Qwen-3

The 'Qwen-3' dataset/benchmark is used to evaluate the performance of large language models (LLMs) across various tasks, including long-context understanding, code comprehension, and mathematical reasoning.

Papers using Qwen-3 (8)

Efficient Training-Free Multi-Token Prediction via Embedding-Space Probing2026DeLS-Spec: Decoupled Long-Short Contexts for Parallel Speculative Drafting2026BaseRT: Best-in-Class LLM Inference on Apple Silicon via Native Metal2026Fast Multi-dimensional Refusal Subspaces via RFM-AGOP2026Fast KVzip: Efficient and Accurate LLM Inference with Gated KV Eviction2026Didactic to Constructive: Turning Expert Solutions into Learnable Reasoning2026SCAN: Sparse Circuit Anchor Interpretable Neuron for Lifelong Knowledge Editing2026Garbage Attention in Large Language Models: BOS Sink Heads and Sink-aware Pruning2026