The field of smart contract generation is rapidly advancing, with a focus on developing reliable and efficient tools for generating high-quality contracts. Recent research has highlighted the challenges of evaluating the effectiveness of large language models (LLMs) in generating Solidity code, particularly in real-world contract development scenarios. To address this, new benchmarks and evaluation frameworks are being developed to assess the capabilities of LLMs in generating functional and gas-efficient contracts. Additionally, innovative approaches such as curriculum-guided reinforcement learning and agentic specification generation are being explored to improve the synthesis of reliable and economically sustainable smart contracts. Notable papers in this area include SolContractEval, which introduces a contract-level benchmark for Solidity code generation, and Curriculum-Guided Reinforcement Learning for Synthesizing Gas-Efficient Financial Derivatives Contracts, which presents a viable methodology for automated synthesis of reliable and economically sustainable smart contracts. Another noteworthy paper is Agentic Specification Generator for Move Programs, which demonstrates the effectiveness of LLM-based specification generation for emerging languages like Move.