Google Research
The research division of Google that conducts fundamental and applied research in artificial intelligence and computer science. It develops methods, models and systems spanning machine learning, natural language processing, computer vision and related fields.
Core Facts
In 2023, Stanford University and Google Research jointly published the paper "Generative Agents: Interactive Simulacra of Human Behavior," deploying 25 LLM-based agents in a virtual town.
思维链（Chain-of-Thought）提示技术由Google Research于2022年提出
The concept of emergent abilities was systematically elaborated by the Google Research team (Wei et al.) in the 2022 paper Emergent Abilities of Large Language Models, which examined more than 200 tasks.
Timeline (last 90 days)
The Google Research team shared an experimental project at the Bilibili AI Summit that lets language-model-driven agents play NPCs in "Werewolf."
The iQuest series has completed over 400,000 tasks, covering 10 languages, and Google will also launch a teacher training module.
Google launched a neuroscience-focused lesson called "Polar Peaks" on its AI Quests learning platform.
Google Research reported a performance improvement of up to +11.5 points on the WorkBuddy-Bench benchmark.
Judging from Google Research's release channels, Harness2 is more likely a research-oriented prototype or conceptual framework rather than a ready-to-use product.
SimpleQA Verified was introduced by Google DeepMind and Google Research.
Google Research has open-sourced Harness2, a framework that gives AI Agents' "scaffolding" the ability to self-improve.
A Google research team open-sourced Harness2, an Apache-2.0 licensed framework.
Google Research and collaborators published a new paper proposing to replace the central orchestrator with sequential auctions to coordinate multi-agent LLM workflows, with workflows emerging from economic incentives rather than a predefined DAG.
Titans (Google Research) introduces a neural long-term memory module that updates memory based on surprise at test time.
17 more timeline events
All Facts (20)
In 2023, Stanford University and Google Research jointly published the paper "Generative Agents: Interactive Simulacra of Human Behavior," deploying 25 LLM-based agents in a virtual town.
80%Verified思维链（Chain-of-Thought）提示技术由Google Research于2022年提出
80%VerifiedThe concept of emergent abilities was systematically elaborated by the Google Research team (Wei et al.) in the 2022 paper Emergent Abilities of Large Language Models, which examined more than 200 tasks.
75%VerifiedSelf-Consistency was formally proposed by Google Research in the 2022 paper 'Self-Consistency Improves Chain of Thought Reasoning in Language Models.'
75%VerifiedGoogle Research has open-sourced Harness2, a framework that gives AI Agents' "scaffolding" the ability to self-improve.
65%UnverifiedIn a 2022 study, Google Research systematically summarized the concept of emergent capabilities, finding that abilities such as chain-of-thought reasoning and few-shot learning are absent in small models but suddenly emerge after the parameter count crosses a critical threshold.
60%UnverifiedSigLIP is a vision-language contrastive learning model proposed by Google Research that uses a sigmoid loss function to replace CLIP's softmax contrastive loss, eliminating the dependency on global negative samples.
60%UnverifiedxAI's technical team brings together researchers from top institutions including DeepMind, OpenAI, and Google Research.
60%UnverifiedThe iQuest series has completed over 400,000 tasks, covering 10 languages, and Google will also launch a teacher training module.
50%UnverifiedGoogle Research reported a performance improvement of up to +11.5 points on the WorkBuddy-Bench benchmark.
50%UnverifiedSimpleQA Verified was introduced by Google DeepMind and Google Research.
50%UnverifiedTitans (Google Research) introduces a neural long-term memory module that updates memory based on surprise at test time.
50%UnverifiedYossi Matias is the Vice President and Head of Google Research.
50%UnverifiedGoogle Research proposed the RRSI (Regularized Recursive Self-Improvement) framework for constraining the recursive self-improvement process of AI agents
50%UnverifiedJAX is developed by Google Research, and its core strengths lie in its functional programming style and the composability of jit, vmap, and grad.
50%UnverifiedData Commons was built by Google Research starting around 2018.
50%UnverifiedFreshQA is an evaluation benchmark released by Google Research in 2023, used to test the ability of language models and search systems to handle time-sensitive questions.
50%UnverifiedTimesFM 3.0 is a foundation model dedicated to time series forecasting, with zero-shot forecasting capability.
50%UnverifiedIn 2022, researchers from Google Research and Stanford systematically described the phenomenon of emergent abilities for the first time in a paper: when model scale crosses a certain threshold, performance on specific tasks jumps from near-random to significantly above baseline.
50%UnverifiedAccording to a Google Research paper, the pre-training data scale of TimesFM reaches approximately 100 billion time points.
50%