LongBench v-2

LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks 🌐 Project Page: https://longbench2.github.io 💻 Github Repo: https://github.com/THUDM/LongBench 📚 Arxiv Paper: https://arxiv.org/abs/2412.15204 LongBench v2 is designed to assess the ability of LLMs to handle long-context pr

Papers using LongBench v-2 (5)

LongAct: Harnessing Intrinsic Activation Patterns for Long-Context Reinforcement Learning2026A Decomposition Perspective to Long-context Reasoning for LLMs2026Overflow Prevention Enhances Long-Context Recurrent LLMs2025LongMagpie: A Self-synthesis Method for Generating Large-scale Long-context Instructions2025Beyond RAG: Task-Aware KV Cache Compression for Comprehensive Knowledge Reasoning2025