{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OV6P2JTV7LJPILFH6UXU4QI52A","short_pith_number":"pith:OV6P2JTV","schema_version":"1.0","canonical_sha256":"757cfd2675fad2f42ca7f52f4e411dd01459177b3dd47ff7c13db0f7d27b9898","source":{"kind":"arxiv","id":"2604.07035","version":2},"attestation_state":"computed","paper":{"title":"Unified Deployment-Aware Evaluation of Open Reasoning Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Accuracy-efficiency tradeoffs in reasoning LLMs depend jointly on architecture, prompting protocol, and task rather than sparse activation alone.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ge Wang, Md Motaleb Hossen Manik","submitted_at":"2026-04-08T12:50:52Z","abstract_excerpt":"Open reasoning language models are often compared under mixed sample sizes, partially standardized prompts, and accuracy-centered summaries, which makes practical model selection difficult to interpret. We present a unified evaluation of seven open reasoning language model configurations across four benchmarks: ARC-Challenge, GSM8K, MATH levels 1 to 3, and TruthfulQA MC1. We test zero-shot, chain-of-thought (CoT), and few-shot CoT prompting on the same 238-example subset for every model--dataset--strategy condition, yielding a complete 7 x 4 x 3 design with 84 conditions and 19,992 evaluated e"},"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":"2604.07035","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-08T12:50:52Z","cross_cats_sorted":[],"title_canon_sha256":"8ca16ca16ac778a2a73492f3ad4f744c08b705095d09c9592cb55b11a4a06d3b","abstract_canon_sha256":"e6c8d5e5a812be39a8d8a0da79f389eafe196787038db3dc44cf29431659897d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:12.592716Z","signature_b64":"ifJ7NWMpNo+OzgY0LXYmlYq5/s7cKNxOvbW2I36LkQFhDN2cfOmdJdabSLEs8ILzDXZmYoaOSt5jHdrGfuGjAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"757cfd2675fad2f42ca7f52f4e411dd01459177b3dd47ff7c13db0f7d27b9898","last_reissued_at":"2026-05-20T01:05:12.591934Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:12.591934Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unified Deployment-Aware Evaluation of Open Reasoning Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Accuracy-efficiency tradeoffs in reasoning LLMs depend jointly on architecture, prompting protocol, and task rather than sparse activation alone.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ge Wang, Md Motaleb Hossen Manik","submitted_at":"2026-04-08T12:50:52Z","abstract_excerpt":"Open reasoning language models are often compared under mixed sample sizes, partially standardized prompts, and accuracy-centered summaries, which makes practical model selection difficult to interpret. We present a unified evaluation of seven open reasoning language model configurations across four benchmarks: ARC-Challenge, GSM8K, MATH levels 1 to 3, and TruthfulQA MC1. We test zero-shot, chain-of-thought (CoT), and few-shot CoT prompting on the same 238-example subset for every model--dataset--strategy condition, yielding a complete 7 x 4 x 3 design with 84 conditions and 19,992 evaluated e"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"These results show that sparse activation alone does not guarantee the best practical operating point: observed accuracy-efficiency tradeoffs depend jointly on architecture, prompting protocol, and task composition.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the four chosen benchmarks and three prompting strategies are representative enough of real-world reasoning workloads to support general claims about accuracy-efficiency tradeoffs.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Gemma-4-E4B with few-shot chain-of-thought reaches the highest weighted accuracy of 0.675 at 14.9 GB VRAM, while the larger Gemma-4-26B-A4B MoE model scores 0.663 but uses 48.1 GB.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Accuracy-efficiency tradeoffs in reasoning LLMs depend jointly on architecture, prompting protocol, and task rather than sparse activation alone.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"455f26886c7e9c24405dc6418b419b58d2eafe09b327f5aac3e5575be5729663"},"source":{"id":"2604.07035","kind":"arxiv","version":2},"verdict":{"id":"93e4922e-cb2b-4db0-bdc4-28d92264b317","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T17:44:59.556352Z","strongest_claim":"These results show that sparse activation alone does not guarantee the best practical operating point: observed accuracy-efficiency tradeoffs depend jointly on architecture, prompting protocol, and task composition.","one_line_summary":"Gemma-4-E4B with few-shot chain-of-thought reaches the highest weighted accuracy of 0.675 at 14.9 GB VRAM, while the larger Gemma-4-26B-A4B MoE model scores 0.663 but uses 48.1 GB.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the four chosen benchmarks and three prompting strategies are representative enough of real-world reasoning workloads to support general claims about accuracy-efficiency tradeoffs.","pith_extraction_headline":"Accuracy-efficiency tradeoffs in reasoning LLMs depend jointly on architecture, prompting protocol, and task rather than sparse activation alone."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.07035/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":"2604.07035","created_at":"2026-05-20T01:05:12.592058+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.07035v2","created_at":"2026-05-20T01:05:12.592058+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.07035","created_at":"2026-05-20T01:05:12.592058+00:00"},{"alias_kind":"pith_short_12","alias_value":"OV6P2JTV7LJP","created_at":"2026-05-20T01:05:12.592058+00:00"},{"alias_kind":"pith_short_16","alias_value":"OV6P2JTV7LJPILFH","created_at":"2026-05-20T01:05:12.592058+00:00"},{"alias_kind":"pith_short_8","alias_value":"OV6P2JTV","created_at":"2026-05-20T01:05:12.592058+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2604.26985","citing_title":"Simple Self-Conditioning Adaptation for Masked Diffusion Models","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22601","citing_title":"From Natural Language to Verified Code: Toward AI Assisted Problem-to-Code Generation with Dafny-Based Formal Verification","ref_index":57,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OV6P2JTV7LJPILFH6UXU4QI52A","json":"https://pith.science/pith/OV6P2JTV7LJPILFH6UXU4QI52A.json","graph_json":"https://pith.science/api/pith-number/OV6P2JTV7LJPILFH6UXU4QI52A/graph.json","events_json":"https://pith.science/api/pith-number/OV6P2JTV7LJPILFH6UXU4QI52A/events.json","paper":"https://pith.science/paper/OV6P2JTV"},"agent_actions":{"view_html":"https://pith.science/pith/OV6P2JTV7LJPILFH6UXU4QI52A","download_json":"https://pith.science/pith/OV6P2JTV7LJPILFH6UXU4QI52A.json","view_paper":"https://pith.science/paper/OV6P2JTV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.07035&json=true","fetch_graph":"https://pith.science/api/pith-number/OV6P2JTV7LJPILFH6UXU4QI52A/graph.json","fetch_events":"https://pith.science/api/pith-number/OV6P2JTV7LJPILFH6UXU4QI52A/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OV6P2JTV7LJPILFH6UXU4QI52A/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OV6P2JTV7LJPILFH6UXU4QI52A/action/storage_attestation","attest_author":"https://pith.science/pith/OV6P2JTV7LJPILFH6UXU4QI52A/action/author_attestation","sign_citation":"https://pith.science/pith/OV6P2JTV7LJPILFH6UXU4QI52A/action/citation_signature","submit_replication":"https://pith.science/pith/OV6P2JTV7LJPILFH6UXU4QI52A/action/replication_record"}},"created_at":"2026-05-20T01:05:12.592058+00:00","updated_at":"2026-05-20T01:05:12.592058+00:00"}