{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:MYX5JO7HZOPAKSBOLXR7VHU7S5","short_pith_number":"pith:MYX5JO7H","canonical_record":{"source":{"id":"2605.28020","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T06:25:00Z","cross_cats_sorted":[],"title_canon_sha256":"01f2054a160816756f05f8a7e2641233df9874246e750e8bb59906fa57390a1b","abstract_canon_sha256":"8137284c129b1f1b28b70b22c90137b7b91e800220cc9e058b95acf72d45b263"},"schema_version":"1.0"},"canonical_sha256":"662fd4bbe7cb9e05482e5de3fa9e9f974c0c791ee13b240534e9173ae3c59f26","source":{"kind":"arxiv","id":"2605.28020","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.28020","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"arxiv_version","alias_value":"2605.28020v1","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28020","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_12","alias_value":"MYX5JO7HZOPA","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_16","alias_value":"MYX5JO7HZOPAKSBO","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_8","alias_value":"MYX5JO7H","created_at":"2026-05-28T01:04:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:MYX5JO7HZOPAKSBOLXR7VHU7S5","target":"record","payload":{"canonical_record":{"source":{"id":"2605.28020","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T06:25:00Z","cross_cats_sorted":[],"title_canon_sha256":"01f2054a160816756f05f8a7e2641233df9874246e750e8bb59906fa57390a1b","abstract_canon_sha256":"8137284c129b1f1b28b70b22c90137b7b91e800220cc9e058b95acf72d45b263"},"schema_version":"1.0"},"canonical_sha256":"662fd4bbe7cb9e05482e5de3fa9e9f974c0c791ee13b240534e9173ae3c59f26","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:56.201300Z","signature_b64":"ZPkSrZNYpanHCzc1++PU559mj32OPOEvbYN1xQUxO22jAlESd7McPntusYcjdM062utQt+nPrXIgexTf76bZCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"662fd4bbe7cb9e05482e5de3fa9e9f974c0c791ee13b240534e9173ae3c59f26","last_reissued_at":"2026-05-28T01:04:56.200905Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:56.200905Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.28020","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-28T01:04:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"us2LjSIO3CUh1t6HxX4K5HQpE+3O9dk0as4vKkdNOwxBabpnY3UFnEKgXGQvYe/vfEUbAZbnXYQvKE93frNDBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T20:12:18.828926Z"},"content_sha256":"fc70d782517c0d8b93a04ff09d95d249b459196942246e1ff217b9727b4b06e7","schema_version":"1.0","event_id":"sha256:fc70d782517c0d8b93a04ff09d95d249b459196942246e1ff217b9727b4b06e7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:MYX5JO7HZOPAKSBOLXR7VHU7S5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Missing Piece in Pre-trained Model Evaluation: Reward-Guided Decoding Unlocks Task-Oriented Behavior Without Parameter Updates","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dayiheng Liu, Guo Chen, Linfeng Zhang, Qingyang Liu, Shaobo Wang, Xingzhang Ren, Zhengyang Tang, Ziyue Wang","submitted_at":"2026-05-27T06:25:00Z","abstract_excerpt":"With the rapid progress of large language models (LLMs), reliably evaluating the capabilities of pre-trained LLMs has become increasingly important. The challenge is that base pre-trained models are optimized for next-token prediction and often fail to follow instructions or produce well-formed answers under standard prompting and direct decoding. As a result, benchmark performance can conflate model capability with decoding-induced failures to produce task-oriented outputs, while exposing such behavior often relies on costly post-training. Recent decodingonly approaches attempt to reshape out"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28020","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/2605.28020/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":"2026-05-28T01:04:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qwJNCYdLzdnsmCzuq8VC4357ZgewrKRyMqL+y7d1St+H96f/VZf0wvV0d2eodYXLTKRJlq5n2huyPZGvKfl5DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T20:12:18.829309Z"},"content_sha256":"108287ef575508b0b18e7027764fc28399e3fb8bdd1100dfebf5e04f106d2dc5","schema_version":"1.0","event_id":"sha256:108287ef575508b0b18e7027764fc28399e3fb8bdd1100dfebf5e04f106d2dc5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MYX5JO7HZOPAKSBOLXR7VHU7S5/bundle.json","state_url":"https://pith.science/pith/MYX5JO7HZOPAKSBOLXR7VHU7S5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MYX5JO7HZOPAKSBOLXR7VHU7S5/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":"2026-05-28T20:12:18Z","links":{"resolver":"https://pith.science/pith/MYX5JO7HZOPAKSBOLXR7VHU7S5","bundle":"https://pith.science/pith/MYX5JO7HZOPAKSBOLXR7VHU7S5/bundle.json","state":"https://pith.science/pith/MYX5JO7HZOPAKSBOLXR7VHU7S5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MYX5JO7HZOPAKSBOLXR7VHU7S5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:MYX5JO7HZOPAKSBOLXR7VHU7S5","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":"8137284c129b1f1b28b70b22c90137b7b91e800220cc9e058b95acf72d45b263","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T06:25:00Z","title_canon_sha256":"01f2054a160816756f05f8a7e2641233df9874246e750e8bb59906fa57390a1b"},"schema_version":"1.0","source":{"id":"2605.28020","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.28020","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"arxiv_version","alias_value":"2605.28020v1","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28020","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_12","alias_value":"MYX5JO7HZOPA","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_16","alias_value":"MYX5JO7HZOPAKSBO","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_8","alias_value":"MYX5JO7H","created_at":"2026-05-28T01:04:56Z"}],"graph_snapshots":[{"event_id":"sha256:108287ef575508b0b18e7027764fc28399e3fb8bdd1100dfebf5e04f106d2dc5","target":"graph","created_at":"2026-05-28T01:04:56Z","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/2605.28020/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With the rapid progress of large language models (LLMs), reliably evaluating the capabilities of pre-trained LLMs has become increasingly important. The challenge is that base pre-trained models are optimized for next-token prediction and often fail to follow instructions or produce well-formed answers under standard prompting and direct decoding. As a result, benchmark performance can conflate model capability with decoding-induced failures to produce task-oriented outputs, while exposing such behavior often relies on costly post-training. Recent decodingonly approaches attempt to reshape out","authors_text":"Dayiheng Liu, Guo Chen, Linfeng Zhang, Qingyang Liu, Shaobo Wang, Xingzhang Ren, Zhengyang Tang, Ziyue Wang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T06:25:00Z","title":"The Missing Piece in Pre-trained Model Evaluation: Reward-Guided Decoding Unlocks Task-Oriented Behavior Without Parameter Updates"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28020","kind":"arxiv","version":1},"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:fc70d782517c0d8b93a04ff09d95d249b459196942246e1ff217b9727b4b06e7","target":"record","created_at":"2026-05-28T01:04:56Z","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":"8137284c129b1f1b28b70b22c90137b7b91e800220cc9e058b95acf72d45b263","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T06:25:00Z","title_canon_sha256":"01f2054a160816756f05f8a7e2641233df9874246e750e8bb59906fa57390a1b"},"schema_version":"1.0","source":{"id":"2605.28020","kind":"arxiv","version":1}},"canonical_sha256":"662fd4bbe7cb9e05482e5de3fa9e9f974c0c791ee13b240534e9173ae3c59f26","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"662fd4bbe7cb9e05482e5de3fa9e9f974c0c791ee13b240534e9173ae3c59f26","first_computed_at":"2026-05-28T01:04:56.200905Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:04:56.200905Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZPkSrZNYpanHCzc1++PU559mj32OPOEvbYN1xQUxO22jAlESd7McPntusYcjdM062utQt+nPrXIgexTf76bZCA==","signature_status":"signed_v1","signed_at":"2026-05-28T01:04:56.201300Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.28020","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fc70d782517c0d8b93a04ff09d95d249b459196942246e1ff217b9727b4b06e7","sha256:108287ef575508b0b18e7027764fc28399e3fb8bdd1100dfebf5e04f106d2dc5"],"state_sha256":"5be0276dc4633d6556b222503719088d12c6b06dbaa26e1a6af07d56b35a4861"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CcS00OuDD3igqxbKTCXoUoEvIjYcfDOuZNeHd+3I4r7RRcGf9lq15Wn5ywia+vurMpJQpG43/RxtP6OEOJNIBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-28T20:12:18.831632Z","bundle_sha256":"44d36d284b06eac09898a02a9a502e24631e29d05cecf0cdfa452de28777168d"}}