{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:4DFQ6YAJC3MMGBBOMNPPJCALBH","short_pith_number":"pith:4DFQ6YAJ","canonical_record":{"source":{"id":"2406.06461","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-10T16:55:08Z","cross_cats_sorted":[],"title_canon_sha256":"7b8712e72ea95d5e6cde497a8128049ba10bc00ce7e82731d950ebd380980b53","abstract_canon_sha256":"c2f0a0b3025097bcf2fdf2dc08f420a60db8c216c28ab9e82ae294f237b68e6a"},"schema_version":"1.0"},"canonical_sha256":"e0cb0f600916d8c3042e635ef4880b09fe1e6af2bad1106c09542075e5991034","source":{"kind":"arxiv","id":"2406.06461","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.06461","created_at":"2026-07-05T08:32:13Z"},{"alias_kind":"arxiv_version","alias_value":"2406.06461v3","created_at":"2026-07-05T08:32:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.06461","created_at":"2026-07-05T08:32:13Z"},{"alias_kind":"pith_short_12","alias_value":"4DFQ6YAJC3MM","created_at":"2026-07-05T08:32:13Z"},{"alias_kind":"pith_short_16","alias_value":"4DFQ6YAJC3MMGBBO","created_at":"2026-07-05T08:32:13Z"},{"alias_kind":"pith_short_8","alias_value":"4DFQ6YAJ","created_at":"2026-07-05T08:32:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:4DFQ6YAJC3MMGBBOMNPPJCALBH","target":"record","payload":{"canonical_record":{"source":{"id":"2406.06461","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-10T16:55:08Z","cross_cats_sorted":[],"title_canon_sha256":"7b8712e72ea95d5e6cde497a8128049ba10bc00ce7e82731d950ebd380980b53","abstract_canon_sha256":"c2f0a0b3025097bcf2fdf2dc08f420a60db8c216c28ab9e82ae294f237b68e6a"},"schema_version":"1.0"},"canonical_sha256":"e0cb0f600916d8c3042e635ef4880b09fe1e6af2bad1106c09542075e5991034","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:32:13.167182Z","signature_b64":"bErfyqUPNx8OwZ3qwml09G0SMQezP9pvXlRWQvgwUM/bGMuyT6neihdooXVO+GV583yUJ3NVctjJTwOvL0kzAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e0cb0f600916d8c3042e635ef4880b09fe1e6af2bad1106c09542075e5991034","last_reissued_at":"2026-07-05T08:32:13.166678Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:32:13.166678Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.06461","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:32:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"f4r63wJazieAXFB3MTU5HIAvcoW9WhaMj+BmTlnLLvg7HG78Il2L3fGJTfMGXQ/yc+47RI2gbvVrDN6/023ICA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T05:47:50.981953Z"},"content_sha256":"362f3e88026bccb907fd8ab26272320cdd7c3a7d4101d5edcac7ee3fff3696ad","schema_version":"1.0","event_id":"sha256:362f3e88026bccb907fd8ab26272320cdd7c3a7d4101d5edcac7ee3fff3696ad"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:4DFQ6YAJC3MMGBBOMNPPJCALBH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reasoning in Token Economies: Budget-Aware Evaluation of LLM Reasoning Strategies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Baishakhi Ray, Ben Athiwaratkun, Dejiao Zhang, Junlin Wang, Siddhartha Jain, Varun Kumar","submitted_at":"2024-06-10T16:55:08Z","abstract_excerpt":"A diverse array of reasoning strategies has been proposed to elicit the capabilities of large language models. However, in this paper, we point out that traditional evaluations which focus solely on performance metrics miss a key factor: the increased effectiveness due to additional compute. By overlooking this aspect, a skewed view of strategy efficiency is often presented. This paper introduces a framework that incorporates the compute budget into the evaluation, providing a more informative comparison that takes into account both performance metrics and computational cost. In this budget-aw"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.06461","kind":"arxiv","version":3},"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/2406.06461/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:32:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oCZdRCnd8xu8WGpIwq3GvBlwsDo3MDBBkz1en3nljvj0JVXbxDaWAjTJnye1NksuZQ8IpnKyN2tXM98iFVbaCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T05:47:50.982341Z"},"content_sha256":"ea29accb3cae54c173d2533982a44cab9b44022bf1714ea8029263e8d7d3e2f9","schema_version":"1.0","event_id":"sha256:ea29accb3cae54c173d2533982a44cab9b44022bf1714ea8029263e8d7d3e2f9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4DFQ6YAJC3MMGBBOMNPPJCALBH/bundle.json","state_url":"https://pith.science/pith/4DFQ6YAJC3MMGBBOMNPPJCALBH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4DFQ6YAJC3MMGBBOMNPPJCALBH/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-07T05:47:50Z","links":{"resolver":"https://pith.science/pith/4DFQ6YAJC3MMGBBOMNPPJCALBH","bundle":"https://pith.science/pith/4DFQ6YAJC3MMGBBOMNPPJCALBH/bundle.json","state":"https://pith.science/pith/4DFQ6YAJC3MMGBBOMNPPJCALBH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4DFQ6YAJC3MMGBBOMNPPJCALBH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:4DFQ6YAJC3MMGBBOMNPPJCALBH","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"c2f0a0b3025097bcf2fdf2dc08f420a60db8c216c28ab9e82ae294f237b68e6a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-10T16:55:08Z","title_canon_sha256":"7b8712e72ea95d5e6cde497a8128049ba10bc00ce7e82731d950ebd380980b53"},"schema_version":"1.0","source":{"id":"2406.06461","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.06461","created_at":"2026-07-05T08:32:13Z"},{"alias_kind":"arxiv_version","alias_value":"2406.06461v3","created_at":"2026-07-05T08:32:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.06461","created_at":"2026-07-05T08:32:13Z"},{"alias_kind":"pith_short_12","alias_value":"4DFQ6YAJC3MM","created_at":"2026-07-05T08:32:13Z"},{"alias_kind":"pith_short_16","alias_value":"4DFQ6YAJC3MMGBBO","created_at":"2026-07-05T08:32:13Z"},{"alias_kind":"pith_short_8","alias_value":"4DFQ6YAJ","created_at":"2026-07-05T08:32:13Z"}],"graph_snapshots":[{"event_id":"sha256:ea29accb3cae54c173d2533982a44cab9b44022bf1714ea8029263e8d7d3e2f9","target":"graph","created_at":"2026-07-05T08:32:13Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2406.06461/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"A diverse array of reasoning strategies has been proposed to elicit the capabilities of large language models. However, in this paper, we point out that traditional evaluations which focus solely on performance metrics miss a key factor: the increased effectiveness due to additional compute. By overlooking this aspect, a skewed view of strategy efficiency is often presented. This paper introduces a framework that incorporates the compute budget into the evaluation, providing a more informative comparison that takes into account both performance metrics and computational cost. In this budget-aw","authors_text":"Baishakhi Ray, Ben Athiwaratkun, Dejiao Zhang, Junlin Wang, Siddhartha Jain, Varun Kumar","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-10T16:55:08Z","title":"Reasoning in Token Economies: Budget-Aware Evaluation of LLM Reasoning Strategies"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.06461","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:362f3e88026bccb907fd8ab26272320cdd7c3a7d4101d5edcac7ee3fff3696ad","target":"record","created_at":"2026-07-05T08:32:13Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"c2f0a0b3025097bcf2fdf2dc08f420a60db8c216c28ab9e82ae294f237b68e6a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-10T16:55:08Z","title_canon_sha256":"7b8712e72ea95d5e6cde497a8128049ba10bc00ce7e82731d950ebd380980b53"},"schema_version":"1.0","source":{"id":"2406.06461","kind":"arxiv","version":3}},"canonical_sha256":"e0cb0f600916d8c3042e635ef4880b09fe1e6af2bad1106c09542075e5991034","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e0cb0f600916d8c3042e635ef4880b09fe1e6af2bad1106c09542075e5991034","first_computed_at":"2026-07-05T08:32:13.166678Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:32:13.166678Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bErfyqUPNx8OwZ3qwml09G0SMQezP9pvXlRWQvgwUM/bGMuyT6neihdooXVO+GV583yUJ3NVctjJTwOvL0kzAw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:32:13.167182Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.06461","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:362f3e88026bccb907fd8ab26272320cdd7c3a7d4101d5edcac7ee3fff3696ad","sha256:ea29accb3cae54c173d2533982a44cab9b44022bf1714ea8029263e8d7d3e2f9"],"state_sha256":"68ff20d04a52fef9b24741d8d6e0403cb670d3b7f1d10f6309de96214ed73721"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JDq5Qdcc/v45pcae3cFdLYtCG9JErNeLwweKjGogX5+BTKR4nHircDwX4MBIGJTHgSQczwoeM+92T2blITa+Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T05:47:50.984979Z","bundle_sha256":"721c30cb67d5ec73d473bcadb905b117d25dd34d19d84c3caecc996d1cbc0e6c"}}