3abWy8eqbnZoN5mh6aCmoPemy+xkDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T16:13:24.134385Z","bundle_sha256":"b5e2b72841dbda7745ecc5f92d32e9ad66738d94fa5ea74209461ed774cddb78"}}