Pessimistic Verification for Open Ended Math Questions Article Swipe
The key limitation of the verification performance lies in the ability of error detection. With this intuition we designed several variants of pessimistic verification, which are simple workflows that could significantly improve the verification of open-ended math questions. In pessimistic verification we construct multiple parallel verifications for the same proof, and the proof is deemed incorrect if any one of them reports an error. This simple technique significantly improves the performance across many math verification benchmarks without incurring substantial computational resources. Its token efficiency even surpassed extended long-CoT in test-time scaling. Our case studies further indicate that the majority of false negatives in stronger models are actually caused by annotation errors in the original dataset, so our method's performance is in fact underestimated. Self-verification for mathematical problems can effectively improve the reliability and performance of language model outputs, and it also plays a critical role in enabling long-horizon mathematical tasks. We believe that research on pessimistic verification will help enhance the mathematical capabilities of language models across a wide range of tasks.
Related Topics
Concepts
Computer science
Intuition
Pessimism
Simple (philosophy)
Theoretical computer science
Reliability (semiconductor)
Construct (python library)
Mathematical proof
Workflow
Range (aeronautics)
Algorithm
Artificial intelligence
Key (lock)
Programming language
Semaphore
Mathematics
Formal verification
Mathematical logic
Mathematical model
Computational complexity theory
Machine learning
Metadata
- Type
- article
- Landing Page
- http://arxiv.org/abs/2511.21522
- https://arxiv.org/pdf/2511.21522
- OA Status
- green
- OpenAlex ID
- https://openalex.org/W7106863019
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W7106863019Canonical identifier for this work in OpenAlex
- Title
-
Pessimistic Verification for Open Ended Math QuestionsWork title
- Type
-
articleOpenAlex work type
- Publication year
-
2025Year of publication
- Publication date
-
2025-11-26Full publication date if available
- Authors
-
Huang Yanxing, Tang Zihan, Lin Zejin, Li Peng, Liu YangList of authors in order
- Landing page
-
https://arxiv.org/abs/2511.21522Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2511.21522Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2511.21522Direct OA link when available
- Concepts
-
Computer science, Intuition, Pessimism, Simple (philosophy), Theoretical computer science, Reliability (semiconductor), Construct (python library), Mathematical proof, Workflow, Range (aeronautics), Algorithm, Artificial intelligence, Key (lock), Programming language, Semaphore, Mathematics, Formal verification, Mathematical logic, Mathematical model, Computational complexity theory, Machine learningTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
0Total citation count in OpenAlex
Full payload
| id | https://openalex.org/W7106863019 |
|---|---|
| doi | |
| ids.openalex | https://openalex.org/W7106863019 |
| fwci | 0.0 |
| type | article |
| title | Pessimistic Verification for Open Ended Math Questions |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T13523 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.30717775225639343 |
| topics[0].domain.id | https://openalex.org/domains/3 |
| topics[0].domain.display_name | Physical Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/1703 |
| topics[0].subfield.display_name | Computational Theory and Mathematics |
| topics[0].display_name | Mathematics, Computing, and Information Processing |
| topics[1].id | https://openalex.org/T11206 |
| topics[1].field.id | https://openalex.org/fields/31 |
| topics[1].field.display_name | Physics and Astronomy |
| topics[1].score | 0.158406063914299 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/3109 |
| topics[1].subfield.display_name | Statistical and Nonlinear Physics |
| topics[1].display_name | Model Reduction and Neural Networks |
| topics[2].id | https://openalex.org/T10028 |
| topics[2].field.id | https://openalex.org/fields/17 |
| topics[2].field.display_name | Computer Science |
| topics[2].score | 0.08058294653892517 |
| topics[2].domain.id | https://openalex.org/domains/3 |
| topics[2].domain.display_name | Physical Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/1702 |
| topics[2].subfield.display_name | Artificial Intelligence |
| topics[2].display_name | Topic Modeling |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C41008148 |
| concepts[0].level | 0 |
| concepts[0].score | 0.6403766870498657 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[0].display_name | Computer science |
| concepts[1].id | https://openalex.org/C132010649 |
| concepts[1].level | 2 |
| concepts[1].score | 0.6146528720855713 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q189222 |
| concepts[1].display_name | Intuition |
| concepts[2].id | https://openalex.org/C9992130 |
| concepts[2].level | 2 |
| concepts[2].score | 0.5798895955085754 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q484954 |
| concepts[2].display_name | Pessimism |
| concepts[3].id | https://openalex.org/C2780586882 |
| concepts[3].level | 2 |
| concepts[3].score | 0.49071305990219116 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q7520643 |
| concepts[3].display_name | Simple (philosophy) |
| concepts[4].id | https://openalex.org/C80444323 |
| concepts[4].level | 1 |
| concepts[4].score | 0.44507598876953125 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q2878974 |
| concepts[4].display_name | Theoretical computer science |
| concepts[5].id | https://openalex.org/C43214815 |
| concepts[5].level | 3 |
| concepts[5].score | 0.41462352871894836 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q7310987 |
| concepts[5].display_name | Reliability (semiconductor) |
| concepts[6].id | https://openalex.org/C2780801425 |
| concepts[6].level | 2 |
| concepts[6].score | 0.4005880355834961 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q5164392 |
| concepts[6].display_name | Construct (python library) |
| concepts[7].id | https://openalex.org/C108710211 |
| concepts[7].level | 2 |
| concepts[7].score | 0.3790738880634308 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q11538 |
| concepts[7].display_name | Mathematical proof |
| concepts[8].id | https://openalex.org/C177212765 |
| concepts[8].level | 2 |
| concepts[8].score | 0.35564008355140686 |
| concepts[8].wikidata | https://www.wikidata.org/wiki/Q627335 |
| concepts[8].display_name | Workflow |
| concepts[9].id | https://openalex.org/C204323151 |
| concepts[9].level | 2 |
| concepts[9].score | 0.33134275674819946 |
| concepts[9].wikidata | https://www.wikidata.org/wiki/Q905424 |
| concepts[9].display_name | Range (aeronautics) |
| concepts[10].id | https://openalex.org/C11413529 |
| concepts[10].level | 1 |
| concepts[10].score | 0.32944709062576294 |
| concepts[10].wikidata | https://www.wikidata.org/wiki/Q8366 |
| concepts[10].display_name | Algorithm |
| concepts[11].id | https://openalex.org/C154945302 |
| concepts[11].level | 1 |
| concepts[11].score | 0.3255106210708618 |
| concepts[11].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[11].display_name | Artificial intelligence |
| concepts[12].id | https://openalex.org/C26517878 |
| concepts[12].level | 2 |
| concepts[12].score | 0.2968299388885498 |
| concepts[12].wikidata | https://www.wikidata.org/wiki/Q228039 |
| concepts[12].display_name | Key (lock) |
| concepts[13].id | https://openalex.org/C199360897 |
| concepts[13].level | 1 |
| concepts[13].score | 0.29546937346458435 |
| concepts[13].wikidata | https://www.wikidata.org/wiki/Q9143 |
| concepts[13].display_name | Programming language |
| concepts[14].id | https://openalex.org/C95203288 |
| concepts[14].level | 2 |
| concepts[14].score | 0.2927919924259186 |
| concepts[14].wikidata | https://www.wikidata.org/wiki/Q221682 |
| concepts[14].display_name | Semaphore |
| concepts[15].id | https://openalex.org/C33923547 |
| concepts[15].level | 0 |
| concepts[15].score | 0.2916664779186249 |
| concepts[15].wikidata | https://www.wikidata.org/wiki/Q395 |
| concepts[15].display_name | Mathematics |
| concepts[16].id | https://openalex.org/C111498074 |
| concepts[16].level | 2 |
| concepts[16].score | 0.28944504261016846 |
| concepts[16].wikidata | https://www.wikidata.org/wiki/Q173326 |
| concepts[16].display_name | Formal verification |
| concepts[17].id | https://openalex.org/C47884741 |
| concepts[17].level | 2 |
| concepts[17].score | 0.28088635206222534 |
| concepts[17].wikidata | https://www.wikidata.org/wiki/Q1166618 |
| concepts[17].display_name | Mathematical logic |
| concepts[18].id | https://openalex.org/C76969082 |
| concepts[18].level | 2 |
| concepts[18].score | 0.2679521143436432 |
| concepts[18].wikidata | https://www.wikidata.org/wiki/Q486902 |
| concepts[18].display_name | Mathematical model |
| concepts[19].id | https://openalex.org/C179799912 |
| concepts[19].level | 2 |
| concepts[19].score | 0.2628382742404938 |
| concepts[19].wikidata | https://www.wikidata.org/wiki/Q205084 |
| concepts[19].display_name | Computational complexity theory |
| concepts[20].id | https://openalex.org/C119857082 |
| concepts[20].level | 1 |
| concepts[20].score | 0.25772765278816223 |
| concepts[20].wikidata | https://www.wikidata.org/wiki/Q2539 |
| concepts[20].display_name | Machine learning |
| keywords[0].id | https://openalex.org/keywords/intuition |
| keywords[0].score | 0.6146528720855713 |
| keywords[0].display_name | Intuition |
| keywords[1].id | https://openalex.org/keywords/pessimism |
| keywords[1].score | 0.5798895955085754 |
| keywords[1].display_name | Pessimism |
| keywords[2].id | https://openalex.org/keywords/simple |
| keywords[2].score | 0.49071305990219116 |
| keywords[2].display_name | Simple (philosophy) |
| keywords[3].id | https://openalex.org/keywords/reliability |
| keywords[3].score | 0.41462352871894836 |
| keywords[3].display_name | Reliability (semiconductor) |
| keywords[4].id | https://openalex.org/keywords/construct |
| keywords[4].score | 0.4005880355834961 |
| keywords[4].display_name | Construct (python library) |
| keywords[5].id | https://openalex.org/keywords/mathematical-proof |
| keywords[5].score | 0.3790738880634308 |
| keywords[5].display_name | Mathematical proof |
| keywords[6].id | https://openalex.org/keywords/workflow |
| keywords[6].score | 0.35564008355140686 |
| keywords[6].display_name | Workflow |
| language | |
| locations[0].id | pmh:oai:arXiv.org:2511.21522 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4306400194 |
| locations[0].source.issn | |
| locations[0].source.type | repository |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | arXiv (Cornell University) |
| locations[0].source.host_organization | https://openalex.org/I205783295 |
| locations[0].source.host_organization_name | Cornell University |
| locations[0].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[0].license | cc-by |
| locations[0].pdf_url | https://arxiv.org/pdf/2511.21522 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | https://openalex.org/licenses/cc-by |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2511.21522 |
| indexed_in | arxiv |
| authorships[0].author.id | https://openalex.org/A2747663557 |
| authorships[0].author.orcid | |
| authorships[0].author.display_name | Huang Yanxing |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Huang, Yanxing |
| authorships[0].is_corresponding | True |
| authorships[1].author.id | https://openalex.org/A2749664720 |
| authorships[1].author.orcid | |
| authorships[1].author.display_name | Tang Zihan |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Tang, Zihan |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A2385073983 |
| authorships[2].author.orcid | |
| authorships[2].author.display_name | Lin Zejin |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Lin, Zejin |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A1992351433 |
| authorships[3].author.orcid | https://orcid.org/0000-0001-5358-6736 |
| authorships[3].author.display_name | Li Peng |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Li, Peng |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A2102234800 |
| authorships[4].author.orcid | https://orcid.org/0000-0001-5133-0470 |
| authorships[4].author.display_name | Liu Yang |
| authorships[4].author_position | last |
| authorships[4].raw_author_name | Liu, Yang |
| authorships[4].is_corresponding | False |
| has_content.pdf | True |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2511.21522 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-11-28T00:00:00 |
| display_name | Pessimistic Verification for Open Ended Math Questions |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-28T23:14:17.795251 |
| primary_topic.id | https://openalex.org/T13523 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.30717775225639343 |
| primary_topic.domain.id | https://openalex.org/domains/3 |
| primary_topic.domain.display_name | Physical Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/1703 |
| primary_topic.subfield.display_name | Computational Theory and Mathematics |
| primary_topic.display_name | Mathematics, Computing, and Information Processing |
| cited_by_count | 0 |
| locations_count | 1 |
| best_oa_location.id | pmh:oai:arXiv.org:2511.21522 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4306400194 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | repository |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | arXiv (Cornell University) |
| best_oa_location.source.host_organization | https://openalex.org/I205783295 |
| best_oa_location.source.host_organization_name | Cornell University |
| best_oa_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| best_oa_location.license | cc-by |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2511.21522 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | https://openalex.org/licenses/cc-by |
| best_oa_location.is_accepted | False |
| best_oa_location.is_published | False |
| best_oa_location.raw_source_name | |
| best_oa_location.landing_page_url | http://arxiv.org/abs/2511.21522 |
| primary_location.id | pmh:oai:arXiv.org:2511.21522 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4306400194 |
| primary_location.source.issn | |
| primary_location.source.type | repository |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | arXiv (Cornell University) |
| primary_location.source.host_organization | https://openalex.org/I205783295 |
| primary_location.source.host_organization_name | Cornell University |
| primary_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| primary_location.license | cc-by |
| primary_location.pdf_url | https://arxiv.org/pdf/2511.21522 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | https://openalex.org/licenses/cc-by |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2511.21522 |
| publication_date | 2025-11-26 |
| publication_year | 2025 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 142, 167 |
| abstract_inverted_index.In | 38 |
| abstract_inverted_index.We | 150 |
| abstract_inverted_index.an | 62 |
| abstract_inverted_index.by | 108 |
| abstract_inverted_index.if | 56 |
| abstract_inverted_index.in | 8, 88, 102, 111, 120, 145 |
| abstract_inverted_index.is | 53, 119 |
| abstract_inverted_index.it | 139 |
| abstract_inverted_index.of | 3, 11, 21, 34, 59, 99, 134, 163, 170 |
| abstract_inverted_index.on | 154 |
| abstract_inverted_index.so | 115 |
| abstract_inverted_index.we | 17, 41 |
| abstract_inverted_index.Its | 81 |
| abstract_inverted_index.Our | 91 |
| abstract_inverted_index.The | 0 |
| abstract_inverted_index.and | 50, 132, 138 |
| abstract_inverted_index.any | 57 |
| abstract_inverted_index.are | 25, 105 |
| abstract_inverted_index.can | 127 |
| abstract_inverted_index.for | 46, 124 |
| abstract_inverted_index.key | 1 |
| abstract_inverted_index.one | 58 |
| abstract_inverted_index.our | 116 |
| abstract_inverted_index.the | 4, 9, 32, 47, 51, 69, 97, 112, 130, 160 |
| abstract_inverted_index.This | 64 |
| abstract_inverted_index.With | 14 |
| abstract_inverted_index.also | 140 |
| abstract_inverted_index.case | 92 |
| abstract_inverted_index.even | 84 |
| abstract_inverted_index.fact | 121 |
| abstract_inverted_index.help | 158 |
| abstract_inverted_index.lies | 7 |
| abstract_inverted_index.many | 72 |
| abstract_inverted_index.math | 36, 73 |
| abstract_inverted_index.role | 144 |
| abstract_inverted_index.same | 48 |
| abstract_inverted_index.that | 28, 96, 152 |
| abstract_inverted_index.them | 60 |
| abstract_inverted_index.this | 15 |
| abstract_inverted_index.wide | 168 |
| abstract_inverted_index.will | 157 |
| abstract_inverted_index.could | 29 |
| abstract_inverted_index.error | 12 |
| abstract_inverted_index.false | 100 |
| abstract_inverted_index.model | 136 |
| abstract_inverted_index.plays | 141 |
| abstract_inverted_index.proof | 52 |
| abstract_inverted_index.range | 169 |
| abstract_inverted_index.token | 82 |
| abstract_inverted_index.which | 24 |
| abstract_inverted_index.across | 71, 166 |
| abstract_inverted_index.caused | 107 |
| abstract_inverted_index.deemed | 54 |
| abstract_inverted_index.error. | 63 |
| abstract_inverted_index.errors | 110 |
| abstract_inverted_index.models | 104, 165 |
| abstract_inverted_index.proof, | 49 |
| abstract_inverted_index.simple | 26, 65 |
| abstract_inverted_index.tasks. | 149, 171 |
| abstract_inverted_index.ability | 10 |
| abstract_inverted_index.believe | 151 |
| abstract_inverted_index.enhance | 159 |
| abstract_inverted_index.further | 94 |
| abstract_inverted_index.improve | 31, 129 |
| abstract_inverted_index.reports | 61 |
| abstract_inverted_index.several | 19 |
| abstract_inverted_index.studies | 93 |
| abstract_inverted_index.without | 76 |
| abstract_inverted_index.actually | 106 |
| abstract_inverted_index.critical | 143 |
| abstract_inverted_index.dataset, | 114 |
| abstract_inverted_index.designed | 18 |
| abstract_inverted_index.enabling | 146 |
| abstract_inverted_index.extended | 86 |
| abstract_inverted_index.improves | 68 |
| abstract_inverted_index.indicate | 95 |
| abstract_inverted_index.language | 135, 164 |
| abstract_inverted_index.long-CoT | 87 |
| abstract_inverted_index.majority | 98 |
| abstract_inverted_index.method's | 117 |
| abstract_inverted_index.multiple | 43 |
| abstract_inverted_index.original | 113 |
| abstract_inverted_index.outputs, | 137 |
| abstract_inverted_index.parallel | 44 |
| abstract_inverted_index.problems | 126 |
| abstract_inverted_index.research | 153 |
| abstract_inverted_index.scaling. | 90 |
| abstract_inverted_index.stronger | 103 |
| abstract_inverted_index.variants | 20 |
| abstract_inverted_index.construct | 42 |
| abstract_inverted_index.incorrect | 55 |
| abstract_inverted_index.incurring | 77 |
| abstract_inverted_index.intuition | 16 |
| abstract_inverted_index.negatives | 101 |
| abstract_inverted_index.surpassed | 85 |
| abstract_inverted_index.technique | 66 |
| abstract_inverted_index.test-time | 89 |
| abstract_inverted_index.workflows | 27 |
| abstract_inverted_index.annotation | 109 |
| abstract_inverted_index.benchmarks | 75 |
| abstract_inverted_index.detection. | 13 |
| abstract_inverted_index.efficiency | 83 |
| abstract_inverted_index.limitation | 2 |
| abstract_inverted_index.open-ended | 35 |
| abstract_inverted_index.questions. | 37 |
| abstract_inverted_index.resources. | 80 |
| abstract_inverted_index.effectively | 128 |
| abstract_inverted_index.performance | 6, 70, 118, 133 |
| abstract_inverted_index.pessimistic | 22, 39, 155 |
| abstract_inverted_index.reliability | 131 |
| abstract_inverted_index.substantial | 78 |
| abstract_inverted_index.capabilities | 162 |
| abstract_inverted_index.long-horizon | 147 |
| abstract_inverted_index.mathematical | 125, 148, 161 |
| abstract_inverted_index.verification | 5, 33, 40, 74, 156 |
| abstract_inverted_index.computational | 79 |
| abstract_inverted_index.significantly | 30, 67 |
| abstract_inverted_index.verification, | 23 |
| abstract_inverted_index.verifications | 45 |
| abstract_inverted_index.underestimated. | 122 |
| abstract_inverted_index.Self-verification | 123 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 5 |
| citation_normalized_percentile.value | 0.83657357 |
| citation_normalized_percentile.is_in_top_1_percent | False |
| citation_normalized_percentile.is_in_top_10_percent | False |