Knowledge
Factual recall and domain expertise
350 datasets
last indexed 17h ago
Model Leaderboard
| No. | Model | Score |
|---|---|---|
| 1 |  Gemini 3 Pro | 1.00 |
| 2 | Grok 4 | 0.60 |
| 3 |  GPT-5 | 0.60 |
| 4 | Qwen3 235B A22B | 0.59 |
| 5 | Gemini 2.5 Pro | 0.58 |
| 6 | o3 | 0.56 |
| 7 | GPT-5 Mini | 0.48 |
| 8 |  Claude Sonnet 4.5 | 0.32 |
| 9 | GPT-4.1 | 0.31 |
| 10 | o1 | 0.30 |
| 11 | Grok 3 | 0.28 |
| 12 | GPT-5.1 | 0.18 |
| 13 | Claude Opus 4.1 | 0.17 |
| 14 | Gemini 2.5 Flash | 0.15 |
| 15 | Claude 3.7 Sonnet | 0.12 |
| 16 |  DeepSeek-R1 | 0.10 |
| 17 | GPT-4o | 0.08 |
| 18 | Claude 4.5 Sonnet (thinking) | 0.06 |
| 19 | DeepSeek-V3 | 0.02 |
| 20 | Claude Opus 4 (thinking) | 0.00 |
| 21 | Claude 3.5 Sonnet | 0.00 |
| Name | Organization | Best Model |
|---|---|---|
| Humanity's Last Exam | Scale AI | Gemini 3 Pro |
| SimpleQA Verified | Google DeepMind | Gemini 3 Pro Preview |
| TimeQA | University of California, Santa Barbara | Human Worker |
| RAGBench | Galileo Technologies Inc. | DeBERTA |
| OMGEval | Northeastern University | GPT-4 |
| SimpleQA | OpenAI | OpenAI o1-preview |
| SuperGPQA | 2077.AI | gpt-5 |
| MedXpertQA | Shanghai Artificial Intelligence Laboratory | o1 |
| RAGTruth | University of Illinois at Urbana-Champaign | GPT-4-0613 |
| POPQA | University of Washington | GPT-3 davinci-003 |
| Visual-RAG | Nanjing University of Science and Technology | Gemini 2.5 Pro |
| ESGenius | Alibaba Group | o3 |
| ChroKnowBench | Korea University | GPT4o-mini |
| Factcheck-Bench | University of Illinois at Urbana-Champaign | Factcheck-GPT (Web) |
| LongFact | Google DeepMind | GPT-4-Turbo |
| LLM-AggreFact | The University of Texas at Austin | Bespoke-Minicheck-7B |
| SciAssess | DP Technology | GPT-4o |
| ARC | Allen Institute for Artificial Intelligence | DGEM |
| CulturalBench | University of Washington | o1 |
| AGIEval | Microsoft | GPT-4o |
| SciFIBench | Google DeepMind | GPT-4o |
| HalluEditBench | Illinois Institute of Technology | Llama3-8B |
| DLAMA | University of Edinburgh | arBERT |
| PubMedQA | University of Pittsburgh | GPT-4 (Medprompt) |
| FinDER | Yonsei University | Claude-3.7-Sonnet |
| HaluBench | Stanford University | LYNX (70B) |
| BizFinBench | Harbin Institute of Technology | ChatGPT-o3 |
| QuALITY | New York University | Human annotators |
| Chinese SimpleQA | Taobao & Tmall Group of Alibaba | o1-preview |
| KGQuiz | University of Washington | Davinci (text-davinci-003) |
| Wikidata5M | Tsinghua University | KEPLER-Cond |
| FinEval | Shanghai University of Finance and Economics | Claude 3.5-Sonnet |
| FaithJudge | University of Waterloo | gemini-2.5-flash |
| STaRK | Stanford University | Claude3 Reranker |
| Pinocchio | University of Cambridge | GPT-3.5-Turbo |
| MedQA-CS | Emory University | Claude 3.5 Sonnet |
| KGQAGen-10k | Rensselaer Polytechnic Institute | GPT-4o (w/ SP) |
| CRUD-RAG | University of Science and Technology of China | GPT-4o |
| MedBrowseComp | Universitat Pompeu Fabra | O3 deepresearch |
| LexRAG | Tsinghua University | Qwen-2.5-72B |
| C-Eval | Shanghai Jiao Tong University | 海信星海 |
| XRAG | Amazon AGI | GPT-4o |
| SecKnowledge-Eval | IBM Research | CyberPal.AI-Mistral |
| SimpleVQA | Beihang University | Gemini-2.0-flash |
| INFOSEEK | Georgia Institute of Technology | CLIP → FiD |
| ICD-Bench | Princeton University | o4-mini |
| A-OKVQA | Carnegie Mellon University | GPV-2 |
| RAG-QA Arena | University of California, Santa Barbara | GPT-4O (without CoT) |
| CommonsenseQA | Allen Institute for Artificial Intelligence | Human |
| VERISCORE | UMass Amherst | GPT-4o |
| MLaKE | Chinese Academy of Sciences | Qwen1.5-7B-Chat |
| FACT-AUDIT | National University of Singapore | GPT-4o |
| RAG-Check | University of Maryland | GPT-4o |
| LawBench | Shanghai AI Laboratory | GPT-4 |
| TheoremQA | University of Waterloo | GPT-4 |
| BoolQ | University of Washington | RankVicuna |
| ChineseEcomQA | Taobao & Tmall Group of Alibaba | DeepSeek-R1 |
| WikiBigEdit | Helmholtz Munich | mistral-7b |
| MedBench | Shanghai Artificial Intelligence Laboratory | GPT-4 |
| MedRGB | University of Oregon | GPT-3.5 |
| ORQA | University of Toronto | Llama3.1-405B-I |
| ReasonVQA | Technical University of Berlin | GPT-4o |
| ToolQA | Georgia Institute of Technology | ReAct (GPT-3.5) |
| HistBench | Fudan University | HistAgent (gpt-4o) |
| FactBench | University of Michigan | GPT-4o |
| KG-LLM-Bench | University of Southern California | claude-3.5-sonnet-v2 |
| SecBench | The Hong Kong Polytechnic University | o1-preview |
| ScienceMeter | University of Washington | OLMO2-32B-INSTRUCT |
| Chinese SafetyQA | Alibaba Group | o1-preview |
| BMMR | Shanghai AI Laboratory | Gemini-2.5-pro |
| DomainRAG | Renmin University of China | Baichuan2-33B-32k |
| SciKnowEval | Zhejiang University | o4-mini |
| WildHallucinations | University of Washington | GPT-4o |
| OKGQA | National University of Singapore | GPT-4o (CoT+SC) |
| GAOKAO-Bench | Fudan University | GPT-4-0314 |
| CFinBench | Huawei Noah’s Ark Lab | Yi1.5-34B |
| Continual Knowledge Learning | KAIST | T5-Modular |
| Hallucinations Leaderboard | University College London | OpenHermes-2.5-Mistral-7B |
| WikiContradict | IBM Research – Thomas J. Watson Research Center | Mistral-7b-inst |
| ICR2 | University of Edinburgh | GPT-4-turbo |
| CFLUE | Alibaba Group | GPT-4 |
| FELM | Carnegie Mellon University | GPT-4 |
| TeleQnA | The Chinese University of Hong Kong | GPT-4 |
| NEPAQuAD | Pacific Northwest National Laboratory | Gemini 1.5 Pro |
| YESciEval | Leibniz Universität Hannover | LLaMA-3.1-8B (SFT+RL adversarial) |
| OpenFactCheck | Monash University | GPT-4 |
| KoLA | Tsinghua University | GPT-4 |
| ECHOQA | University of California, Santa Barbara | OpenAI o1 |
| SciFact | University of Washington | Oracle Rationale |
| MMKE-Bench | University of Science and Technology of China | LLaVA-1.5 (IKE) |
| SUBARU | Ant Group | Llama3.1-8B-Instruct |
| ArabicMMLU | New York University | GPT-4 |
| ChineseSimpleVQA | Alibaba Group | o1-preview (0901) |
| PubHealthBench | UK Health Security Agency | GPT-4.5 |
| EWoK | Harvard University | Human |
| BMIKE-53 | University of Oxford | Llama 3.1 8B |
| GuessArena | Renmin University of China | OpenAI-o1 |
| MILU | Indian Institute of Technology Madras | GPT-4o |
| MEMERAG | Amazon | Qwen 2.5 32B |
| FRANK | Carnegie Mellon University | BERTScore Precision |
| HalluQA | Shanghai AI Laboratory | ERNIE-Bot |
| KorNAT | KAIST | HyperCLOVA X |
| InfiBench | University of Illinois at Urbana-Champaign | GPT-4o-0125 |
| FACTRBENCH | University of Michigan | GPT4o |
| SeaExam and SeaBench | Nanyang Technological University | Gemma-2-9b-it |
| CAQA | Monash University | Mistral-v0.3 (7B) |
| ORD-QA | Shanghai AI Laboratory | RAG-EDA (ours) |
| OpenMedQA | Barcelona Supercomputing Center | DeepSeek-R1 with CR |
| MIRAGE-Bench | University of Waterloo | GPT-4o |
| UniKnow | Seoul National University | QWEN-14B |
| FeTaQA | Yale University | T5-large |
| AA-Omniscience | Artificial Analysis | Gemini 3 Pro Preview |
| EverGreenQA | HSE University | EG-E5 |
| Head-to-Tail | Meta Reality Labs | GPT-4 |
| KCIF | IBM Research AI | GPT-4o-2024-08-06 |
| DefAn | The University of Western Australia | GPT-3.5 |
| TEMPREASON | Alibaba Group | TempT5 (T5-base) |
| WixQA | Wix.com | GPT-4o |
| HAE-RAE Bench | OneLineAI | GPT-4 |
| ProfBench | NVIDIA | GPT-5 (high) |
| ContextualBench | Salesforce AI Research | SFR-RAG-9B |
| MULTI | Shanghai Jiao Tong University | expert |
| DebateQA | University of Cambridge | GPT-4o |
| ZNO-Eval | Odesа Polytechnic National University | GPT-4o |
| CPsyExam | South China University of Technology | GPT-4 |
| EVOUNA | Northeastern University | Another Human |
| ComprehendEdit | Northwestern Polytechnical University | BLIP-2 OPT |
| CliMedBench | Shanghai Jiao Tong University | ERNIE-Bot |
| ATEB | Google DeepMind | Gecko |
| GEOHALUBENCH | Tsinghua University | Gemini-2.0-flash |
| FACTOR | AI21 Labs | OPT-66B |
| MKQA | Apple | XLM-R (Gold Passages, Translate-Train) |
| InsQABench | Fudan University | GLM4 (Fine-tuned) + RAG-ReAct 9B |
| VND-Bench | Visa | GPT-4 |
| ArabLegalEval | KAUST | GPT-4o |
| Question Answering with Conflicting Contexts | University of Pennsylvania | Phi-3 Medium (finetuned) |
| RoleEval | Tianjin University | GPT-4-0613 |
| FinCDM | Wuhan University | GLM-4 |
| CHARM | Wuhan University | Qwen-72B |
| FAMMA | Ant Group | PoT + GPT-o1 |
| FACTORY | Meta | Qwen3 |
| MMLU-Pro+ | Autodesk AI Research | O1-preview |
| EvoWiki | Fudan University | SFT + Open-book (Mistral-7B-Instruct-v0.3) |
| SciRerankBench | Chinese Academy of Sciences | MXBAI |
| CLIcK | KAIST | Claude 2 |
| MedThink-Bench | University of Georgia | MedGemma-27B |
| DRAGON | HSE | Qwen 2.5 32b Instruct |
| BioKGBench | Zhejiang University | GPT-4 |
| CREAK | The University of Texas at Austin | T5-3B (In-Domain) |
| KnowMT-Bench | HKUST | GPT-4o |
| KnowShiftQA | the University of Tokyo | o1-preview |
| ArXivBench | UCLA | Claude-3.5-sonnet |
| SportQA | University of California, Santa Barbara | GPT-4(5S,CoT) |
| TUMLU | KAIST | Claude 3.5 Sonnet |
| MISBENCH | The Hong Kong University of Science and Technology (Guangzhou) | Qwen2.5-14B |
| KaLMA | University of California, Santa Barbara | GPT-4 (temp=0.5) |
| LegalBench.PT | Instituto Superior Técnico | GPT-4o |
| OneEval | Monash University | o3 |
| X-FaKT | Harvard University | Llama-3-70B |
| MaScQA | Indian Institute of Technology Delhi | GPT-4-CoT |
| AICrypto | Chinese Academy of Sciences | human |
| KG-FPQ | Beijing Jiaotong University | Llama2-13B-Chat |
| K-QA | The Hebrew University of Jerusalem | GPT-4+ICL |
| BnMMLU | University of Malaya | gemini-2.0-flash |
| HaystackCraft | National University of Singapore | GPT-5 |
| Pedagogy Benchmark | Fab Inc. | Gemini-2.5 Pro |
| RARE | Carnegie Mellon University | Llama-3.2-90B-Vision-Instruct |
| FAC²E | Université de Montréal | GPT-4 |
| SciTabQA | Indian Institute of Technology Gandhinagar | OmniTab |
| KoNET | NAVER Cloud AI | gpt-4o-2024-05-13 |
| KoLasSimpleQA | Shanghai Artificial Intelligence Laboratory | GPT-4o |
| EMPEC | The University of Manchester | GPT-4 |
| RoMQA | University of Washington | seq2seq+retrieval |
| EvolvingQA | KAIST | DPR |
| CAMB | Georgia Institute of Technology | Qwen3-235B-A22B-Instruct |
| Fakepedia | EPFL | Mistral-7B-Instruct-v0.1 |
| PROBELM | University of Cambridge | PYTHIA-2.8B |
| X-FACT | University of Utah | XLM-R large |
| SummEdits | Salesforce AI | Human Performance |
| GTSQA | Graphcore Research | SubgraphRAG (200) |
| TrendFact | Northeastern University | FactISR(QwQ-32B) |
| DFIR-Metric | University of Oslo | GPT-4.1 |
| CUB | University of Copenhagen | COMMAND A |
| FactIR | Delft University of Technology | Snowflake-arctic-embed-s |
| RAG Hallucination Benchmark | San Jose State University | TLM |
| MedREQAL | Technical University of Munich | Mixtral |
| ScholarBench | KAIST | o3-mini |
| Futurepedia | Alibaba Group | Qwen2-72B-Instruct |
| ParamBench | Indian Institute of Management Indore | Llama-3.3-70B |
| MultiHoax | Politecnico di Milano | Gemini-2.0-pro-exp |
| AncientDoc | Fudan University | Gemini2.5-Pro |
| KBL | Hanyang University | Mistral-7B-v0.2-Instruct |
| NoMIRACL | University of Waterloo | Mistral-7B |
| SciEx | Karlsruhe Institute of Technology | Claude-3-opus-20240229 |
| QuanTemp | Delft University of Technology | FinQA-Roberta-Large |
| DQABench | Tsinghua University | GPT-4 |
| StructFact | The Hong Kong University of Science and Technology (Guangzhou) | GPT-4o-mini |
| QualBench | Hong Kong Polytechnic University | Qwen2.5-7b-instruct |
| PediaBench | East China Normal University | Qwen-MAX |
| ALCUNA | Peking University | gpt-4o-2024-04-09 |
| KQA Pro | Huawei Noah’s Ark Lab | BART KoPL |
| NEWTON | University of Washington | GPT-4 |
| SECQUE | Microsoft | GPT-4o |
| GaRAGe | Amazon | Nova Pro |
| DocTER | Xiamen University | ChatGPT |
| CROLIN-MQUAKE | Google DeepMind | GPT-3.5-turbo-instruct |
| EarthSE | Shanghai Artificial Intelligence Laboratory | Gemini-2.5 |
| COMPKE | South China University of Technology | GPT-3.5-TURBO |
| Xiezhi | Fudan University | GPT-4 |
| KFinEval-Pilot | KAIST | gpt-o1 |
| Explain-Query-Test | Autodesk Research | Sonnet-3.5 |
| Temporally Consistent Factuality Probe | Indian Institute of Technology Delhi | GPT-J [6B] |
| AC-EVAL | Beijing University of Posts and Telecommunications | ERNIE-Bot 4.0 |
| MedicineQA | Shanghai AI Laboratory | RagPULSE (20B) |
| CONNER | National University of Singapore | ChatGPT (text-davinci-003) (Zero-shot) |
| BertaQA | University of the Basque Country (UPV/EHU) | GPT-4 Turbo |
| Uhura | Imperial College London | GPT-o1-preview |
| CLR-Fact | HKUST | GPT-4o |
| INSEva | Fudan University | Doubao-1.5 |
| FoundaBench | Shanghai Artificial Intelligence Laboratory | InternLM-123B |
| LLMzSzŁ | Adam Mickiewicz University | Mistral-Large-Instruct-2407 |
| ObfusQA | Indian Institute of Technology Patna | DeepSeek R1 |
| EnviroExam | Harbin Institute of Technology | deepseek-67b-chat |
| Polish Cultural Competency | National Information Processing Institute | Gemini-Exp-1206 |
| NetEval | Tsinghua University | gpt-4 |
| EpiK-Eval | Université de Montréal | Flan-T5-XL |
| AgriEval | Harbin Institute of Technology | Qwen-Plus |
| ArxEval | Vellore Institute of Technology | Qwen-2.5 7B |
| ComparisonQA | HKUST | GPT-2 |
| LegalScore | Universidade Estadual Paulista “Júlio de Mesquita Filho” | Claude Sonnet 3.5 |
| ECKGBench | City University of Hong Kong | Qwen2-max |
| SeaEval | Nanyang Technological University | GPT-4 (gpt-4-0613) |
| HSSBench | Chinese Academy of Sciences | Human |
| SKA-Bench | Ant Group | TableGPT-2 |
| PRGB Benchmark | Ant Group | Gemini-2.5-pro-preview |
| MultiHal | Aalborg University | Gemini 2.0 Flash |
| TransportationGames | Beijing Jiaotong University | GPT-4 |
| MultifacetEval | Tsinghua University | Gemini-pro |
| CoLoTa | University of Toronto | OpenAI-o1 |
| OntoURL | Leiden University | LLaMA3.1-8B |
| DailyQA | Renmin University of China | Qwen2.5-72B-Instruct |
| MultiNativQA | Qatar Computing Research Institute | Gemini-1.5 Flash |
| PretexEval | Tsinghua University | GPT-4o |
| LPFQA | University of Michigan | GPT-5 |
| NorQA | University of Oslo | NorMistral-7B-warm |
| Temporal Wiki | University of Tübingen | Llama 3.1 70B |
| KaRR | University of California, Santa Barbara | OPT (175B) |
| FinEval-KR | Shanghai University of Finance and Economics | DeepSeek-R1 |
| BELIEF | the University of Tokyo | Llama3-70B |
| MAGIC | Hanyang University | GPT-4o-mini |
| LHMKE | Tianjin University | Baichuan2-13B-Chat |
| TAXI | Arizona State University | Human |
| M3KE | Tianjin University | GPT-4 |
| EESE | Shanghai Artificial Intelligence Laboratory | GPT-5 |
| LLM-KG-Bench | Leipzig University | LLMKE |
| BaRDa | Allen Institute for AI | GPT-4 |
| CPSDBench | People’s Public Security University of China | GPT-4 |
| TuringQ | Sharif University of Technology | GPT-4 |
| HEAD-QA | Universidade da Coruña | Liu et al. (2020) |
| GeoGLUE | Alibaba Group | Nezha |
| FACT-BENCH | AWS AI Labs | GPT-4 |
| MedMKEB | Peking University | LLaVA-Med |
| Alvorada-Bench | Inteli | O3 Pro |
| EFO-1-QA | Tsinghua University | LogicE |
| CG-Eval | Besteasy Language Technology Co., Ltd | GPT-4 |
| BeerQA | Stanford University | IRRR (SQuAD + HotpotQA) |
| Down and Across | University of Massachusetts Lowell | RAG wiki |
| COPEN | Huawei Noah’s Ark Lab | Human |
| QuantumBench | NVIDIA | GPT-5-high |
| Knowledge Crosswords | University of Washington | VERIFY-ALL (GPT-4) |
| NuclearQA | Pacific Northwest National Laboratory | Llama 2 |
| ThaiCLI | Chulalongkorn University | GPT-4o |
| MQA-AEVAL | Shenzhen University | GPT-3.5-TURBO-INSTRUCT |
| CUS-QA | University of Oslo | Llama-3.3-70B-Instruct |
| SANSKRITI | Indian Institute of Technology Patna | GPT-4o |
| MINED | University of Science and Technology of China | Gemini-2.5-Pro |
| M-QALM | University of Manchester | Flan-T5 (11B) |
| CoDEx | University of Michigan | ConvE |
| ZhuJiu | Chinese Academy of Sciences | GPT-3.5-turbo |
| KoSimpleQA | Seoul National University | HCX SEED 14B |
| KEO QA Benchmark | University of Notre Dame | gemma-3-it |
| XL-BEL | University of Cambridge | XLMR + SAP all syn |
| Entity Cloze by Date | The University of Texas at Austin | T5-large |
| SLAQ | University of Hamburg | Gemma-3 1B |
| Self-Diagnostic Atomic Knowledge | Shenzhen Research Institute of Big Data | GPT-4 |
| Multilingual Compositional Wikidata Questions | University of Copenhagen | mT5-base+RIR |
| OKBench | University of Chicago | GPT-4o |
| SCiPS-QA | Indian Institute of Technology | meta-llama-3-70B |
| CK-Arena | Beijing Normal University | DeepSeek-V3 |
| KG Attributes Benchmark | Megagon Labs | GPT-4 |
| ONTOLAMA | University of Oslo | RoBERTa-large-pm-m3-voc |
| KULTURE Bench | Yonsei University | Human |
| BIOLAMA | Korea University | Bio-LM |
| LoFTI | Indian Institute of Technology, Bombay | GPT-4 |
| BEAR | Humboldt-Universität zu Berlin | Meta-Llama-3-8B |
| FinLFQA | New York University | GPT-4o |
| LiveSearchBench | Shanghai AI Laboratory | gpt-5 |
| Hakka Benchmark | National United University | Llama 3.1 with RAG |
| QUENCH | IIIT-Delhi | GPT-4-Turbo |
| Japanese Financial Benchmark | Preferred Networks. Inc. | Claude 3.5 Sonnet |
| RAmBLA | Imperial College London | GPT-3.5 |
| MuLan | Google DeepMind | Alpaca |
| RealFactBench | University College London | Claude-3.7-Sonnet |
| YpathR | Tsinghua University Shenzhen International Graduate School | YpathRAG |
| InsCoQA | University of Science and Technology of China | GPT-4 |
| GeoSQA | Nanjing University | PMI |
| FailureSensorIQ | IBM T.J. Watson Research Center | o1 |
| MedFact | Xunfei Healthcare Technology Co., Ltd. | XiaoYi |
| CheckThat! 2020 | Qatar University | Buster.AI |
| ClimateEval | Uppsala University | Mistral 24B |
| URDUFACTBENCH | MBZUAI | GPT-4O |
| Ambiguous Entity Retrieval | University of California, Irvine | Bootleg |
| MultiWikiQA | Alexandra Institute | Mistral-Small-3.1-24B-Instruct-2503 |
| MedBench-IT | University of Graz | DeepSeek-R1 |
| CultSportQA | Universitas Gadjah Mada | GPT-4o |
| X-FACTR | Carnegie Mellon University | M-BERT |
| LM-PUB-QUIZ | Humboldt-Universität zu Berlin | Meta-Llama-3-8B |
| Materials Knowledge Benchmark | Technische Universität Dresden | phi-4 |
| ChnEval | Beijing Normal University | RoBERTa-wwm-ext |
| DentalBench | Zhejiang University | GPT-4o |
| ReFACT | University of Potsdam | GPT-4o |
| MultiReQA | Carnegie Mellon University | USE-QA (fine-tuned) |
| TempQA-WD | IBM Research | SYGMA |
| MediQAl | CNRS | o3 |
| DEEPAMBIGQA | UCSB | GPT-5 |
| FinS-Pilot | Renmin University of China | Xiaofa-1.0 |
| ConvQuestions | MPI for Informatics | Oracle + Convex |
| evolveQA | Purdue University | GPT-5-mini |
| Swedish Facts | Linköping University | gemma-3-27b |
| Chinese Commonsense Multi-hop Reasoning | ByteDance | Gemini-2.5-Pro |
| Nunchi-Bench | EPFL | gemini-2.5-pro-preview |
| Agri-Query | Technical University of Munich (TUM) | Gemini 2.5 Flash |
| FATHOMS-RAG | Louisiana State University | claude-sonnet-4 |
| ADAM | Virginia Tech | Gemini Flash 2.5 |
| SinhalaMMLU | Nara Institute of Science and Technology | Claude 3.5 Sonnet |
| XLQA | Korea University | Oracle LM (GPT-4.1) |
| TripJudge | Leiden University | Ensemble |
| RoBiologyDataChoiceQA | University of Bucharest | gemini-2.0-flash |
| MedLAMA | University of Cambridge | SapBERT |
| ArcMMLU | Wuhan University | GPT-4 (gpt-4-0613) |
| KEO | University of Notre Dame | gemma-3-it |
| EFO_k-CQA | Tsinghua University | CQD |
| FactChecker | The Chinese University of Hong Kong | GPT-4 |
| ECLeKTic | Google DeepMind | Gemini 2.0 Pro |
| KACC | Tsinghua University | AttH |
| ParRoT | Allen Institute for AI | Macaw-11B |
| PalmX-GC | Qatar Computing Research Institute | GPT-4.1 |
| CIKQA | University of Washington | G2T (Bian et al., 2021) |
| TEXTWORLDSQA | Carnegie Mellon University | DrQA-M |
| KMIR | Renmin University of China | RoBERTa |