Skip to main content

Share story

Science

Preprint proposes causal method to prune LLM-agent communication links

A new arXiv preprint describes E2-Explainer, a framework for identifying communication links that matter in LLM-based multi-agent systems. The method measures how masking individual channels changes task outcomes and final-response stability, then distills selected subgraphs into an explainer for deployment. The authors report that the resulting subgraphs preserved successful collaboration on reasoning and coding benchmarks while allowing redundant communication edges to be pruned. The results are preliminary and have not been independently verified.
Sources
Published by Tech & Business, a media brand covering technology and business. This story was sourced from cs.AI updates on arXiv.org and reviewed by the T&B editorial agent team.
Back to Newswire
Keep reading
Full wire
Capital
Capital

Taisei backs $150 million fund for AI and robotics startups

Japanese construction company Taisei Corp. is bankrolling a $150 million fund to invest in AI and robotics startups, Bloomberg reported. The 153-year-old company is seeking technological ways to address a shrinking workforce through its investment in those businesses.

Capital
Capital

Edelbyte emerges from stealth with $70 million in funding

Edelbyte has emerged from stealth with $70 million in seed and Series A funding to compete with data infrastructure providers including Snowflake, Databricks and ClickHouse. Sequoia and Quiet Capital led the financing, with participation from Greenoaks and Swish Ventures.

Capital
Capital

Nidus raises $30M across pre-seed and seed rounds

Nidus raised $30M across pre-seed and seed rounds, Upstarts Media reported. Backed by Kleiner Perkins and Costanoa, the startup is developing AI-powered technology for manufacturing robots used by US defense industrial companies.