Can Language Models Play the Wikipedia Game?
Language models increasingly support or automate information retrieval from document collections too large to index.
ProofPaper ↗
Key points
- We used the Wikipedia Game to test language models on an analogous information retrieval task.
- We compared sentence embedding models and LLMs across five Wikipedia languages and five academic-literature datasets.
- We construct very large local graphs of Wikipedia pages and of paper citations using OpenAlex and provide code to build and navigate them.
- The Wikipedia Game format enables robust comparisons across techniques, languages, and domains.
Sources (1)
- [1]Can Language Models Play the Wikipedia Game?arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 05:59 PM
Language models increasingly support or automate information retrieval from document collections too large to index.
We used the Wikipedia Game to test language models on an analogous information retrieval task.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 9, 2026A Critical Look at Power Sampling when Test Time Sharpening LLMs
- Oct 9, 2026Taking Tokenization off the Host for Agentic LLM Serving
- Oct 9, 2026Forecasting Corpus-Sampling Disagreement in Masked Text Reconstruction
- Oct 9, 2026Collaborative Theory of Mind: Learning to Self-Revise by Anticipating User Responses
- Oct 9, 2026Surfing the Activation Wave: Analyzing the Information and Geometry of LLM Hidden States for Speculative Decoding
- Oct 8, 2026A Closer Look at Agentic BBO: Benchmarking LLM Agents for Black-Box Optimization