5. VideoRAG & V-RAGBench
5. VideoRAG 与 V-RAGBench
本文提出 V-RAGBench——一个由 ⟨query, evidence chunk, answer⟩ 三元组构成的基准，可对检索与生成进行忠实且解耦的评估；同时提出 CARVE，一种在多种配置下并行运行检索器、并通过分块自适应重排序为每个分块挑选最优配置的简易方法。V-RAGBench is introduced, a benchmark of $\langle$ query, evidence chunk, answer$\rangle$ triplets that enables faithful, decoupled evaluation of retrieval and generation, and CARVE, a simple method that runs parallel retrievers across configurations and employs chunk-adaptive reranking to identify the winning configuration for each chunk.
1️⃣2️⃣ arXiv · RAGPerf: End-to-End RAG Benchmarking Framework（⭐⭐⭐ 参考）
arXiv · RAGPerf：端到端 RAG 基准测试框架（⭐⭐⭐ 参考）
提出一个面向 RAG 的 AI 系统基准测试（RAGPerf）框架，用于刻画 RAG pipeline 的系统行为，并证明其引入的性能开销可忽略不计。The design and implementation of a RAG-based AI system benchmarking (RAGPerf) framework for characterizing the system behaviors of RAG pipelines is presented and it is shown that RAGPerf incurs negligible performance overhead.
5️⃣ arXiv · Is Agentic RAG Worth It? An Experimental Comparison of RAG Approaches（⭐⭐⭐⭐ 高优先级）
5️⃣ arXiv · Agentic RAG 是否值得？RAG 方法的实验对比（⭐⭐⭐⭐ 高优先级）
基于实证对 "Enhanced" 与 "Agentic" RAG 范式进行评估，为真实场景中选取最有效的 RAG 设计（兼顾性能与成本）提供指导。An empirically driven evaluation of the "Enhanced" and "Agentic" RAG paradigms is conducted, offering guidance on selecting the most effective RAG design for real-world applications, considering both performance and costs.
Taste-aware music retrieval from audio embeddings
基于音频嵌入的品味感知音乐检索
将预测的味觉空间作为基于内容的检索索引，对 309 项条目池的排序比 CLAP-text 基线（处于随机水平）忠实得多；ridge probes 与 audio-bandstop knockout 在已记载的声-味对应关系上读出了最强表征。Operationalised as a content-based retrieval index, the predicted taste space ranks a 309-item pool far more faithfully than a CLAP-text baseline, which sits at chance; ridge probes and an audio-bandstop knockout read the strongest representations against documented sound-taste correspondences.
MIRROR: Novelty-Constrained Memory-Guided MCTS Red-Teaming for Agentic RAG
MIRROR：面向智能体 RAG 的新颖性约束记忆引导 MCTS 红队对抗
提出 MIRROR——一个统一的跨表层框架，在显式新颖性约束下以检索到的上下文为条件生成候选，并执行记忆引导的蒙特卡洛树搜索，使检索可影响搜索先验，同时避免提示词级别的复制。MIRROR is presented, a unified cross-surface framework that performs memory-guided Monte Carlo tree search while conditioning candidate generation on retrieved context under an explicit novelty constraint, allowing retrieval to inform search priors without enabling prompt copying.
HAKARI-Bench: A Lightweight Benchmark for Comparing Retrieval Architectures and Efficiency Settings under Unified Conditions
HAKARI-Bench: A Lightweight Benchmark for Comparing Retrieval Architectures and Efficiency Settings under Unified Conditions
HAKARI-Bench 是一个轻量级基准，将现有检索套件重建为统一格式的小型数据集（Nano-sets），支持在同一条件下对五类检索方法及其效率变体进行与模型无关的对比。HAKARI-Bench is a lightweight benchmark that reconstructs existing retrieval suites into small datasets (Nano-sets) in a unified format, enabling same-condition, model-agnostic comparison of five retrieval families and their efficiency variants.
When Does Streaming Tool Use Help? Characterizing Tool-Intent Stabilization in Streaming Retrieval-Augmented Generation
流式工具使用何时有效？刻画流式 RAG 中的工具意图稳定化
本文命名并测量了推测性查询的检索收敛到包含答案结果时的输入流位置——"工具意图稳定化"：即推测性查询的检索收敛到包含答案结果的输入流位置。This work names and measures the point in the input stream at which a speculative query's retrieval converges on the answer-bearing result, tool-intent stabilization: the point in the input stream at which a speculative query's retrieval converges on the answer-bearing result.
KAMR: Grounding Generation via Knowledge-Aligned Multi-hop Retrieval
KAMR: 基于知识对齐多跳检索的接地生成
提出知识对齐的多跳检索器 KAMR，区分受 query 强约束的 anchor triplet 和弱对齐但在结构上与 anchor 相连的 connected triplet，持续提升多跳检索及下游问答性能。A knowledge-aligned multi-hop retriever, KAMR, which distinguishes anchor triplets that are strongly constrained by the query from connected triplets that are weakly aligned yet structurally linked to the anchors, which consistently improves multi-hop retrieval and downstream question answering performance.