a1: Steep Test-time Scaling Law via Environment Augmented Generation
Preprint
Abstract
Large Language Models (LLMs) have made remarkable breakthroughs in reasoning, yet continue to struggle with hallucinations, logical errors, and inability to self-correct during complex multi-step tasks. Current approaches like chain-of-thought prompting offer limited reasoning capabilities that fail when precise step validation is required. We propose Environment Augmented Generation (EAG), a framework that enhances LLM reasoning through: (1) real-time environmental feedback validating each reasoning step, (2) dynamic branch exploration for investigating alternative solution paths when faced with errors, and (3) experience-based learning from successful reasoning trajectories. Unlike existing methods, EAG enables deliberate backtracking and strategic replanning through tight integration of execution feedback with branching exploration. Our a1-32B model achieves state-of-the-art performance among similar-sized models across all benchmarks, matching larger models like o1 on competition mathematics while outperforming comparable models by up to 24.4 percentage points. Analysis reveals EAG's distinctive scaling pattern: initial token investment in environment interaction yields substantial long-term performance dividends, with advantages amplifying proportionally to task complexity. EAG's theoretical framework demonstrates how environment interactivity and systematic branch exploration together establish a new paradigm for reliable machine reasoning, particularly for problems requiring precise multi-step calculation and logical verification.Keywords
All Related Versions
This publication has 26 references indexed in Scilit:
- From System 1 to System 2: A Survey of Reasoning Large Language ModelsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2025
- Mv-Math: Evaluating Multimodal Math Reasoning in Multi-Visual Contexts2025
- Neural Reasoning Networks: Efficient Interpretable Neural Networks with Automatic Textual ExplanationsProceedings of the AAAI Conference on Artificial Intelligence, 2025
- Graph of Thoughts: Solving Elaborate Problems with Large Language ModelsProceedings of the AAAI Conference on Artificial Intelligence, 2024
- From the Least to the Most: Building a Plug-and-Play Visual Reasoner via Data Synthesis2024
- SLANG: New Concept Comprehension of Large Language Models2024
- Large Language Models are Versatile Decomposers: Decomposing Evidence and Questions for Table-based Reasoning2023
- Code4Struct: Code Generation for Few-Shot Event Structure Prediction2023
- Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models2023
- Language Models of Code are Few-Shot Commonsense Learners2022