Code Generation
cs.CL · 8 papers
Reflexion: Language Agents with Verbal Reinforcement Learning
Pass@1: 91.0%Noah Shinn, Federico Cassano +4
Large language models (LLMs) have been increasingly used to interact with external environments (e.g., games, compilers, APIs) as goal-driven agents. However, it remains…
Voyager: An Open-Ended Embodied Agent with LLMs
Minecraft (MineDojo): 63Guanzhi Wang, Yuqi Xie +6
We introduce Voyager, the first LLM-powered embodied lifelong learning agent in Minecraft that continuously explores the world, acquires diverse skills, and makes novel…
Qwen3 Technical Report: Unified Reasoning & Dynamic Thinking Budgets
AIME'24: 85.7An Yang, Anfeng Li +58
In this work, we present Qwen3, the latest version of the Qwen model family. Qwen3 comprises a series of large language models (LLMs) designed to advance performance, efficiency,…
DeepSeek-V3 Technical Report
MMLU-Pro: 75.9%DeepSeek-AI, Aixin Liu +198
We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. To achieve efficient inference and…
Textbooks Are All You Need
Suriya Gunasekar, Yi Zhang +17
We introduce phi-1, a new large language model for code, with significantly smaller size than competing models: phi-1 is a Transformer-based model with 1.3B parameters, trained…
Gemini: A Family of Highly Capable Multimodal Models — Deep Dive
MMLU: 90.04%Gemini Team, Rohan Anil +1349
This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family…
PaLM: Scaling Language Modeling with Pathways
Model FLOPs Utilization (MFU): 46.2%Aakanksha Chowdhery, Sharan Narang +65
Large language models have been shown to achieve remarkable performance across a variety of natural language tasks using few-shot learning, which drastically reduces the number…
LLaMA: Open and Efficient Foundation Language Models
MMLU: 63.4%Hugo Touvron, Thibaut Lavril +12
We introduce LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. We train our models on trillions of tokens, and show that it is possible to…