Beyond Final Scores: A Systematic Evaluation of Agents for Long-Horizon AI Research and Development
Score
76首次系统性地将过程指标引入长期研发型Agent评估，揭示了最终分数之外的丰富行为细节
研究设计严谨、分析层次丰富，基于7个前沿模型和36个任务的受控对比具有较高可信度
提出的评估框架与改进方向可直接用于Agent训练、经验管理与框架设计实践
当前正值自主智能体快速发展期，该论文聚焦长期研发能力评估，具有很强时效性
来源为arXiv AI/CL/LG子类，属于学术预印本，虽未经同行评审但来自可靠学术发布渠道
Base 84.5 × T2 Source weight ×0.9 = 76
Curation line 70Selected
Why selected
High information gain as it is the first systematic integration of process metrics into long-horizon R&D agent evaluation, revealing complex effects of experience reuse and harness design.
AI commentary
「This paper deserves close reading by anyone concerned with AI's autonomous research capabilities: it moves beyond coarse 'can agents score high' judgments and dissects the process. You'll find that many seemingly clever results come from experience reuse and harness support, while agents that genuinely design new methods from scratch are rare. For those working on agent training or evaluation frameworks, the analysis here can help you avoid the 'score trap'.」
Quick overview
This paper presents a new framework using rule-based metrics to characterize within-run behavior through Solution Framing, Execution, and Feedback Control, and performs controlled comparisons of seven frontier models on 36 long-horizon tasks to assess experience reuse. Results show current agents operate more like engineering optimizers than fully autonomous researchers: they can formulate and implement practical solutions, but performance varies substantially across runs, their strongest solutions mainly adapt or combine established techniques, and genuine methodological novelty remains rare. Detailed analysis reveals distinct process bottlenecks behind similar final outcomes, experience reuse that can help or mislead, and harness designs that affect stability, suggesting concrete directions for improving model training, inference-time strategies, experience management, and harness design.
- Proposes an evaluation framework with Solution Framing, Execution, and Feedback Control stages, using rule-based metrics to analyze agent behavior beyond final scores
- Controlled comparison of seven frontier models on 36 long-horizon R&D tasks finds current agents resemble engineering optimizers more than autonomous researchers
- Experience reuse can help or mislead subsequent decisions, and harness design significantly affects performance stability
- Identifies distinct process bottlenecks behind similar final outcomes, offering concrete directions for improving model training and inference strategies
Full text
This is an AI-generated Chinese digest; use "View original" for the source text.
Related items
85
FeaturedSetoka: A Benchmark for Hierarchical User Understanding in Personalized Agents over Heterogeneous Data
Setoka：面向异构数据个性化Agent的层级用户理解基准
This paper is well worth attention, especially for researchers and engineers working on personalized agents, dialogue systems, and memory-augmented AI. It not only provides a benchmark but defines levels of user understanding from psychology, revealing systematic failures of current systems at higher levels. If you are designing memory-based agents, this paper reminds you that memory systems need to support cross-temporal, cross-source abstract reasoning, not just retrieval.
85
FeaturedBAND: Bayesian Network Distribution Regression for High-Dimensional Distribution Estimation – Breaking the Curse
BAND：高维分布估计的贝叶斯网络回归方法——打破维度诅咒
This paper is a must-read for anyone working on high-dimensional statistics, probabilistic modeling, and generative AI. It provides a simple yet powerful theoretical answer: sparsity can break the curse of dimensionality, and Bayesian networks offer a natural framework. Compared to black-box methods like VAEs and GANs, BAND is interpretable and provably convergent, especially valuable for industry needs like reliable confidence intervals. If you research time series forecasting or risk management, this paper may change how you approach high-dimensional density estimation.
84
FeaturedCycle-Consistent and Uncertainty-Aware Neural Surrogates for Tokamak Edge Plasmas
循环一致与不确定性感知的托卡马克边缘等离子体神经代理模型
This paper is worth attention because it demonstrates how advanced AI techniques (cycle consistency, uncertainty quantification) directly address a core computational bottleneck in nuclear fusion. For fusion researchers and engineers, it offers a practical tool to speed up simulations by orders of magnitude without sacrificing accuracy; for AI practitioners, it provides an excellent case of self-supervised learning and inverse problem solving in scientific applications.
83
FeaturedEquivariant Learning of a Transferable Three-Dimensional Classical Density Functional
等变学习可迁移的三维经典密度泛函
The most striking aspect of this paper is its direct coupling of machine learning to a core problem in statistical mechanics, using equivariant networks to learn a transferable density functional without any labels. For researchers in liquid physics or soft matter, this could be a new paradigm: do we need to run separate simulations for every new system anymore? For AI researchers, it also shows how to embed physical symmetries and variational principles into models rather than just fitting data. I recommend focusing on the methods, especially how equivariance and variational consistency are implemented—that's the real essence.
83
FeaturedComputational and Statistical Guarantees of the c-Rectified Flow
c-Rectified Flow 的计算与统计保证
This paper moves rectified flow from 'works in practice' to 'theoretically controllable,' which is especially important for those working on the underlying algorithms of generative models. If you are tuning pipelines like FLUX or SD3, understanding that ordinary rectified flow can fail when covariances do not commute may help you avoid hidden training pitfalls.
83
FeaturedLearning Deployable Manipulation Policies from High-Fidelity UMI Data Alone
从高保真UMI数据学习可部署操作策略
The most striking point of this paper is proving that pure robot-free data can match real teleoperation performance, breaking the convention of needing a small real-robot anchor for post-training. For research groups looking to reduce hardware costs and industrial robot application developers, HiFi-UMI's data production system and open-source dataset are directly usable resources. Pay attention to the engineering details of head-mounted SLAM and microsecond synchronization—they are key to achieving the reported accuracy.
83
FeaturedCapital Markets LLM Reliability Score: From Plausible to Bankable
资本市场LLM可靠性评分：从“看似合理”到“可银行化”
This paper is valuable for both finance and AI communities: it provides an actionable, workflow-oriented evaluation framework that's closer to real needs than generic benchmarks. If you're deploying LLMs in fintech or responsible for explainability/auditability, this is a must-read. The finding on decision usefulness is particularly striking—judges agree it's critical, yet models vary the most on it.
83
FeaturedWhere Should Optimizer State Live? Tiered State Allocation for Memory-Efficient Mixture-of-Experts Training
优化器状态应放在哪里？面向内存高效MoE训练的分层状态分配
This paper is highly valuable for teams training MoE models: it demonstrates that with careful optimizer state allocation, one can dramatically reduce memory without sacrificing accuracy—and even improve performance. It is especially relevant for labs or enterprises constrained by 40 GB or 80 GB GPU memory. The tiered strategy and ablation on momentum retention are worth studying for optimizing training frameworks like DeepSpeed or Megatron.