{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:Z2RFIHFHKG636IZCDZYONUX4IS","short_pith_number":"pith:Z2RFIHFH","schema_version":"1.0","canonical_sha256":"cea2541ca751bdbf23221e70e6d2fc4491399bc4337d411ce6abe90f048bb885","source":{"kind":"arxiv","id":"2501.00658","version":2},"attestation_state":"computed","paper":{"title":"Understanding and Mitigating Bottlenecks of State Space Models through the Lens of Recency and Over-smoothing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jiajun Zhu, Pan Li, Peihao Wang, Pragya Srivastava, Ruisi Cai, Yuehao Wang, Zhangyang Wang","submitted_at":"2024-12-31T22:06:39Z","abstract_excerpt":"Structured State Space Models (SSMs) have emerged as alternatives to transformers. While SSMs are often regarded as effective in capturing long-sequence dependencies, we rigorously demonstrate that they are inherently limited by strong recency bias. Our empirical studies also reveal that this bias impairs the models' ability to recall distant information and introduces robustness issues. Our scaling experiments then discovered that deeper structures in SSMs can facilitate the learning of long contexts. However, subsequent theoretical analysis reveals that as SSMs increase in depth, they exhibi"},"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":"2501.00658","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-31T22:06:39Z","cross_cats_sorted":[],"title_canon_sha256":"982ffa5a6ef2a8d203cb9b3f521932ed472b56ded48152e696cd9e7404269dda","abstract_canon_sha256":"4a117c9aa63cf0427b9293a47a94cf1fece7eedbfbf05dba9aaf9309230e4c39"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:28:31.530084Z","signature_b64":"FI71hEM+Mwh7nXQrxExLeJbrIfH/++WojhiHVGVGfbZ4ELN/4KUasphsQsz+R/etPJg+iRE4HcFruVNwjlESCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cea2541ca751bdbf23221e70e6d2fc4491399bc4337d411ce6abe90f048bb885","last_reissued_at":"2026-07-05T10:28:31.529569Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:28:31.529569Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Understanding and Mitigating Bottlenecks of State Space Models through the Lens of Recency and Over-smoothing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jiajun Zhu, Pan Li, Peihao Wang, Pragya Srivastava, Ruisi Cai, Yuehao Wang, Zhangyang Wang","submitted_at":"2024-12-31T22:06:39Z","abstract_excerpt":"Structured State Space Models (SSMs) have emerged as alternatives to transformers. While SSMs are often regarded as effective in capturing long-sequence dependencies, we rigorously demonstrate that they are inherently limited by strong recency bias. Our empirical studies also reveal that this bias impairs the models' ability to recall distant information and introduces robustness issues. Our scaling experiments then discovered that deeper structures in SSMs can facilitate the learning of long contexts. However, subsequent theoretical analysis reveals that as SSMs increase in depth, they exhibi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.00658","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/2501.00658/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":"2501.00658","created_at":"2026-07-05T10:28:31.529630+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.00658v2","created_at":"2026-07-05T10:28:31.529630+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.00658","created_at":"2026-07-05T10:28:31.529630+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z2RFIHFHKG63","created_at":"2026-07-05T10:28:31.529630+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z2RFIHFHKG636IZC","created_at":"2026-07-05T10:28:31.529630+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z2RFIHFH","created_at":"2026-07-05T10:28:31.529630+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.06997","citing_title":"Echo: KV-Cache-Free Associative Recall with Spectral Koopman Operators","ref_index":37,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z2RFIHFHKG636IZCDZYONUX4IS","json":"https://pith.science/pith/Z2RFIHFHKG636IZCDZYONUX4IS.json","graph_json":"https://pith.science/api/pith-number/Z2RFIHFHKG636IZCDZYONUX4IS/graph.json","events_json":"https://pith.science/api/pith-number/Z2RFIHFHKG636IZCDZYONUX4IS/events.json","paper":"https://pith.science/paper/Z2RFIHFH"},"agent_actions":{"view_html":"https://pith.science/pith/Z2RFIHFHKG636IZCDZYONUX4IS","download_json":"https://pith.science/pith/Z2RFIHFHKG636IZCDZYONUX4IS.json","view_paper":"https://pith.science/paper/Z2RFIHFH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.00658&json=true","fetch_graph":"https://pith.science/api/pith-number/Z2RFIHFHKG636IZCDZYONUX4IS/graph.json","fetch_events":"https://pith.science/api/pith-number/Z2RFIHFHKG636IZCDZYONUX4IS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z2RFIHFHKG636IZCDZYONUX4IS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z2RFIHFHKG636IZCDZYONUX4IS/action/storage_attestation","attest_author":"https://pith.science/pith/Z2RFIHFHKG636IZCDZYONUX4IS/action/author_attestation","sign_citation":"https://pith.science/pith/Z2RFIHFHKG636IZCDZYONUX4IS/action/citation_signature","submit_replication":"https://pith.science/pith/Z2RFIHFHKG636IZCDZYONUX4IS/action/replication_record"}},"created_at":"2026-07-05T10:28:31.529630+00:00","updated_at":"2026-07-05T10:28:31.529630+00:00"}