NeutronGym: Physics-Graded Neutron Instrument Design for LLM Agents
cs.AI, physics.ins-det
Submitted: 2026-10-02
Updated: 2026-10-02
Code: https://github.com/ljding94/NeutronGym
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- ReST meets ReAct: Self-Improvement for Multi-Step Reasoning LLM Agent
- Reinforcement Learning with Verifiable Physics: Post-training LLMs with Continuous Rewards
- Frontier-Eng: Benchmarking Self-Evolving Agents on Real-World Engineering Tasks with Generative Optimization
- Agentic-AI Detector Co-design and Optimization in Vertically-Integrated Differentiable Full Simulations
- Collider-Bench: Benchmarking AI Agents with Particle Physics Analysis Reproduction
- BoxingGym: Benchmarking Progress in Automated Experimental Design and Model Discovery
- OPTIAGENT: A Physics-Driven Agentic Framework for Automated Optical Design
- GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators
- LLMs Gaming Verifiers: RLVR can Lead to Reward Hacking
- GRACE: an Agentic AI for Particle Physics Experiment Design and Simulation
- Auto-Configuring Scientific Simulators with Lightweight Coding-Agent Adapters
- Autonomous agentic design for photonics
- MDGYM: Benchmarking AI Agents on Molecular Simulations
- Rongzai agent: A Large Language Model-Based Autonomous Assistant for Rietveld Refinement of Neutron Diffraction Data
- OPT-BENCH: Evaluating the Iterative Self-Optimization of LLM Agents in Large-Scale Search Spaces
- Stargazer: A Scalable Model-Fitting Benchmark Environment for AI Agents under Astrophysical Constraints
- EnvSimBench: A Benchmark for Evaluating and Improving LLM-Based Environment Simulation
- GEM: A Gym for Agentic LLMs
- D3-Gym: Constructing Real-World Verifiable Environments for Data-Driven Discovery
- YaRN: Efficient Context Window Extension of Large Language Models
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems