{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:G62PEFGYPKVKCJ7XRKJSMRO3B6","short_pith_number":"pith:G62PEFGY","schema_version":"1.0","canonical_sha256":"37b4f214d87aaaa127f78a932645db0f82f5581c75b2555df03cc8c280f3ce1b","source":{"kind":"arxiv","id":"2606.21562","version":1},"attestation_state":"computed","paper":{"title":"Compressing Observation History into Agent Memory: Distilling Transformers into Recurrent Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"B\\\"ulent Mert Sariyildiz, Christian Wolf, Gianluca Monaci, Guillaume Bono, Philippe Weinzaepfel","submitted_at":"2026-06-19T15:58:36Z","abstract_excerpt":"Transformers are AI's workhorse with strong performance in modeling sequential data, but their computational cost becomes prohibitive when processing long sequences. We target long-horizon streaming vision and robotics applications like map-free pose estimation, where it is particularly impractical to store and maintain a history of observations. Recurrent Transformers address this limitation by maintaining fixed-size memory but their performance lags behind that of transformers operating over the full observation history. We argue that this gap does not stem from architectural limitations, bu"},"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":"2606.21562","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-06-19T15:58:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"66cdbe59aacbff093915bba4ed0dbf3f3d0ee14d669dabe0f24d94c8819405fa","abstract_canon_sha256":"3f8da4dba115560c95dbdb9f99143a0170de5df56e1ff0f419f0a6fb622878c2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:13:14.707000Z","signature_b64":"Fyf7uHbkzivohZogjpoBC+cgksiHYvcM61HqngVEqz+Qk2S8EnBbVje2obLiNTJ30Tfe/ROtvbhqPpHyUSXaCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"37b4f214d87aaaa127f78a932645db0f82f5581c75b2555df03cc8c280f3ce1b","last_reissued_at":"2026-06-23T01:13:14.706516Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:13:14.706516Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Compressing Observation History into Agent Memory: Distilling Transformers into Recurrent Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"B\\\"ulent Mert Sariyildiz, Christian Wolf, Gianluca Monaci, Guillaume Bono, Philippe Weinzaepfel","submitted_at":"2026-06-19T15:58:36Z","abstract_excerpt":"Transformers are AI's workhorse with strong performance in modeling sequential data, but their computational cost becomes prohibitive when processing long sequences. We target long-horizon streaming vision and robotics applications like map-free pose estimation, where it is particularly impractical to store and maintain a history of observations. Recurrent Transformers address this limitation by maintaining fixed-size memory but their performance lags behind that of transformers operating over the full observation history. We argue that this gap does not stem from architectural limitations, bu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.21562","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/2606.21562/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":"2606.21562","created_at":"2026-06-23T01:13:14.706597+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.21562v1","created_at":"2026-06-23T01:13:14.706597+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.21562","created_at":"2026-06-23T01:13:14.706597+00:00"},{"alias_kind":"pith_short_12","alias_value":"G62PEFGYPKVK","created_at":"2026-06-23T01:13:14.706597+00:00"},{"alias_kind":"pith_short_16","alias_value":"G62PEFGYPKVKCJ7X","created_at":"2026-06-23T01:13:14.706597+00:00"},{"alias_kind":"pith_short_8","alias_value":"G62PEFGY","created_at":"2026-06-23T01:13:14.706597+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G62PEFGYPKVKCJ7XRKJSMRO3B6","json":"https://pith.science/pith/G62PEFGYPKVKCJ7XRKJSMRO3B6.json","graph_json":"https://pith.science/api/pith-number/G62PEFGYPKVKCJ7XRKJSMRO3B6/graph.json","events_json":"https://pith.science/api/pith-number/G62PEFGYPKVKCJ7XRKJSMRO3B6/events.json","paper":"https://pith.science/paper/G62PEFGY"},"agent_actions":{"view_html":"https://pith.science/pith/G62PEFGYPKVKCJ7XRKJSMRO3B6","download_json":"https://pith.science/pith/G62PEFGYPKVKCJ7XRKJSMRO3B6.json","view_paper":"https://pith.science/paper/G62PEFGY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.21562&json=true","fetch_graph":"https://pith.science/api/pith-number/G62PEFGYPKVKCJ7XRKJSMRO3B6/graph.json","fetch_events":"https://pith.science/api/pith-number/G62PEFGYPKVKCJ7XRKJSMRO3B6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G62PEFGYPKVKCJ7XRKJSMRO3B6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G62PEFGYPKVKCJ7XRKJSMRO3B6/action/storage_attestation","attest_author":"https://pith.science/pith/G62PEFGYPKVKCJ7XRKJSMRO3B6/action/author_attestation","sign_citation":"https://pith.science/pith/G62PEFGYPKVKCJ7XRKJSMRO3B6/action/citation_signature","submit_replication":"https://pith.science/pith/G62PEFGYPKVKCJ7XRKJSMRO3B6/action/replication_record"}},"created_at":"2026-06-23T01:13:14.706597+00:00","updated_at":"2026-06-23T01:13:14.706597+00:00"}