{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YY6JZTQLN2VHEINS4SW555J777","short_pith_number":"pith:YY6JZTQL","schema_version":"1.0","canonical_sha256":"c63c9cce0b6eaa7221b2e4addef53fffd687d63b59bf3fa6f6c00b54421f4d06","source":{"kind":"arxiv","id":"2605.19775","version":1},"attestation_state":"computed","paper":{"title":"Understanding Inference Scaling for LLMs: Bottlenecks, Trade-offs, and Performance Principles","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.PF"],"primary_cat":"cs.DC","authors_text":"Avinash Maurya, Bogdan Nicolae, Moiz Arif, Sudharshan Vazhkudai","submitted_at":"2026-05-19T12:43:51Z","abstract_excerpt":"The transition from standard generative AI to \\emph{reasoning-centric architectures}, exemplified by models capable of extensive Chain-of-Thought~(CoT) processing, marks a fundamental paradigm shift in system requirements. Unlike traditional workloads dominated by compute-bound prefill, reasoning workloads generate long chains of reasoning tokens that shift inference into a \\emph{Capacity-Bound regime}. This paper presents a comprehensive system characterization, evaluating models ranging from 8B to 671B parameters on GPUs clusters. By systematically exploring the interplay between Data, Tenso"},"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":"2605.19775","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-19T12:43:51Z","cross_cats_sorted":["cs.PF"],"title_canon_sha256":"fbe4aafa61e817c1d24c7528589c37cbafc464ea5ee995fa9638bee5c122d692","abstract_canon_sha256":"cd7297cdec4f20f63b29b9e8c6a69b1a7b2608de49df895201eb2493a9a69258"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:06:13.165186Z","signature_b64":"uX3fC61FfW4m5oKFwZ8yggVfjZAgd+iSB3m31Kz8Y2NIHG5ev33KFriTwCR6ZG2fbjIlZ1SHHL+JFJd22s2IBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c63c9cce0b6eaa7221b2e4addef53fffd687d63b59bf3fa6f6c00b54421f4d06","last_reissued_at":"2026-05-20T01:06:13.164388Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:06:13.164388Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Understanding Inference Scaling for LLMs: Bottlenecks, Trade-offs, and Performance Principles","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.PF"],"primary_cat":"cs.DC","authors_text":"Avinash Maurya, Bogdan Nicolae, Moiz Arif, Sudharshan Vazhkudai","submitted_at":"2026-05-19T12:43:51Z","abstract_excerpt":"The transition from standard generative AI to \\emph{reasoning-centric architectures}, exemplified by models capable of extensive Chain-of-Thought~(CoT) processing, marks a fundamental paradigm shift in system requirements. Unlike traditional workloads dominated by compute-bound prefill, reasoning workloads generate long chains of reasoning tokens that shift inference into a \\emph{Capacity-Bound regime}. This paper presents a comprehensive system characterization, evaluating models ranging from 8B to 671B parameters on GPUs clusters. By systematically exploring the interplay between Data, Tenso"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19775","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/2605.19775/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":"2605.19775","created_at":"2026-05-20T01:06:13.164525+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.19775v1","created_at":"2026-05-20T01:06:13.164525+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19775","created_at":"2026-05-20T01:06:13.164525+00:00"},{"alias_kind":"pith_short_12","alias_value":"YY6JZTQLN2VH","created_at":"2026-05-20T01:06:13.164525+00:00"},{"alias_kind":"pith_short_16","alias_value":"YY6JZTQLN2VHEINS","created_at":"2026-05-20T01:06:13.164525+00:00"},{"alias_kind":"pith_short_8","alias_value":"YY6JZTQL","created_at":"2026-05-20T01:06:13.164525+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/YY6JZTQLN2VHEINS4SW555J777","json":"https://pith.science/pith/YY6JZTQLN2VHEINS4SW555J777.json","graph_json":"https://pith.science/api/pith-number/YY6JZTQLN2VHEINS4SW555J777/graph.json","events_json":"https://pith.science/api/pith-number/YY6JZTQLN2VHEINS4SW555J777/events.json","paper":"https://pith.science/paper/YY6JZTQL"},"agent_actions":{"view_html":"https://pith.science/pith/YY6JZTQLN2VHEINS4SW555J777","download_json":"https://pith.science/pith/YY6JZTQLN2VHEINS4SW555J777.json","view_paper":"https://pith.science/paper/YY6JZTQL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.19775&json=true","fetch_graph":"https://pith.science/api/pith-number/YY6JZTQLN2VHEINS4SW555J777/graph.json","fetch_events":"https://pith.science/api/pith-number/YY6JZTQLN2VHEINS4SW555J777/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YY6JZTQLN2VHEINS4SW555J777/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YY6JZTQLN2VHEINS4SW555J777/action/storage_attestation","attest_author":"https://pith.science/pith/YY6JZTQLN2VHEINS4SW555J777/action/author_attestation","sign_citation":"https://pith.science/pith/YY6JZTQLN2VHEINS4SW555J777/action/citation_signature","submit_replication":"https://pith.science/pith/YY6JZTQLN2VHEINS4SW555J777/action/replication_record"}},"created_at":"2026-05-20T01:06:13.164525+00:00","updated_at":"2026-05-20T01:06:13.164525+00:00"}