分层过程奖励模型是符号视觉学习器
Hierarchical Process Reward Models are Symbolic Vision Learners
用符号自编码器把图表解析成点/线/形状等结构化基元，并以分层过程奖励约束几何一致性，可替代 MLLM 的视觉编码器或直接为 LLM 提供可解释的符号解析输入，为几何/图表类视觉推理提供了一条非像素化的表征路径。
原文依据 one replaces the MLLM's visual encoder with the symbolic encoder; the other feeds interpretable symbolic parses directly to the LLM to guide reasoning
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论符号计算机视觉通过显式逻辑规则与结构化表示来表达图表，从而在机器视觉中实现可解释的理解。这要求与基于像素的视觉模型根本不同的学习范式。符号视觉学习器将图表解析为几何基元——点、线与形状——而基于像素的学习器则作用于纹理与颜色。我们提出一种新颖的自监督符号自编码器，其中编码器将图表解析为结构化基元及其关系，解码器将其编译回可执行代码以复现原始图表。该架构的核心是 SymHPR（符号分层过程奖励建模），它施加分层步骤级奖励，以强制点在线、线在形状、形状在关系上的一致性。由于强化学习在朴素 GRPO 下会崩溃，我们提出的稳定化策略平衡了探索与利用。在符号学习器之上，我们设计了两条流水线：一条用符号编码器替换 MLLM 的视觉编码器；另一条将可解释的符号解析直接送入 LLM 以引导推理。在重建、感知与推理任务上的评估证明了本方法的有效性：几何图表重建的 MSE 降低 98.2%，在图表重建上以 7B 模型超越 GPT-4o 0.6%，在 MathGlance 感知基准上提升 +13%，在 MathVerse 与 GeoQA 推理基准上提升 +3%。
英文原文摘要 1566 字符
Symbolic computer vision represents diagrams through explicit logical rules and structured representations, enabling interpretable understanding in machine vision. This requires fundamentally different learning paradigms from pixel-based visual models. Symbolic visual learners parse diagrams into geometric primitives--points, lines, and shapes--whereas pixel-based learners operate on textures and colors. We propose a novel self-supervised symbolic auto-encoder, where the encoder parses diagrams into structured primitives and their relations, and the decoder compiles them back into executable code that reproduces the original diagrams. Central to this architecture is SymHPR (Symbolic Hierarchical Process Reward Modeling), which applies hierarchical step-level rewards to enforce point-on-line, line-on-shape, and shape-on-relation consistency. Since reinforcement learning collapses under vanilla GRPO, our proposed stabilized strategies balance exploration and exploitation. Built upon the symbolic learner, we design two pipelines: one replaces the MLLM's visual encoder with the symbolic encoder; the other feeds interpretable symbolic parses directly to the LLM to guide reasoning. Evaluations across reconstruction, perception, and reasoning tasks demonstrate the effectiveness of our approach: achieving a 98.2% reduction in MSE for geometric diagram reconstruction, surpassing GPT-4o by 0.6% with a 7B model on chart reconstruction, and improving by +13% on the MathGlance perception benchmark, and by +3% on MathVerse and GeoQA reasoning benchmarks.