Tag
该推文介绍一项关于自我改进智能体的研究，将 harness 自改方法映射到进化算法，并提出准入规则：只有在全新独立任务上以一定幅度超越当前版本时才接受编辑，以防止过拟合与技能遗忘。
这篇关于 Evo² 的文章提出将自进化 Agent 的两条路径（修改模型权重与修改 Harness）连接起来，让 Proposer 与 Reflector 双角色进入持续训练循环，在 Polyglot 上达到 82.1%、SWE-bench Verified 上达到 48.3%，且演化出的 Harness 可直接迁移复用。
This paper introduces EvoGen-Harness, a generator-agnostic framework that evolves multiple external responsibilities around frozen text-to-image models using a Trace method for failure attribution and coordinated search, achieving significant benchmark gains over existing single-dimension adaptation baselines.
Video-RSI 提出一种通过 harness 进化实现视频理解智能体递归自我改进的框架，无需更新模型权重，通过重新审视训练视频、收集证据并诊断失败来迭代改进提示、工具和记忆，从而在更少帧处理量下提升准确率。
The paper introduces CHASE, a counterfactual-guided method for harness evolution in AI agent evaluation that mitigates benchmark shortcuts, enhancing reliability through constraint generation and validity checks.
ModularRSI introduces a modular and generalizable framework for recursive self-improvement in AI agent harnesses, using contrastive learning across tasks to evolve modules independently and enhance performance on unseen tasks.
The paper introduces an on-policy expert-correction pipeline to co-evolve harnesses and models, enabling weaker AI models to improve by correcting failing turns without full imitation, preserving compatibility and enhancing performance on enterprise agent tasks.
JIT-Agent is an open-source 27B model that dynamically generates task-specific harnesses for AI agents, outperforming hand-built systems with improved token efficiency.
The paper introduces EVOHARNESSBENCH, a benchmark for evaluating LLM agents under evolving tool, skill, and agent harnesses, revealing gaps in retention and adaptation.
This technical report introduces Harness-Aware Training (HAT) to train compact AI agents that can evolve with changing configurations, improving performance and reducing latency for real-time interactions in live e-commerce digital avatars.
Introduces Evo-Bench, the first benchmark for evaluating language models' intrinsic harness-evolving capabilities across Search, Office, and General agent domains, showing top models achieve large gains but struggle on Office workflows.
The paper introduces Hierarchical Self-Improvement (HSI), a framework that enhances frozen LLM agents by evolving task-specific harnesses through hierarchical self-modification, achieving substantial gains on moderate tasks while being limited by feedback quality and backbone capabilities.
This paper re-evaluates the methodology of automatic harness evolution for LLM agents, highlighting that its gains may stem from additional test-time search rather than improved harness design, and that evaluation on the same benchmark risks overfitting. Experiments show that harness evolution does not consistently outperform simpler test-time scaling methods.
This paper rethinks how automatic harness evolution for agents should be evaluated, showing that gains may be due to increased compute rather than genuine improvements, and that evolved harnesses transfer poorly to unseen tasks.