SciMatrix 1.5 puts researchers inside multi-agent scientific discussions
The Chinese Academy of Sciences’ Computer Network Information Center announced SciMatrix 1.5 on November 11, adding human intervention, selectable agent discussion rooms and a pilot collaborator-recommendation feature. This is newly covered here, not a November 16 launch. Official announcement.
The platform’s developer says researchers can interrupt an agent, revise its assumptions and add constraints while a discussion continues. Users can select one or several of its 25 disciplinary agents. The team-recommendation pilot draws on an expert database to suggest potential research collaborators and explain its choices. These are described capabilities, not independently tested results.
Access is not presented as unrestricted public availability: the announcement says existing trial applicants will receive access gradually and new users must request a trial.
Analysis: The interesting design choice is placing expert correction inside the conversation rather than waiting for a final report. For research workflows, that could make assumptions easier to challenge. It does not establish that simulated disagreement improves scientific validity, or that suggested collaborators are suitable. Those questions require evaluation beyond the feature announcement.
The source supplies no independently verified accuracy, productivity or discovery benchmark. No account was created and no trial was run.
Sources and citations
Revision history
Published .