TechniqueRLHF / Alignment2 recent entries8 Apr 2026Improving the academic workflow: Introducing two AI agents for better figures and peer reviewGoogle Research introduced two AI agents to streamline academic workflows: **PaperVizAgent**, a visualizer agent for drawing academic figures, and **ScholarPeer**, a reviewer agent that automatica...→9 Apr 2026ConvApparel: Measuring and bridging the realism gap in user simulatorsConvApparel is a human-AI conversation dataset and comprehensive evaluation framework designed to quantify the 'realism gap' in LLM-based user simulators and improve the training of robust conversa...
TechniqueRAG1 recent entries5 Jun 2026Unlocking dependable responses with Gemini Enterprise Agent Platform’s Agentic RAGGoogle's RAG Engine securely connects private enterprise data to LLMs to improve answer accuracy and reduce hallucinations , making it a key component of the Gemini Enterprise Agent Platform for build
TechniqueAgents6 recent entries8 Apr 2026Improving the academic workflow: Introducing two AI agents for better figures and peer reviewGoogle Research introduced two AI agents to streamline academic workflows: **PaperVizAgent**, a visualizer agent for drawing academic figures, and **ScholarPeer**, a reviewer agent that automatica...→9 Apr 2026ConvApparel: Measuring and bridging the realism gap in user simulatorsConvApparel is a human-AI conversation dataset and comprehensive evaluation framework designed to quantify the 'realism gap' in LLM-based user simulators and improve the training of robust conversa...→16 Apr 2026Designing synthetic datasets for the real world: Mechanism design and reasoning from first principlesThis Google Research work presents guidelines for synthetic data mechanism design and provides insights into generating and evaluating synthetic data at scale. The research introduces a reasoning-driv→21 Apr 2026ReasoningBank: Enabling agents to learn from experienceReasoningBank is a memory framework that distills generalizable reasoning strategies from an agent's successful and failed experiences, which the agent retrieves at test time to inform interactions wh→28 May 2026A New Era of Innovation: Google Research at I/O 2026Google Research at I/O 2026 showcased new Gemini AI models including Gemini Omni, which can create content from any input starting with video, and Gemini 3.5 Flash, combining frontier intelligence wit→5 Jun 2026Unlocking dependable responses with Gemini Enterprise Agent Platform’s Agentic RAGGoogle's RAG Engine securely connects private enterprise data to LLMs to improve answer accuracy and reduce hallucinations , making it a key component of the Gemini Enterprise Agent Platform for build
TechniqueMultimodal2 recent entries16 Apr 2026Designing synthetic datasets for the real world: Mechanism design and reasoning from first principlesThis Google Research work presents guidelines for synthetic data mechanism design and provides insights into generating and evaluating synthetic data at scale. The research introduces a reasoning-driv→9 Jul 2026SensorFM: Towards a general intelligence and interface for wearable health dataSensorFM is a foundation model that analyzes multimodal wearable sensor signals from devices like smartwatches and fitness trackers to generate insights into health and activities. The model was train
TechniqueSafety1 recent entries5 Jun 2026Unlocking dependable responses with Gemini Enterprise Agent Platform’s Agentic RAGGoogle's RAG Engine securely connects private enterprise data to LLMs to improve answer accuracy and reduce hallucinations , making it a key component of the Gemini Enterprise Agent Platform for build