PhySciBench benchmark shows current AI models achieve at most 33.5% accuracy on physical science tasks; DelveAgent framework improves accuracy by up to 7.5 points and cuts costs to one-third.
Title resolution pending
19 Pith papers cite this work, alongside 806 external citations. Polarity classification is still indexing.
19
Pith papers citing it
806
external citations · OpenAlex
citation-role summary
background 2
method 1
citation-polarity summary
fields
cond-mat.mtrl-sci 4 cs.AI 4 physics.comp-ph 3 cs.CY 1 cs.DC 1 cs.LG 1 eess.SY 1 math-ph 1 physics.app-ph 1 physics.ed-ph 1years
2026 19representative citing papers
SCAPE enables 90-99% sparse gradient communication in sharded Adam-style LLM training by deriving masks from first-moment statistics, achieving up to 43.3% faster pre-training on Llama-500M with no loss in validation loss or downstream accuracy.
SAGE with MHFA improves failure recovery in autonomous research agents, raising metrics-bearing outputs from 42% to 92% on a 12-topic benchmark versus single-reflection baselines.
LAP is a new protocol extending A2A and MCP with four physical-world primitives for agent-to-instrument interaction in autonomous laboratories.
LLM agents run closed-loop design of photonic components and a full modulator by proposing, simulating, and refining against acceptance criteria.
ColPackAgent integrates a custom colpack Python package wrapping HOOMD-blue with MCP tools and an agent skill to enable reliable autonomous workflows for colloidal packing simulations across interactive, prompt-driven, and autoresearch modes.
ShardTensor is a domain-parallelism system for SciML that enables flexible scaling of extreme-resolution spatial datasets by removing the constraint of batch size one per device.
A two-layer certification framework decouples knowledge validity from human authorship to accommodate AI-enabled research in existing publication systems.
QMP-Bench supplies a realistic test set for AI on quantum many-body problems while PhysVEC uses integrated verifiers to turn unreliable LLM generations into code that passes both syntax and physics checks, outperforming baselines.
Autonomous controllers cannot remove decision risk from target components absent from their pre-action physical records; closing that gap requires expanded access, cost, or restricted deployment.
The paper introduces Experiment-as-Code Labs as a declarative stack synthesizing AI agents, systems orchestration, and physical lab control for AI-driven discovery.
An affordable Arduino-based IoT setup generates real-time optical data for students to compare traversal, Bayesian, and deep learning methods in a self-driving experimental workflow.
A trust-region Bayesian optimization framework integrates LEED multiple scattering models to jointly optimize structural and experimental parameters for automated surface reconstruction.
A new AutoML framework packages spectral featurization, interpretable modeling, and feature selection into a Python package and web app, demonstrated on XANES/PDF bond-length regression and grape sugar prediction.
AIMBio-Mat is a conceptual blueprint for an AI-native, FAIR, governance-aware decision layer that formulates biomedical-materials discovery as constrained multi-objective optimization under uncertainty.
Proposes a regional data-centric materials science ecosystem for the Great Plains, identifying five barriers to data sharing and outlining a staged roadmap illustrated by a high-purity germanium pilot.
Infrastructure is the primary obstacle to embodied AI for science in the Global South, and addressing it turns automation into essential capacity rather than a luxury.
A survey of generative crystal modeling, multimodal learning, and closed-loop inverse design pipelines for crystalline solids, including failure modes and evaluation practices.
citing papers explorer
-
Deep Research in Physical Sciences: A Multi-Agent Framework and Comprehensive Benchmark
PhySciBench benchmark shows current AI models achieve at most 33.5% accuracy on physical science tasks; DelveAgent framework improves accuracy by up to 7.5 points and cuts costs to one-third.
-
SCAPE: Accurate and Efficient LLM Training with Extreme Sparse Communication
SCAPE enables 90-99% sparse gradient communication in sharded Adam-style LLM training by deriving masks from first-moment statistics, achieving up to 43.3% faster pre-training on Llama-500M with no loss in validation loss or downstream accuracy.
-
One Reflection Is Not Enough: Self-Correcting Autonomous Research via Multi-Hypothesis Failure Attribution
SAGE with MHFA improves failure recovery in autonomous research agents, raising metrics-bearing outputs from 42% to 92% on a 12-topic benchmark versus single-reflection baselines.
-
LAP: An Agent-to-Instrument Protocol for Autonomous Science
LAP is a new protocol extending A2A and MCP with four physical-world primitives for agent-to-instrument interaction in autonomous laboratories.
-
Autonomous agentic design for photonics
LLM agents run closed-loop design of photonic components and a full modulator by proposing, simulating, and refining against acceptance criteria.
-
ColPackAgent: Agent-Skill-Guided Hard-Particle Monte Carlo Workflows for Colloidal Packing
ColPackAgent integrates a custom colpack Python package wrapping HOOMD-blue with MCP tools and an agent skill to enable reliable autonomous workflows for colloidal packing simulations across interactive, prompt-driven, and autoresearch modes.
-
ShardTensor: Domain Parallelism for Scientific Machine Learning
ShardTensor is a domain-parallelism system for SciML that enables flexible scaling of extreme-resolution spatial datasets by removing the constraint of batch size one per device.
-
Rethinking Publication: A Certification Framework for AI-Enabled Research
A two-layer certification framework decouples knowledge validity from human authorship to accommodate AI-enabled research in existing publication systems.
-
Towards Verifiable and Self-Correcting AI Physicists for Quantum Many-Body Simulations
QMP-Bench supplies a realistic test set for AI on quantum many-body problems while PhysVEC uses integrated verifiers to turn unreliable LLM generations into code that passes both syntax and physics checks, outperforming baselines.
-
Measurement-Access Risk Frontiers for Autonomous Scientific Control
Autonomous controllers cannot remove decision risk from target components absent from their pre-action physical records; closing that gap requires expanded access, cost, or restricted deployment.
-
Experiment-as-Code Labs: A Declarative Stack for AI-Driven Scientific Discovery
The paper introduces Experiment-as-Code Labs as a declarative stack synthesizing AI agents, systems orchestration, and physical lab control for AI-driven discovery.
-
Building an Affordable Self-Driving Lab: Practical Machine Learning Experiments for Physics Education Using Internet-of-Things
An affordable Arduino-based IoT setup generates real-time optical data for students to compare traversal, Bayesian, and deep learning methods in a self-driving experimental workflow.
-
Physics-informed automated surface reconstructing via low-energy electron diffraction based on Bayesian optimization
A trust-region Bayesian optimization framework integrates LEED multiple scattering models to jointly optimize structural and experimental parameters for automated surface reconstruction.
-
Spectra-Scope : A toolkit for automated and interpretable characterization of material properties from spectral data
A new AutoML framework packages spectral featurization, interpretable modeling, and feature selection into a Python package and web app, demonstrated on XANES/PDF bond-length regression and grape sugar prediction.
-
AIMBio-Mat: An AI-Native FAIR Platform for Closed-Loop Materials Discovery and Biomedical Translation
AIMBio-Mat is a conceptual blueprint for an AI-native, FAIR, governance-aware decision layer that formulates biomedical-materials discovery as constrained multi-objective optimization under uncertainty.
-
Building a Regional Data-Centric Materials Science Ecosystem for Processing-Rich Materials Innovation in the Great Plains
Proposes a regional data-centric materials science ecosystem for the Great Plains, identifying five barriers to data sharing and outlining a staged roadmap illustrated by a high-purity germanium pilot.
-
Infrastructure First: Enabling Embodied AI for Science in the Global South
Infrastructure is the primary obstacle to embodied AI for science in the Global South, and addressing it turns automation into essential capacity rather than a luxury.
-
Towards Automated Discovery: A Review of Generative Models, Multimodal Learning and Closed-Loop Workflows in Inverse Materials Design
A survey of generative crystal modeling, multimodal learning, and closed-loop inverse design pipelines for crystalline solids, including failure modes and evaluation practices.
- Optimizing Expert-Designed Crystal Graph Networks for Band-Gap Prediction with an Autonomous LLM Research Loop