Gate News message, April 11, AI infrastructure company Ramp Labs released research findings called “Latent Briefing,” enabling efficient memory sharing among multi-agent systems by directly compressing large-model KV caches, greatly reducing Token consumption without losing accuracy. In mainstream multi-agent architectures, the orchestrator breaks down tasks and repeatedly calls worker model instances; as the inference chain grows longer, Token usage expands exponentially. The core idea behind Latent Briefing is to use the attention mechanism to identify the truly crucial parts of the context, discard redundant information directly at the representation layer, rather than relying on slow LLM summarization or RAG retrieval with less stable results. On the LongBench v2 benchmark, the method performed impressively: the worker model’s Token consumption dropped by 65%, the Token savings’ median for medium-length documents (32k to 100k) reached 49%, overall accuracy improved by about 3 percentage points versus the baseline, and the additional time spent per compression was only about 1.7 seconds—roughly a 20x speedup compared with the original algorithm. The experiments used Claude Sonnet 4 as the orchestrator and Qwen3-14B as the worker model, covering a wide range of document scenarios including academic papers, legal documents, novels, and government reports. The study also found that the optimal compression threshold varies with task difficulty and document length—hard problems are better suited to aggressive compression to filter speculative reasoning noise, while long documents are better suited to lighter compression to preserve dispersed key information.
Disclaimer: The information on this page may come from third parties and does not represent the views or opinions of Gate. The content displayed on this page is for reference only and does not constitute any financial, investment, or legal advice. Gate does not guarantee the accuracy or completeness of the information and shall not be liable for any losses arising from the use of this information. Virtual asset investments carry high risks and are subject to significant price volatility. You may lose all of your invested principal. Please fully understand the relevant risks and make prudent decisions based on your own financial situation and risk tolerance. For details, please refer to
Disclaimer.
Related Articles
Agile Soda Launches Agentic OCR Platform with 98% Document Classification Accuracy
Agile Soda launched Agentic OCR, an AI-driven document automation platform that eliminates pre-training and allows for instant deployment. It offers high accuracy in classification and extraction, improving continuously through user corrections, with plans for future enhancements.
GateNews1h ago
American Express to Acquire AI Expense Startup Hyper in Q2 2026
American Express will acquire AI startup Hyper to enhance its expense management tools for commercial clients. The acquisition, expected to close in Q2 2026, follows a partnership that launched a co-branded rewards card in 2024.
GateNews3h ago
Singapore Cloud Startup OrtCloud Raises $1.7M in Pre-Seed Round Led by Golden Gate Ventures
OrtCloud, a Singaporean startup, raised $1.7 million in pre-seed funding for its specialized cloud infrastructure designed for AI workloads. With clients like OpenAI and Samsung, it aims to enhance product development and expand in Asia Pacific and the U.S.
GateNews3h ago
Canva Launches AI 2.0 Platform, Expanding from Design Tool to Unified Work OS
Canva AI 2.0 transforms Canva from a design tool to a comprehensive work operations platform. It uses generative AI to streamline workflows, enabling users to create and edit designs via natural language, automate tasks, and integrate with various applications.
GateNews4h ago
Sahara AI Launches Investment Agent Sorin Supporting Crypto, Stocks, and Prediction Markets
Sahara AI has launched Sorin, an investment agent for trading across various assets like cryptocurrencies and stocks. It offers autonomous trading, quantitative strategy automation, and personalized risk management to all users, following testing with 20,000 participants.
GateNews17h ago
AlphaNet Raises $10M Seed Round Led by Joffre Capital to Launch Institutional-Grade Quantitative Trading Platform
Quantitative AI trading platform AlphaNet secures $10 million in seed funding, preparing for a public launch with over 30 high-performing strategies. Plans include an Open Platform for strategy integration by 2026, aiming for 100+ strategies.
GateNews20h ago
