paper-with-me

Papers

PhantomWiki: On-Demand Datasets for Reasoning and Retrieval Evaluation

2025-02-27 · Albert Gong, Kamilė Stankevičiūtė, Chao Wan, Anmol Kabra, Raphael Thesmar, Johann Lee, Julius Klenke, Carla P. Gomes, Kilian Q. Weinberger

High-quality benchmarks are essential for evaluating reasoning and retrieval capabilities of large language models (LLMs). However, curating datasets for this purpose is not a permanent solution as they are prone to data leakage and inflated performance results. To address these challenges, we propose PhantomWiki: a pipeline to generate unique, factually consistent document corpora with diverse question-answer pairs. Unlike prior work, PhantomWiki is neither a fixed dataset, nor is it based on any existing data. Instead, a new PhantomWiki instance is generated on demand for each evaluation. We vary the question difficulty and corpus size to disentangle reasoning and retrieval capabilities respectively, and find that PhantomWiki datasets are surprisingly challenging for frontier LLMs. Thus, we contribute a scalable and data leakage-resistant framework for disentangled evaluation of reasoning, retrieval, and tool-use abilities. Our code is available at https://github.com/kilian-group/phantom-wiki.

📄 PDF Abstract BibTeX arXiv:2502.20377

Code (1)

kilian-group/phantom-wiki 공식 구현

Tasks

Retrieval

Similar Papers 제목 키워드 기반

Memento: Note-Taking for Your Future Self

2025-06-25 · Chao Wan, Albert Gong, Mihir Mishra, Carl-Leander Henneking 외

Large language models (LLMs) excel at reasoning-only tasks, but struggle when reasoning must be tightly coupled with retrieval, as in multi-hop question answering. To overcome these limitations, we introduce a prompting …

Multi-hop Question AnsweringQuestion AnsweringRAG

MR$^2$-Bench: Going Beyond Matching to Reasoning in Multimodal Retrieval

2025-09-30 · Junjie Zhou, Ze Liu, Lei Xiong, Jin-Ge Yao 외 arxiv

Multimodal retrieval is becoming a crucial component of modern AI applications, yet its evaluation lags behind the demands of more realistic and challenging scenarios. Existing benchmarks primarily probe surface-level se…

Semantic correspondenceCausal Inference

CoAL-RAG: A Complexity-Aware Legal Retrieval-Augmented Generation Method

2026-08-18 · Jin Su, Zhuofeng Zhao, Huanhuan Wang, Hao Chen arxiv

Legal consultation questions exhibit multi-level complexity. A single retrieval strategy often leads to over-reasoning for simple questions and poor interpretability for complex ones, making it difficult to meet the requ…

Semantic RetrievalLogical Reasoning

NoTeS-Bank: Benchmarking Neural Transcription and Search for Scientific Notes Understanding

2025-04-12 · Aniket Pal, Sanket Biswas, Alloy Das, Ayush Lodh 외

Understanding and reasoning over academic handwritten notes remains a challenge in document AI, particularly for mathematical equations, diagrams, and scientific notations. Existing visual question answering (VQA) benchm…

BenchmarkingDocument AIdocument understandingMultimodal Reasoning+6

MC-Search: Evaluating and Enhancing Multimodal Agentic Search with Structured Long Reasoning Chains

2026-03-01 · Xuying Ning, Dongqi Fu, Tianxin Wei, Mengting Ai 외 arxiv

With the increasing demand for step-wise, cross-modal, and knowledge-grounded reasoning, multimodal large language models (MLLMs) are evolving beyond the traditional fixed retrieve-then-generate paradigm toward more soph…

Multimodal Reasoning