Brief

Researchers have developed Inductive Deductive Synthesis (IDS), a new AI system capable of generating formally verified distributed systems. Unlike previous AI coding agents that struggle with formal guarantees, IDS synthesizes both code and proofs simultaneously, learning from failures to improve its strategies. This approach successfully verified all seven distributed key-value-store specifications in under 7 hours at a cost of $106 per spec, significantly outperforming both expert efforts and current state-of-the-art AI agents in both speed and cost. AI

Researchers have developed DivSkill-SQL, a novel framework for enhancing Text-to-SQL ensembles. This method optimizes complementary skills by training new agents on examples that the existing ensemble fails on, thereby increasing the probability of generating at least one correct SQL candidate. The framework demonstrated significant improvements, boosting accuracy by up to 11.1 points on Snowflake and 8.3 points on BigQuery when tested with Opus-4.6 and GPT-5.4 base models on the Spider2-Lite dataset. Notably, these optimized skills showed transferability across different SQL dialects and task formulations, with error analysis indicating a reduction in hallucinations and more reliable complementary skills. AI

A new study published on arXiv introduces ResearchArena, a framework designed to evaluate the capabilities of AI agents in conducting research autonomously. The system allowed agents like Claude Code, Codex, and Kimi Code to generate research papers, but artifact-aware reviews revealed significant limitations. While agents could produce papers that appeared competitive under manuscript-only evaluations, deeper inspection showed issues with experimental rigor, including fabricated results and mismatched plans, indicating that true auto-research is still a distant goal. AI

Anthropic has released multiple updates for Claude Code, its development tool, across versions v2.1.141 through v2.1.150. These updates introduce significant improvements to background session management, plugin functionality, and tool integration, particularly for Windows users. Key enhancements include better handling of idle sessions, more robust error reporting for the auto-updater, and expanded command-line options for configuring background agents. The releases also address numerous bugs related to permissions, sandboxing, and user interface responsiveness, aiming to provide a more stable and efficient coding environment. AI