# naive bayes

Published articles for naive bayes.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Counting Clues: A Lightweight Probabilistic Baseline Can Match an LLM

DevFeed: [Counting Clues: A Lightweight Probabilistic Baseline Can Match an LLM](<https://devfeed.tech/articles/counting-clues-a-lightweight-probabilistic-baseline-can-match-an-llm-102925.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2512.12868>)

Author: Furong Jia, Yuan Pu, Finn Guo, Monica Agrawal

Published: 2026-10-28T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [LLM evaluation / benchmarking](<https://devfeed.tech/topics/llm-evaluation-benchmarking.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [diagnosis](<https://devfeed.tech/tags/diagnosis.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llm-inference](<https://devfeed.tech/tags/llm-inference.md>), [naive-bayes](<https://devfeed.tech/tags/naive-bayes.md>), [overlap](<https://devfeed.tech/tags/overlap.md>), [performance](<https://devfeed.tech/tags/performance.md>), [pretraining](<https://devfeed.tech/tags/pretraining.md>), [probabilistic](<https://devfeed.tech/tags/probabilistic.md>)

### AI overview

On MedQA questions asking for the most likely diagnosis, a lightweight Frequency-Based Probabilistic Ranker (FBPR) matched the corresponding LLMs when it used concept-diagnosis co-occurrence statistics from Dolma or RedPajama. FBPR scores answer options with a smoothed Naive Bayes model. Its answers overlapped with LLM inference barely above chance, indicating complementary strengths; a simple hybrid combining them gained notably over either method alone. The abstract presents FBPR as evidence that explicit probabilistic baselines can explain a substantial portion of LLM benchmark performance, while providing a complementary signal. It does not specify numerical scores or the hybrid procedure.

### Source excerpt

arXiv:2512.12868v2 Announce Type: replace-cross Abstract: Large language models (LLMs) excel on multiple-choice clinical diagnosis benchmarks, yet it is unclear how much of this performance reflects underlying probabilistic reasoning. We study this through questions from MedQA that ask for the most likely diagnosis. We introduce Frequency-Based Probabilistic Ranker (FBPR), a lightweight method that scores options with a smoothed Naive Bayes over concept-diagnosis co-occurrence statistics from a large corpus. When co-occurrence statistics are sourced from Dolma, the pretraining corpus for OLMo, and RedPajama, an open reproduction of the LLaMA pretraining-data recipe, FBPR achieves comparable performance to the corresponding LLMs. Furthermore, direct LLM inference and FBPR exhibit complementary strengths, correctly answering largely different subsets of questions with overlap barely above chance. Combining them with a simple hybrid method yields a notable gain over either method alone. These findings show that explicit probabilistic baselines can account for a substantial portion of the benchmark performance on diagnosis questions, while serving as historically grounded reference points and complementary signals for hybridization.

## Word Segmentation with Google's N-Gram Corpus

DevFeed: [Word Segmentation with Google's N-Gram Corpus](<https://devfeed.tech/articles/word-segmentation-or-makingsenseofthis-40255.md>)

Original publisher: [Read original article](<https://www.jeremykun.com/2012/01/15/word-segmentation/>)

Published: 2012-01-15T11:10:51Z

Content type: tutorial

Language: en

Sources: [Jeremy Kun](<https://devfeed.tech/sources/jeremy-kun.md>)

Topics: [Programming](<https://devfeed.tech/topics/programming.md>), [Algorithms](<https://devfeed.tech/topics/algorithms.md>), [Google](<https://devfeed.tech/topics/google.md>), [Code](<https://devfeed.tech/topics/code.md>), [Python](<https://devfeed.tech/topics/python.md>)

Tags: [algorithm](<https://devfeed.tech/tags/algorithm.md>), [article](<https://devfeed.tech/tags/article.md>), [code](<https://devfeed.tech/tags/code.md>), [dynamic-programming](<https://devfeed.tech/tags/dynamic-programming.md>), [google](<https://devfeed.tech/tags/google.md>), [linguistics](<https://devfeed.tech/tags/linguistics.md>), [mathematics](<https://devfeed.tech/tags/mathematics.md>), [naive-bayes](<https://devfeed.tech/tags/naive-bayes.md>), [ngrams](<https://devfeed.tech/tags/ngrams.md>), [programming](<https://devfeed.tech/tags/programming.md>), [python](<https://devfeed.tech/tags/python.md>)

### AI overview

This tutorial explains word segmentation, counts the exponentially many possible segmentations, and introduces a dynamic-programming approach using a simple model based on a subset of Google's n-gram corpus.

### Source excerpt

A First Look at Google's N-Gram Corpus In this post we will focus on the problem of finding the appropriate word boundaries in strings like "homebuiltairplanes", as is common in web URLs like www.homebuiltairplanes.com. This is an interesting problem because humans do it so easily, but there is no obvious programmatic solution. We will begin this article by addressing the complexity of this problem, continue by implementing a simple model using a subset of Google's n-gram corpus, and finish by describing our future plans to enhance the model.