ScienceClaw: Benchmarking Continual Self-Evolution of AI-for-Science Agents Across the Natural and Social Sciences
cs.AI
Submitted: 2026-10-06
Updated: 2026-10-06
Comments: 28 pages
Code: https://github.com/beita6969/ScienceClaw
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- AI Coding Agents Can Reproduce Social Science Findings
- ChemCrow: Augmenting large-language models with chemistry tools
- From Atomic Actions to Standard Operating Procedures: Iterative Tool Optimization for Self-Evolving LLM Agents
- SEA-Eval: A Benchmark for Evaluating Self-Evolving Agents Beyond Episodic Assessment
- Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses
- PACE: Two-Timescale Self-Evolution for Small Language Model Agents
- Benchmarking AI Agents for Addressing Scientific Challenges Across Scales
- The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery
- Beyond Static Tools: Test-Time Tool Evolution for Scientific Reasoning
- AIRS-Bench: a Suite of Tasks for Frontier AI Research Science Agents
- EvoScientist: Towards Multi-Agent Evolving AI Scientists for End-to-End Scientific Discovery
- Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills
- SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agents
- S1-NexusAgent: a Self-Evolving Agent Framework for Multidisciplinary Scientific Research
- Voyager: An Open-Ended Embodied Agent with Large Language Models
- Executable Code Actions Elicit Better LLM Agents
- AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation
- ResearchClawBench: A Benchmark for End-to-End Autonomous Scientific Research
- SkillOpt: Executive Strategy for Self-Evolving Agent Skills
- ReAct: Synergizing Reasoning and Acting in 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