跳到正文- Source: XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in LLMs
- Source: Why we no longer evaluate SWE-bench Verified (OpenAI, 2026)
- Source: Wei et al. 2025: BrowseComp
- Source: ViDoRe V3 (Loison et al., ACL 2026)
- Source: The TREC-8 Question Answering Track Report (Voorhees)
- Source: Terminal-Bench (Merrill et al., 2026)
- Source: SWE-Bench Pro (Deng et al., Scale AI, 2025)
- Source: Spider 2.0 (Lei et al., ICLR 2025) and Leaderboard
- Source: Spider (Yu et al., EMNLP 2018)
- Source: RULER: What's the Real Context Size of Your Long-Context Language Models? (NVIDIA)
- Source: Raising the bar on SWE-bench Verified with Claude 3.5 Sonnet (Anthropic, 2025)
- Source: Pervasive Annotation Errors Break Text-to-SQL Benchmarks (Jin et al., 2026)
- Source: OSWorld Benchmark
- Source: OmniDocBench (Ouyang et al., CVPR 2025) and Leaderboard
- Source: NoLiMa: Long-Context Evaluation Beyond Literal Matching (ICML 2025)
- Source: MTEB: Massive Text Embedding Benchmark (Muennighoff et al.)
- Source: MMTEB: Massive Multilingual Text Embedding Benchmark (Enevoldsen et al.)
- Source: Mem0, the token-efficient memory algorithm (Mem0, 2026)
- Source: LongMemEval (Wu et al., 2024)
- Source: Lies, Damn Lies, & Statistics: Is Mem0 Really SOTA in Agent Memory? (Zep, 2025)
- Source: JSONSchemaBench (Generating Structured Outputs from Language Models)
- Source: BIRD Benchmark (Li et al., NeurIPS 2023) and Leaderboard
- Source: Benchmarking AI Agent Memory: Is a Filesystem All You Need? (Letta, 2025)
- Source: BEIR: A Heterogeneous Benchmark for Zero-shot Evaluation of IR Models (Thakur et al.)
- Source: AgentDojo
- Coding Agents