{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:AEMCUIRODLLNGTJNXYLIVYKZDI","short_pith_number":"pith:AEMCUIRO","schema_version":"1.0","canonical_sha256":"01182a222e1ad6d34d2dbe168ae1591a3a54ac47815646feacbab87d95daaa3f","source":{"kind":"arxiv","id":"2502.08180","version":2},"attestation_state":"computed","paper":{"title":"Enhancing LLM Character-Level Manipulation via Divide and Conquer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bryan Hooi, Nanyun Peng, Yiwei Wang, Yujun Cai, Zhecheng Li, Zhen Xiong","submitted_at":"2025-02-12T07:37:39Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated strong generalization capabilities across a wide range of natural language processing (NLP) tasks. However, they exhibit notable weaknesses in character-level string manipulation, struggling with fundamental operations such as character deletion, insertion, and substitution. These challenges stem primarily from tokenization constraints, despite the critical role of such operations in data preprocessing and code generation. Through systematic analysis, we derive two key insights: (1) LLMs face significant difficulties in leveraging intrinsic token "},"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":"2502.08180","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-12T07:37:39Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"30d21783f836868ccb66477bdf99db5e2c22066bbcd272ad62c5e35f56fe3f3f","abstract_canon_sha256":"5dbf7ce631a8a4e10ae5eeca89d4cb6e141c9632fce0ea4e580ddda4f8cc1b7d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:40:16.836365Z","signature_b64":"dPfzfUsi3Sif9zjPBhVAlh1obBHdAi7ZFAg9zsOuedZWCsQyelZ2nCicYRlzOG5fLuks8IQ/ufQG5aHS6MqCAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"01182a222e1ad6d34d2dbe168ae1591a3a54ac47815646feacbab87d95daaa3f","last_reissued_at":"2026-07-05T10:40:16.835883Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:40:16.835883Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing LLM Character-Level Manipulation via Divide and Conquer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bryan Hooi, Nanyun Peng, Yiwei Wang, Yujun Cai, Zhecheng Li, Zhen Xiong","submitted_at":"2025-02-12T07:37:39Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated strong generalization capabilities across a wide range of natural language processing (NLP) tasks. However, they exhibit notable weaknesses in character-level string manipulation, struggling with fundamental operations such as character deletion, insertion, and substitution. These challenges stem primarily from tokenization constraints, despite the critical role of such operations in data preprocessing and code generation. Through systematic analysis, we derive two key insights: (1) LLMs face significant difficulties in leveraging intrinsic token "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.08180","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/2502.08180/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":"2502.08180","created_at":"2026-07-05T10:40:16.835939+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.08180v2","created_at":"2026-07-05T10:40:16.835939+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.08180","created_at":"2026-07-05T10:40:16.835939+00:00"},{"alias_kind":"pith_short_12","alias_value":"AEMCUIRODLLN","created_at":"2026-07-05T10:40:16.835939+00:00"},{"alias_kind":"pith_short_16","alias_value":"AEMCUIRODLLNGTJN","created_at":"2026-07-05T10:40:16.835939+00:00"},{"alias_kind":"pith_short_8","alias_value":"AEMCUIRO","created_at":"2026-07-05T10:40:16.835939+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08317","citing_title":"Blind-Spots-Bench: Evaluating Blind Spots in Multimodal Models","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2606.06840","citing_title":"Characterize Then Distill: Mechanistic Reasoning in Large Output Spaces","ref_index":146,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AEMCUIRODLLNGTJNXYLIVYKZDI","json":"https://pith.science/pith/AEMCUIRODLLNGTJNXYLIVYKZDI.json","graph_json":"https://pith.science/api/pith-number/AEMCUIRODLLNGTJNXYLIVYKZDI/graph.json","events_json":"https://pith.science/api/pith-number/AEMCUIRODLLNGTJNXYLIVYKZDI/events.json","paper":"https://pith.science/paper/AEMCUIRO"},"agent_actions":{"view_html":"https://pith.science/pith/AEMCUIRODLLNGTJNXYLIVYKZDI","download_json":"https://pith.science/pith/AEMCUIRODLLNGTJNXYLIVYKZDI.json","view_paper":"https://pith.science/paper/AEMCUIRO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.08180&json=true","fetch_graph":"https://pith.science/api/pith-number/AEMCUIRODLLNGTJNXYLIVYKZDI/graph.json","fetch_events":"https://pith.science/api/pith-number/AEMCUIRODLLNGTJNXYLIVYKZDI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AEMCUIRODLLNGTJNXYLIVYKZDI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AEMCUIRODLLNGTJNXYLIVYKZDI/action/storage_attestation","attest_author":"https://pith.science/pith/AEMCUIRODLLNGTJNXYLIVYKZDI/action/author_attestation","sign_citation":"https://pith.science/pith/AEMCUIRODLLNGTJNXYLIVYKZDI/action/citation_signature","submit_replication":"https://pith.science/pith/AEMCUIRODLLNGTJNXYLIVYKZDI/action/replication_record"}},"created_at":"2026-07-05T10:40:16.835939+00:00","updated_at":"2026-07-05T10:40:16.835939+00:00"}