{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:6RVCV2MFWNQ5VJU7QJYEQND3RJ","short_pith_number":"pith:6RVCV2MF","canonical_record":{"source":{"id":"2505.19634","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T07:51:30Z","cross_cats_sorted":[],"title_canon_sha256":"081c2c4b1aa089a7c2bb51f66393f5678ade2e434879cb86120fbcce8de29f71","abstract_canon_sha256":"fb9b87b628a25ea97ec432f897eedb67fb1e2141b2da112629728fb6bf844feb"},"schema_version":"1.0"},"canonical_sha256":"f46a2ae985b361daa69f827048347b8a619434601dea0425f50967c6fa11c699","source":{"kind":"arxiv","id":"2505.19634","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19634","created_at":"2026-07-05T12:09:44Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19634v4","created_at":"2026-07-05T12:09:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19634","created_at":"2026-07-05T12:09:44Z"},{"alias_kind":"pith_short_12","alias_value":"6RVCV2MFWNQ5","created_at":"2026-07-05T12:09:44Z"},{"alias_kind":"pith_short_16","alias_value":"6RVCV2MFWNQ5VJU7","created_at":"2026-07-05T12:09:44Z"},{"alias_kind":"pith_short_8","alias_value":"6RVCV2MF","created_at":"2026-07-05T12:09:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:6RVCV2MFWNQ5VJU7QJYEQND3RJ","target":"record","payload":{"canonical_record":{"source":{"id":"2505.19634","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T07:51:30Z","cross_cats_sorted":[],"title_canon_sha256":"081c2c4b1aa089a7c2bb51f66393f5678ade2e434879cb86120fbcce8de29f71","abstract_canon_sha256":"fb9b87b628a25ea97ec432f897eedb67fb1e2141b2da112629728fb6bf844feb"},"schema_version":"1.0"},"canonical_sha256":"f46a2ae985b361daa69f827048347b8a619434601dea0425f50967c6fa11c699","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:09:44.508284Z","signature_b64":"grr4SEtcgyEDwp7OMs3ChsXFn2sTzSmf3LwRKvsFvARJEDXc/hXCiCL+r2zENB/3VS1ZCeeElT91GSdAqCn8Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f46a2ae985b361daa69f827048347b8a619434601dea0425f50967c6fa11c699","last_reissued_at":"2026-07-05T12:09:44.507811Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:09:44.507811Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.19634","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T12:09:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vsp5Z60yOze/ZnYOoW3uMcCdGHMQJQPCwfgX5w8qBtbw25YWWxc7NIDio13ep4+T4Wo0Y7myIIrGFHGFthLEDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T04:15:19.766641Z"},"content_sha256":"f2f772ad02502b73fda007327add377ca4f03bc07d8fc8945fad3cc752999213","schema_version":"1.0","event_id":"sha256:f2f772ad02502b73fda007327add377ca4f03bc07d8fc8945fad3cc752999213"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:6RVCV2MFWNQ5VJU7QJYEQND3RJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Faster and Better LLMs via Latency-Aware Test-Time Scaling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Haoli Bai, Lei Zhu, Lu Hou, Shiming Xiang, Tianyu Zhang, Wulong Liu, Xianzhi Yu, Zili Wang","submitted_at":"2025-05-26T07:51:30Z","abstract_excerpt":"Test-Time Scaling (TTS) has proven effective in improving the performance of Large Language Models (LLMs) during inference. However, existing research has overlooked the efficiency of TTS from a latency-sensitive perspective. Through a latency-aware evaluation of representative TTS methods, we demonstrate that a compute-optimal TTS does not always result in the lowest latency in scenarios where latency is critical. To address this gap and achieve latency-optimal TTS, we propose two key approaches by optimizing the concurrency configurations: (1) branch-wise parallelism, which leverages multipl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19634","kind":"arxiv","version":4},"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/2505.19634/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":"2026-07-05T12:09:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ute/YOv5vAAXrBesoWLMShl0y9o87TKIil5hZpT1aCggGWdT+49wqmPWqymFJE5nQ+eHHjR9YZMsed97IAGDBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T04:15:19.767309Z"},"content_sha256":"8c442141abe256c607c13a3302d7d59a67792b7af71cf45be2d935b26acc18e4","schema_version":"1.0","event_id":"sha256:8c442141abe256c607c13a3302d7d59a67792b7af71cf45be2d935b26acc18e4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6RVCV2MFWNQ5VJU7QJYEQND3RJ/bundle.json","state_url":"https://pith.science/pith/6RVCV2MFWNQ5VJU7QJYEQND3RJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6RVCV2MFWNQ5VJU7QJYEQND3RJ/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":"2026-08-10T04:15:19Z","links":{"resolver":"https://pith.science/pith/6RVCV2MFWNQ5VJU7QJYEQND3RJ","bundle":"https://pith.science/pith/6RVCV2MFWNQ5VJU7QJYEQND3RJ/bundle.json","state":"https://pith.science/pith/6RVCV2MFWNQ5VJU7QJYEQND3RJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6RVCV2MFWNQ5VJU7QJYEQND3RJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:6RVCV2MFWNQ5VJU7QJYEQND3RJ","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":"fb9b87b628a25ea97ec432f897eedb67fb1e2141b2da112629728fb6bf844feb","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T07:51:30Z","title_canon_sha256":"081c2c4b1aa089a7c2bb51f66393f5678ade2e434879cb86120fbcce8de29f71"},"schema_version":"1.0","source":{"id":"2505.19634","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19634","created_at":"2026-07-05T12:09:44Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19634v4","created_at":"2026-07-05T12:09:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19634","created_at":"2026-07-05T12:09:44Z"},{"alias_kind":"pith_short_12","alias_value":"6RVCV2MFWNQ5","created_at":"2026-07-05T12:09:44Z"},{"alias_kind":"pith_short_16","alias_value":"6RVCV2MFWNQ5VJU7","created_at":"2026-07-05T12:09:44Z"},{"alias_kind":"pith_short_8","alias_value":"6RVCV2MF","created_at":"2026-07-05T12:09:44Z"}],"graph_snapshots":[{"event_id":"sha256:8c442141abe256c607c13a3302d7d59a67792b7af71cf45be2d935b26acc18e4","target":"graph","created_at":"2026-07-05T12:09:44Z","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/2505.19634/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Test-Time Scaling (TTS) has proven effective in improving the performance of Large Language Models (LLMs) during inference. However, existing research has overlooked the efficiency of TTS from a latency-sensitive perspective. Through a latency-aware evaluation of representative TTS methods, we demonstrate that a compute-optimal TTS does not always result in the lowest latency in scenarios where latency is critical. To address this gap and achieve latency-optimal TTS, we propose two key approaches by optimizing the concurrency configurations: (1) branch-wise parallelism, which leverages multipl","authors_text":"Haoli Bai, Lei Zhu, Lu Hou, Shiming Xiang, Tianyu Zhang, Wulong Liu, Xianzhi Yu, Zili Wang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T07:51:30Z","title":"Faster and Better LLMs via Latency-Aware Test-Time Scaling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19634","kind":"arxiv","version":4},"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:f2f772ad02502b73fda007327add377ca4f03bc07d8fc8945fad3cc752999213","target":"record","created_at":"2026-07-05T12:09:44Z","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":"fb9b87b628a25ea97ec432f897eedb67fb1e2141b2da112629728fb6bf844feb","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T07:51:30Z","title_canon_sha256":"081c2c4b1aa089a7c2bb51f66393f5678ade2e434879cb86120fbcce8de29f71"},"schema_version":"1.0","source":{"id":"2505.19634","kind":"arxiv","version":4}},"canonical_sha256":"f46a2ae985b361daa69f827048347b8a619434601dea0425f50967c6fa11c699","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f46a2ae985b361daa69f827048347b8a619434601dea0425f50967c6fa11c699","first_computed_at":"2026-07-05T12:09:44.507811Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:09:44.507811Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"grr4SEtcgyEDwp7OMs3ChsXFn2sTzSmf3LwRKvsFvARJEDXc/hXCiCL+r2zENB/3VS1ZCeeElT91GSdAqCn8Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T12:09:44.508284Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.19634","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f2f772ad02502b73fda007327add377ca4f03bc07d8fc8945fad3cc752999213","sha256:8c442141abe256c607c13a3302d7d59a67792b7af71cf45be2d935b26acc18e4"],"state_sha256":"e307fd332e30547aefcec88bf6c74da1575a96defbecbc44c29239eae9907766"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DeqUfNh/KsqzO6ODICPzb0vwOV94vQb8ojYrckaW7zFimdr/w9MM6L+XXp9Zba39Ne+6Usv0Hh1wQSijruKPCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T04:15:19.772847Z","bundle_sha256":"73f62ade9a7d89e0360d65bddfcc3d9d8153aa5c6907174ab8e929db9e9894c2"}}