Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2602 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Competitive Programming | LeetCode Weekly & Biweekly Contests OOD Generalization (Apr 25–May 31, 2026) (test) | 14 | Jun 16, 2026 | ||
| Video Grounding | UAV-SAVG (test) | 14 | Jun 16, 2026 | ||
| Multi-hop QA | HotpotQA (test) | 14 | Jun 16, 2026 | ||
| Multi-hop QA | NQ (test) | 14 | Jun 16, 2026 | ||
| Fingerprint Spoofing | MET on UltraChat (test) | 14 | Jun 16, 2026 | ||
| Fingerprint Spoofing | LLM-idio on UltraChat (test) | 14 | Jun 16, 2026 | ||
| Fingerprint Spoofing | LLMmap (test) | 14 | Jun 16, 2026 | ||
| Cross-family weak-to-weak spoofing | LLMmap Gemma-2B to Qwen2-1.5B | 14 | Jun 16, 2026 | ||
| Image Quality Assessment | Dehaze | 14 | Jun 16, 2026 | ||
| Scenario classification | Customer-service conversations | 14 | Jun 16, 2026 | ||
| Tactic classification | Customer-service conversations | 14 | Jun 16, 2026 | ||
| Question Answering | Story | 14 | Jun 16, 2026 | ||
| Robot Manipulation | LIBERO (sub-suites) | 14 | Jun 16, 2026 | ||
| Causal Reasoning | e-CARE n = 200 | 14 | Jun 16, 2026 | ||
| Federated Class-Incremental Learning | Cora | 14 | Jun 16, 2026 | ||
| Harmlessness Evaluation | DNA (test) | 14 | Jun 16, 2026 | ||
| Helpfulness evaluation | LINGUASAFE | 14 | Jun 16, 2026 | ||
| Helpfulness evaluation | DNA | 14 | Jun 16, 2026 | ||
| Scanpath Generation | MIT-FV MIT1003 | 14 | Jun 16, 2026 | ||
| Scanpath Generation | COCO-FV (COCO-FreeView) | 14 | Jun 16, 2026 | ||
| Conversational function-calling | τ-bench retail (public leaderboard) | 14 | Jun 16, 2026 | ||
| Mathematical Proof Reasoning | IMO-ProofBench Advanced (full set) | 14 | Jun 16, 2026 | ||
| Code Generation | HumanEval | 14 | Jun 16, 2026 | ||
| Science Question Answering | ScienceQA | 14 | Jun 16, 2026 | ||
| Multitask Language Understanding | MMLU | 14 | Jun 16, 2026 |