اخبار هوش مصنوعی و ارزیابی نوآوری علمی: گزارش جدید AI2 در بستر علمی کاوش پژوهشی به‌روز در AI2 که بررسی می‌کند LLMها در ارزیابی نوآوری‌های علمی به چه اندازه قابل اعتماد هستند و بنچمارک جدیدی ارائه می‌دهد. ۵ آبان ۱۴۰۵ ۲ دقیقه سیاست و قانون Embedded AI evaluators won’t be an easy solution, Hugging Face’s policy chief says AI labs have staked their reputations on model benchmarking — the supposedly neutral and accurate system of evaluations that are designed to measure whether o ۲۸ مهر ۱۴۰۵ ۲ دقیقه اخبار The final Disrupt Stage lineup: Three days of conversations you won’t hear anywhere outside of TechCrunch Disrupt 2026 The full TechCrunch Disrupt Stage lineup revealed, featuring Max Hodak, Mark Wahlberg, Benchmark partners, and more. Register now to save up to $100 on your p ۲۱ مهر ۱۴۰۵ ۲ دقیقه اخبار The final Disrupt Stage lineup: Three days of conversations you won’t hear anywhere outside of TechCrunch Disrupt 2026 The full TechCrunch Disrupt Stage lineup revealed, featuring Max Hodak, Mark Wahlberg, Benchmark partners, and more. Register now to save up to $100 on your p ۱۳ مهر ۱۴۰۵ ۲ دقیقه اخبار Introducing MentalHealthBench MentalHealthBench is an expert-informed benchmark for evaluating helpful and safe AI responses across realistic mental health conversations. ۱ مهر ۱۴۰۵ ۲ دقیقه پژوهش How UK AISI and EvalEval Are Making Benchmark Results Reproducible We’re on a journey to advance and democratize artificial intelligence through open source and open science. ۳۱ شهریور ۱۴۰۵ ۲ دقیقه کسب‌وکار Where will the next breakout startup come from? Benchmark’s full partnership weighs in at TechCrunch Disrupt 2026 Where will the next breakout startup come from? Benchmark’s full partnership weighs in on the main stage at TechCrunch Disrupt 2026. Save up to $200 before Se ۳۰ شهریور ۱۴۰۵ ۲ دقیقه اخبار Vals, backed by Andreessen Horowitz, is looking to become the gold standard for AI benchmarking Vals AI is hoping to make AI benchmarking a more neutral and trustworthy resource in a world increasingly inundated by AI models. ۲۸ شهریور ۱۴۰۵ ۲ دقیقه مدل‌ها Separating signal from noise in coding evaluations A new analysis from OpenAI reveals issues in SWE-Bench Pro, a popular coding benchmark, raising concerns about reliability and accuracy in evaluating AI model ۱۷ تیر ۱۴۰۵ ۲ دقیقه پژوهش ScarfBench: Benchmarking AI Agents for Enterprise Java Framework Migration A Blog post by IBM Research on Hugging Face ۹ تیر ۱۴۰۵ ۲ دقیقه