• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
  • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI
    Report

    Meta's Muse Spark reportedly helped answer five open research questions

    A user says Muse Spark generated code that helped find a 384-element counterexample to a 2024 group-theory conjecture.

    1 Source, 7h ago, first seen 7h ago

    TLDR

    A user says Meta researchers worked with mathematicians on six papers using Muse Spark 1.1 and 1.2, with five answering previously open research questions. The account describes work across several fields, including probability, differential equations and group theory. It says the researchers used Meta AI's standard chat interface without a custom research scaffold.

    Combined views

    —

    1 Source, first seen 7h ago

    Combined views

    —

    1 Source, first seen 7h ago

    — likes
    — likes
    — comments
    — saves
    — reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    Featured Source
    — comments
    — saves
    — reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    1 Source

    @Dr_Singularityhuge AI news from Meta OpenAI, Anthropic, and now Meta AI started making amazing math discoveries and accelerate progress. just day after 1 man using Claude explored roughly 400 scientific problems across 18 fields in just three months, helping produce 36 manuscripts and uncover new results in physics, ecology, genetics, mathematics, and more. 👀 ___________________________ Meta researchers worked with mathematicians using Muse Spark 1.1 and 1.2 on 6 research papers across multiple fields. Five of the papers answer previously open research questions, using the normal Meta AI chat interface, without a custom research scaffold. The results: 1. Probability - Gaussian ellipsoid fitting The team identified a sharp threshold for when random high-dimensional Gaussian points can all lie on a centered ellipsoid: below the threshold, such an ellipsoid exists with high probability; above it, it almost certainly does not. The exact behavior at the threshold remains open. 2. Differential equations - wave collapse They proved finite time blow up for radial negative energy solutions of the mass critical biharmonic nonlinear Schrödinger equation in dimensions two and higher, settling a question left open in 2015 and confirming a prediction from earlier simulations for this setting. 3. Group theory - a conjecture disproved Muse Spark generated GAP code that helped find a 384 element counterexample showing that semiabelian finite groups do not necessarily have to be monomial, disproving a 2024 conjecture. 4. Optimization - when a relaxation is exact. The researchers characterized when a cycle based relaxation of a certain binary polynomial optimization problem exactly reproduces the original problem, and when it necessarily leaves a gap. Muse Spark helped reframe the problem probabilistically, find a counterexample, and develop the proof strategy. 5. Arithmetic physics - connecting string theory and number theory. The team established a broader connection between p-adic string two point functions and height functions on curves, extending a relationship previously known in a more restricted setting. Muse Spark helped identify the connection, generate candidate proofs, and draft several technical sections. 6. Non associative algebra - another conjecture fallen. Muse Spark generated a 3 dimensional counterexample to a proposed characterization of solvable evolution algebras. The work went further and produced an alternative criterion based on whole subspaces rather than individual elements. ________________________________ acceleration accelerates9h