{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:ECRKBHSRQLNCLXZGKE3Z3FSAEQ","short_pith_number":"pith:ECRKBHSR","canonical_record":{"source":{"id":"2606.29914","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-29T07:51:22Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1dacf6f07d3df3601abe65e24d613851e2d7310a124f57369d23ba8c37e01525","abstract_canon_sha256":"3fd71b435c85e0decd6dadb527f2e543192ba01cd5e95629c889351f1d2ba06d"},"schema_version":"1.0"},"canonical_sha256":"20a2a09e5182da25df2651379d9640243306cc82288c8ab458968b8dde801daf","source":{"kind":"arxiv","id":"2606.29914","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.29914","created_at":"2026-06-30T02:17:40Z"},{"alias_kind":"arxiv_version","alias_value":"2606.29914v1","created_at":"2026-06-30T02:17:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.29914","created_at":"2026-06-30T02:17:40Z"},{"alias_kind":"pith_short_12","alias_value":"ECRKBHSRQLNC","created_at":"2026-06-30T02:17:40Z"},{"alias_kind":"pith_short_16","alias_value":"ECRKBHSRQLNCLXZG","created_at":"2026-06-30T02:17:40Z"},{"alias_kind":"pith_short_8","alias_value":"ECRKBHSR","created_at":"2026-06-30T02:17:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:ECRKBHSRQLNCLXZGKE3Z3FSAEQ","target":"record","payload":{"canonical_record":{"source":{"id":"2606.29914","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-29T07:51:22Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1dacf6f07d3df3601abe65e24d613851e2d7310a124f57369d23ba8c37e01525","abstract_canon_sha256":"3fd71b435c85e0decd6dadb527f2e543192ba01cd5e95629c889351f1d2ba06d"},"schema_version":"1.0"},"canonical_sha256":"20a2a09e5182da25df2651379d9640243306cc82288c8ab458968b8dde801daf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T02:17:40.772719Z","signature_b64":"ojMkCgz2d58A88T12eDAwrP8dEOBLFqmHloV+MCOJIcPB3eLRfqFW6gBBYUvwuw3ydUm8BMoz7Vyn/SEs5u1Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"20a2a09e5182da25df2651379d9640243306cc82288c8ab458968b8dde801daf","last_reissued_at":"2026-06-30T02:17:40.772132Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T02:17:40.772132Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.29914","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":"2026-06-30T02:17:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mCbUh88x8cSON/3alfC9C7wyP78T/+gIJq+PB7DDY5aFU4g6hB2e6A1BzN3P7MdNcU8ELdVzeQiCoHM3oIPwAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-01T06:47:38.075266Z"},"content_sha256":"0bd986000801811345880fa76cf1d94b67fd039daf68c4662a2ae796e9770730","schema_version":"1.0","event_id":"sha256:0bd986000801811345880fa76cf1d94b67fd039daf68c4662a2ae796e9770730"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:ECRKBHSRQLNCLXZGKE3Z3FSAEQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Kuan Wang","submitted_at":"2026-06-29T07:51:22Z","abstract_excerpt":"Agent memory systems are increasingly evaluated against RAG and full-context baselines, but reported gains often mix changes in the memory method with changes in the language model, embedding model, or retrieval pipeline, making it unclear what is actually being measured. We present MemDelta, a controlled evaluation protocol that varies one component at a time on LongMemEval-S (500 questions, 50+ sessions, three model families). Four findings emerge: (1) verbatim RAG matches full-context GPT-4o-mini (47.2% vs. 49.8%, p = 0.34), but the ranking reverses across models: Gemini gains +14pp from fu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.29914","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/2606.29914/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":"2026-06-30T02:17:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"N+zXPQuBTznMyIolghemE8qlc/dQoNVG8HX4klEcH17PUvp3Ci+pEHP7mAOZU5SVFtsoDpfBozqB4qVhRRz3BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-01T06:47:38.075685Z"},"content_sha256":"8ccfc6fca1f16ae83b7e60a205088ff3d9c24e2cbc78497937220b3ad7159768","schema_version":"1.0","event_id":"sha256:8ccfc6fca1f16ae83b7e60a205088ff3d9c24e2cbc78497937220b3ad7159768"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ECRKBHSRQLNCLXZGKE3Z3FSAEQ/bundle.json","state_url":"https://pith.science/pith/ECRKBHSRQLNCLXZGKE3Z3FSAEQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ECRKBHSRQLNCLXZGKE3Z3FSAEQ/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":"2026-07-01T06:47:38Z","links":{"resolver":"https://pith.science/pith/ECRKBHSRQLNCLXZGKE3Z3FSAEQ","bundle":"https://pith.science/pith/ECRKBHSRQLNCLXZGKE3Z3FSAEQ/bundle.json","state":"https://pith.science/pith/ECRKBHSRQLNCLXZGKE3Z3FSAEQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ECRKBHSRQLNCLXZGKE3Z3FSAEQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ECRKBHSRQLNCLXZGKE3Z3FSAEQ","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":"3fd71b435c85e0decd6dadb527f2e543192ba01cd5e95629c889351f1d2ba06d","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-29T07:51:22Z","title_canon_sha256":"1dacf6f07d3df3601abe65e24d613851e2d7310a124f57369d23ba8c37e01525"},"schema_version":"1.0","source":{"id":"2606.29914","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.29914","created_at":"2026-06-30T02:17:40Z"},{"alias_kind":"arxiv_version","alias_value":"2606.29914v1","created_at":"2026-06-30T02:17:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.29914","created_at":"2026-06-30T02:17:40Z"},{"alias_kind":"pith_short_12","alias_value":"ECRKBHSRQLNC","created_at":"2026-06-30T02:17:40Z"},{"alias_kind":"pith_short_16","alias_value":"ECRKBHSRQLNCLXZG","created_at":"2026-06-30T02:17:40Z"},{"alias_kind":"pith_short_8","alias_value":"ECRKBHSR","created_at":"2026-06-30T02:17:40Z"}],"graph_snapshots":[{"event_id":"sha256:8ccfc6fca1f16ae83b7e60a205088ff3d9c24e2cbc78497937220b3ad7159768","target":"graph","created_at":"2026-06-30T02:17:40Z","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/2606.29914/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Agent memory systems are increasingly evaluated against RAG and full-context baselines, but reported gains often mix changes in the memory method with changes in the language model, embedding model, or retrieval pipeline, making it unclear what is actually being measured. We present MemDelta, a controlled evaluation protocol that varies one component at a time on LongMemEval-S (500 questions, 50+ sessions, three model families). Four findings emerge: (1) verbatim RAG matches full-context GPT-4o-mini (47.2% vs. 49.8%, p = 0.34), but the ranking reverses across models: Gemini gains +14pp from fu","authors_text":"Kuan Wang","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-29T07:51:22Z","title":"MemDelta: Controlled Baselines and Hidden Confounds in Agent Memory Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.29914","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":"sha256:0bd986000801811345880fa76cf1d94b67fd039daf68c4662a2ae796e9770730","target":"record","created_at":"2026-06-30T02:17:40Z","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":"3fd71b435c85e0decd6dadb527f2e543192ba01cd5e95629c889351f1d2ba06d","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-29T07:51:22Z","title_canon_sha256":"1dacf6f07d3df3601abe65e24d613851e2d7310a124f57369d23ba8c37e01525"},"schema_version":"1.0","source":{"id":"2606.29914","kind":"arxiv","version":1}},"canonical_sha256":"20a2a09e5182da25df2651379d9640243306cc82288c8ab458968b8dde801daf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"20a2a09e5182da25df2651379d9640243306cc82288c8ab458968b8dde801daf","first_computed_at":"2026-06-30T02:17:40.772132Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-30T02:17:40.772132Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ojMkCgz2d58A88T12eDAwrP8dEOBLFqmHloV+MCOJIcPB3eLRfqFW6gBBYUvwuw3ydUm8BMoz7Vyn/SEs5u1Bg==","signature_status":"signed_v1","signed_at":"2026-06-30T02:17:40.772719Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.29914","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0bd986000801811345880fa76cf1d94b67fd039daf68c4662a2ae796e9770730","sha256:8ccfc6fca1f16ae83b7e60a205088ff3d9c24e2cbc78497937220b3ad7159768"],"state_sha256":"e546b1d6491794605755318dc4bac5c222d7f8757ec7c2e8daa7077ab44e16af"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Yd1su3kAeA6Oyf6w4EBvZlVHEkP9sWo47hvP//BTUEoMmb2l6yZwRVBehduWOdEoRkB3m4piElKyND9ZJKbZCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-01T06:47:38.078208Z","bundle_sha256":"9de73b91371042de6bf1840ba7d3e3760a95c8073adb57d9786be0aebad5168a"}}