A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Zeyu Ren, Ling Yue, Ran Li, et al.
FlowEvo is a training-free framework where workflows and executable skills co-evolve at inference time, improving agent performance across benchmarks while reducing token usage.
Mingxuan Zheng, Yujin Zhou, Chuxue Cao, et al.
SkillProx introduces a proximal-gradient-inspired framework for self-evolving agent skills, combining closed-loop diagnostic evolution with utility-aware proximal refinement to improve accuracy by 3.0 points over gradient-based baselines.
Boxiu Li, Zimo Wen, Yijia Fan, et al.
Argus is a persistent, self-evolving agentic runtime that separates stable user intent from operational objectives, enabling long-horizon reasoning with fixed model weights and achieving ~78% on SWE-Bench Pro.
Siyuan Wang, Zhuohan Long, Zhihao Fan, et al.
This paper introduces a multi-agent framework for dynamically evolving benchmarks to keep pace with rapidly advancing LLMs.
Unknown
This paper investigates reinforcement learning for interactive tool-using agents, highlighting user model fine-tuning as a critical component for training such agents.
Unknown
SPHERE is a self-evolving data generation pipeline that enhances mathematical reasoning in small language models by iteratively generating, correcting, and diversifying reasoning chains.
Yuyao Wang, Zhongjian Zhang, Mo Chi, et al.
EvoMemBench benchmarks agent memory from a self-evolving perspective, providing a unified testbed for adaptive memory in execution-oriented tasks.