Skip to content
Tagged articles

Google Research

9 articles · Page 1 of 1
Machine Heart
Machine Heart
Oct 31, 2026 · Artificial Intelligence

Google's CLASHarness Benchmarks 15 LLMs on Multi-Agent Conflict in Computer Use

Google's CLASHarness evaluates 15 leading LLMs on 600 multi-agent conflict tasks with 4K visual evidence, revealing top models achieve only 25–28% end-to-end pass rates despite 55–67% per-step scores, and that visual verification is critical for detecting teammate deception.

CLASHarnessGoogle ResearchLLM benchmarking
0 likes · 13 min read
Google's CLASHarness Benchmarks 15 LLMs on Multi-Agent Conflict in Computer Use
Machine Learning Algorithms & Natural Language Processing
Machine Learning Algorithms & Natural Language Processing
Oct 25, 2026 · Artificial Intelligence

Google Open-Sources Harness²: Dual-Layer RSI Framework Boosts Agent Performance Without Weight Updates

Google's Harness² framework introduces a dual-layer recursive self-improvement architecture that optimizes AI agent performance by iteratively refining eight editable components — system prompt, guardrails, memory, tool guidance, sub-agents, skills, scripts, and plugins — achieving up to +11.5 points on WorkBuddy-Bench and +11.8% on LAB with Gemini 3.5 Flash and GLM-5 in single-step optimization.

AI agentsGLM-5Gemini 3.5 Flash
0 likes · 11 min read
Google Open-Sources Harness²: Dual-Layer RSI Framework Boosts Agent Performance Without Weight Updates
TonyBai
TonyBai
Oct 19, 2026 · Artificial Intelligence

Google's Harness²: Agents Self-Improve Their Own Harnesses Without Weight Updates

Google's Harness² introduces a dual-layer recursive self-improvement framework where agents optimize their own harnesses—prompt, tools, memory, skills—via behavioral comparison instead of ground truth, achieving up to 11.5-point gains on benchmarks and enabling weaker models to surpass stronger ones with static harnesses.

AI agentsAgent HarnessGoogle Research
0 likes · 19 min read
Google's Harness²: Agents Self-Improve Their Own Harnesses Without Weight Updates
Data Party THU
Data Party THU
Oct 10, 2026 · Artificial Intelligence

Google TimesFM-3: Native Multivariate Forecasting with Alternating Attention

Google's TimesFM-3 introduces native multivariate time series forecasting with alternating causal temporal and full variate attention, non-autoregressive decoding via contiguous patch masking, and support for historical and future-known covariates, achieving top rankings on Gift-Eval, FEV-Bench, and TIME benchmarks without sacrificing univariate performance.

Google ResearchTimesFM-3alternating attention
0 likes · 12 min read
Google TimesFM-3: Native Multivariate Forecasting with Alternating Attention
Data Party THU
Data Party THU
Sep 18, 2026 · Artificial Intelligence

Google AI Cyclone Model Outperforms High-Res Systems at 28km Resolution

Google's WeatherNext Cyclones (WN-C) AI model, published in Nature, achieves state-of-the-art tropical cyclone track and intensity forecasts using coarse 0.25° resolution data, outperforming specialized high-resolution models like HAFS and global systems like ENS and GenCast, with ensemble forecasts providing calibrated uncertainty and economic value, now operational at NHC.

AI weather modelingGoogle ResearchIBTrACS
0 likes · 7 min read
Google AI Cyclone Model Outperforms High-Res Systems at 28km Resolution
Architect
Architect
Sep 12, 2026 · Artificial Intelligence

Google's Multi-Agent Research: Task Structure, Not Agent Count, Determines Architecture Value

Google's research on 260 multi-agent configurations across six benchmarks shows centralized architectures improve parallel tasks by 81% but hurt sequential planning by 39-70%. Teamwork framework adds critique-synthesis loops that retain failed branches. The key insight: agent count isn't an architecture metric—task decomposability, verifiable sub-results, and coordination costs should drive design.

AI agentsGoogle ResearchTeamwork framework
0 likes · 18 min read
Google's Multi-Agent Research: Task Structure, Not Agent Count, Determines Architecture Value
Machine Heart
Machine Heart
Apr 17, 2026 · Artificial Intelligence

Combining Transformers and RNNs: Google’s Memory Caching Unlocks Ultra‑Long Context

Google Research introduces Memory Caching (MC), a technique that gives RNNs growing memory capacity, bridging the gap with Transformers to enable ultra‑long context processing while reducing memory demands, and demonstrates its effectiveness through extensive language‑modeling and recall experiments.

AI ArchitectureGoogle ResearchMemory Caching
0 likes · 7 min read
Combining Transformers and RNNs: Google’s Memory Caching Unlocks Ultra‑Long Context