Research Papers 论文研究 4h ago Updated 29m ago 更新于 29分钟前 46

ADE: Agentic Data Evolution Framework for Human-Centered Objectives ADE:面向人类中心目标的智能体数据演化框架

ADE introduces a data-centric framework that treats synthetic supervision as evolving data snapshots rather than static datasets The core OVS (Observation-Variation-Selection) closed-loop procedure uses a steady-state admission mechanism as a quality ratchet to ensure only consistently improving updates are accepted On DEV300, ADE raises intrinsic win rate from 50% to 75.81% and extrinsic win rate from 55.20% to 68.86% Blind expert evaluation shows a 66.11% preference for ADE-evolved answers ove 提出ADE(Agentic Data Evolution)框架,通过闭环OVS(Observation-Variation-Selection)过程实现合成数据的持续迭代优化 引入稳态准入机制作为"质量棘轮",保守控制数据更新以保障跨轮次的持续改进 在DEV300基准上,ADE将内在胜率从50%提升至75.81%,外在胜率从55.20%提升至68.86% 盲评专家对演化答案的偏好率达66.11%,验证了数据质量提升的实际效果 该方法具有跨模型规模、跨训练方法和跨任务类型的泛化能力

58
Hot 热度
72
Quality 质量
68
Impact 影响力

Analysis 深度分析

TL;DR

  • ADE introduces a data-centric framework that treats synthetic supervision as evolving data snapshots rather than static datasets
  • The core OVS (Observation-Variation-Selection) closed-loop procedure uses a steady-state admission mechanism as a quality ratchet to ensure only consistently improving updates are accepted
  • On DEV300, ADE raises intrinsic win rate from 50% to 75.81% and extrinsic win rate from 55.20% to 68.86%
  • Blind expert evaluation shows a 66.11% preference for ADE-evolved answers over originals
  • Gains generalize across post-training methods, model scales, and tasks beyond the target weakly verifiable educational objectives

Why It Matters

This work addresses a critical bottleneck in LLM alignment: when human-centered objectives are non-executable and context-dependent, verification becomes unreliable and synthetic data selection replaces generation as the limiting factor. ADE's conservative quality ratchet directly tackles the problem of noisy signals causing silent regressions during iterative refinement, offering a principled approach to data evolution that practitioners can adopt for alignment pipelines.

Technical Details

  • OVS Loop: The Observation-Variation-Selection procedure forms a closed feedback cycle where data snapshots are continuously observed, varied through agentic generation, and selected via a steady-state admission gate that only accepts improvements demonstrating sustained cross-round gains
  • Steady-State Admission Mechanism: Acts as a quality ratchet, conservatively gating updates to prevent regressions; this prevents the common failure mode where noisy verification signals destabilize iterative refinement
  • Evaluation Framework: Dual validation through intrinsic trend tracking (measuring data quality directly) and extrinsic post-training evaluation (measuring downstream model performance)
  • Benchmarks: DEV300 dataset for educational objectives, with additional validation across diverse benchmarks, multiple model scales, and various post-training methods
  • Expert Validation: Blind evaluation by domain experts confirmed 66.11% preference for evolved answers, providing human-grounded validation beyond automated metrics

Industry Insight

  • The shift from generation-focused to selection-focused synthetic data pipelines represents a fundamental reorientation in alignment research; practitioners should invest in robust verification and admission mechanisms rather than simply scaling data generation
  • The steady-state admission ratchet design is broadly applicable beyond educational tasks—any domain with weakly verifiable, context-dependent objectives (e.g., creative writing, nuanced reasoning, value alignment) can benefit from this conservative evolution strategy
  • The cross-method and cross-scale generalization of ADE's gains suggests data evolution may become a standard preprocessing step in alignment workflows, potentially reducing dependence on expensive human annotation at scale

TL;DR

  • 提出ADE(Agentic Data Evolution)框架,通过闭环OVS(Observation-Variation-Selection)过程实现合成数据的持续迭代优化
  • 引入稳态准入机制作为"质量棘轮",保守控制数据更新以保障跨轮次的持续改进
  • 在DEV300基准上,ADE将内在胜率从50%提升至75.81%,外在胜率从55.20%提升至68.86%
  • 盲评专家对演化答案的偏好率达66.11%,验证了数据质量提升的实际效果
  • 该方法具有跨模型规模、跨训练方法和跨任务类型的泛化能力

为什么值得看

本文针对大模型对齐中"弱可验证目标"这一核心难题,提出以数据为中心的系统性解决方案,突破了传统合成数据生成中"重生成轻筛选"的瓶颈。其稳态准入机制为数据质量管控提供了可复用的工程范式,对教育、医疗等依赖上下文判断的高价值领域具有直接参考价值。

技术解析

  • 核心架构:ADE将合成监督数据组织为"演进数据快照",通过Observation(观测)-Variation(变异)-Selection(选择)的闭环流程实现数据迭代优化,而非一次性生成。
  • 质量管控机制:稳态准入机制(steady-state admission mechanism)作为"质量棘轮",仅当新数据在跨轮次对比中持续优于现有快照时才允许更新,有效防止噪声信号导致的静默退化。
  • 验证体系:采用内在趋势追踪(intrinsic trend tracking)与外在后训练评估(extrinsic post-training evaluation)双重验证,确保数据质量提升可转化为模型性能增益。
  • 实验基准:在DEV300数据集上验证,同时覆盖多样化基准测试,证明方法的有效性不仅限于单一任务。

行业启示

  • 数据质量优于数据规模:在弱可验证目标场景下,数据筛选机制的质量比单纯扩大合成数据量更重要,"精选"比"多生"更能突破性能瓶颈。
  • 闭环迭代是数据演进的必然路径:OVS框架为AI数据工程提供了可落地的迭代范式,建议团队在数据密集型任务中引入类似的闭环验证机制。
  • 跨领域泛化潜力:该方法在超出目标教育领域的任务中仍有效,预示其在医疗、法律、客服等需要上下文理解的高价值垂直领域具有广泛应用前景。

Disclaimer: The above content is generated by AI and is for reference only. 免责声明:以上内容由 AI 生成,仅供参考。

LLM 大模型 Alignment 对齐 Agent Agent Dataset 数据集 Research 科学研究