Agent Skill Evaluation and Evolution: Frameworks and Benchmarks
cs.CL
Submitted: 2026-06-09
Updated: 2026-11-02
Code: https://github.com/Cassie07/AgentSkill_Survey
Terminology
Sources
- Program Synthesis with Large Language Models
- $\tau^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment
- Building Self-Evolving Agents via Experience-Driven Lifelong Learning: A Framework and Benchmark
- Evaluating Large Language Models Trained on Code
- SkillCraft: Can LLM Agents Learn to Use Tools Skillfully?
- WildClawBench: A Benchmark for Real-World, Long-Horizon Agent Evaluation
- A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence
- SkillReducer: Optimizing LLM Agent Skills for Token Efficiency
- Beyond Seeing: Evaluating Multimodal LLMs on Tool-Enabled Image Perception, Transformation, and Reasoning
- SWE-Skills-Bench: Do Agent Skills Actually Help in Real-World Software Engineering?
- Measuring Massive Multitask Language Understanding
- XSkill: Continual Learning from Experience and Skills in Multimodal Agents
- Two Heads are Better Than One: Test-time Scaling of Multi-agent Collaborative Reasoning
- EmbodiSkill: Skill-Aware Reflection for Self-Evolving Embodied Agents
- Organizing, Orchestrating, and Benchmarking Agent Skills at Ecosystem Scale
- TIR-Bench: A Comprehensive Benchmark for Agentic Thinking-with-Images Reasoning
- MM-BrowseComp: A Comprehensive Benchmark for Multimodal Browsing Agents
- SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks
- ARISE: Agent Reasoning with Intrinsic Skill Evolution in Hierarchical Reinforcement Learning
- From Skill Text to Skill Structure: The Scheduling-Structural-Logical Representation for Agent Skills
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data