Tiny Model, Big Logic: Diversity-Driven Optimization Elicits Large-Model Reasoning Ability in VibeThinker-1.5B
[NeurIPS 2025] AGI-Elo: How Far Are We From Mastering A Task?
Benchmark methodology, task sets, and evaluation results for RA²R
“U.S. Provisional Patent Application No. 63/978,753, filed February 9, 2026” Aleph Generalizable Intelligence Convergence Engine (AGICE) - Evidence-Native, Policy-Governed Reasoning with Rollback and Geometrical Multi-Dimensional Transformation (GMDT) via the Complex Operator (−i)
LoRA fine-tuning evaluation pipeline for code generation models on LiveCodeBench benchmark
Interactive benchmark for evaluating LLMs on clarifying ambiguous code requirements. Paper: arXiv:2607.00711
Reproducible systems + methodology case study benchmarking three open-weight MoE coding models (gpt-oss-120b, Qwen3-Coder-30B, Nemotron-3-Super-120B) locally on one DGX Spark (GB10) via vLLM / TensorRT-LLM + OpenCode. Serving feasibility, task success, and efficiency across three layers.
Executable-code benchmark and verifier-guided post-training sandbox in the L20 project family.
Add a description, image, and links to the livecodebench topic page so that developers can more easily learn about it.
To associate your repository with the livecodebench topic, visit your repo's landing page and select "manage topics."