根据业务问题、候选表列及选表规则，智能决策最合适的数据表。
LLM 角色和任务描述 Freewheel 广告业务介绍 术语黑话映射表: 包括了常见的黑话和通过 RAG 方式召回的部分 数据目录与仓库概述 候选表和列：也就是上一步召回的表和列，包括每个表的描述和选表逻辑 通用选表规则：应该遵守的规则，例如什么时候应该选择事实表或者维度表，什么时候应该选择多张表 选表样例：我们收集了多样化的问题并标注了最优选择的表，存储到选表样例向量数据库中，通过 RAG 的方式找到和用户提问相似的例子。
cs.IR提交于 已译
矛盾下的覆盖率：面向否定感知检索的基准
Coverage Under Contradiction: A Benchmark for Negation-Aware Retrieval
Pasha Abrishamchian · Eugene Yang · Andrew Yates · Nazli Goharian · Ophir Frieder
中文摘要
现有覆盖率评估基准主要量化检索证据对信息需求的覆盖情况，但多基于预训练组件可能已知的陈述，未验证当语料库证据与检索器参数化知识（parametric knowledge）矛盾时的表现。本研究通过为基于信息单元（nugget）的基准添加受控矛盾来探索这一行为。具体方法是为每个信息单元创建否定对应项，并编辑支持文档以支持该否定同时保留上下文。将此过程应用于 CoverageBench，构建了涵盖六个数据集、约 63,000 个原始-编辑证据对的 CoverageBench-Negate。基准附带丰富的构建元数据、所有对的 LLM 判断及样本人工审计。基线实验显示，BM25 区分原始与编辑近亲证据的能力显著优于稠密检索器（Qwen3 和 Jina），后者表现接近随机水平。该基准及构建材料已在 Hugging Face Datasets 发布。
关键要点
- 01现有基准未验证检索器在语料库证据与预训练知识矛盾时的表现。
- 02通过为信息单元创建否定对应项并编辑支持文档，构建受控矛盾基准。
- 03基于 CoverageBench 创建 CoverageBench-Negate，包含约 63,000 个证据对。
- 04BM25 在区分矛盾证据上显著优于 Qwen3 和 Jina 等稠密检索器。
- 05基准数据、构建元数据及 LLM 判断已在 Hugging Face 公开。
解读
尚无解读。
原始英文摘要
arXiv:2610.37162v1 Announce Type: cross Abstract: Coverage-evaluation benchmarks quantify whether retrieved evidence covers the different pieces of information needed to satisfy an information need. Existing benchmarks mostly consider factual claims that may already be known to pretrained retrieval components, leaving open whether retrievers find corpus evidence when it contradicts those priors. We study how retrievers behave when their parametric knowledge and the information in a corpus disagree. To do so, we develop a process to augment a nugget-based coverage benchmark with controlled contradictions. For each nugget, we create a negated counterpart and edit its supporting documents to support that negation while preserving the context. Applying this process to CoverageBench, a benchmark measuring information coverage, we create CoverageBench-Negate, spanning six datasets and roughly 63,000 original-edited evidence pairs (one per nugget-document link). We attach rich construction metadata to every pair and additionally provide out-of-family LLM judgments over all pairs and a human audit of a sample. In baseline experiments, BM25 separates original from edited near-twin evidence substantially better than dense retrievers (Qwen3 and Jina), which stay near chance. We release the benchmark on Hugging Face Datasets (nuggets, documents, pairs) together with the construction materials, including the prompts used to create it.