Record · corpus.blog/posts/01a08784-a37a-718e-8ddc-ed9fc5cfc1e7

Chunking in RAG: why 512 tokens is not a strategy

ccomkhj.github.io · published 14 July 2026

A record is what was published and who pointed at it. The text of the post is not held here: read it at the source, or at its Wayback capture.

Read the post at ccomkhj.github.io Wayback capture

0blogs citing

First cited: —
Most recent: —
Rank this month: not ranked
Links from this post: 21
Words captured: 2,475

Cited by

No blog we hold has cited this post yet.

Links from this post

Link

Host

https://www.anthropic.com/engineering/contextual-retrieval

anthropic.com

https://arxiv.org/abs/2409.04701

arxiv.org

https://aclanthology.org/2024.emnlp-main.845/

aclanthology.org

https://aclanthology.org/2025.findings-naacl.114/

aclanthology.org

https://aclanthology.org/2026.acl-long.1372/

aclanthology.org

https://docs.langchain.com/oss/python/integrations/splitters/recursive_text_splitter

docs.langchain.com

https://docling-project.github.io/docling/concepts/chunking/

docling-project.github.io

https://github.com/superlinear-ai/raglite

github.com

https://github.com/jina-ai/late-chunking

github.com

https://aclanthology.org/2022.naacl-main.272/

aclanthology.org

https://developers.llamaindex.ai/python/framework/integrations/retrievers/auto_merging_retriever/

developers.llamaindex.ai

https://docs.ragas.io/en/stable/concepts/metrics/available_metrics/

docs.ragas.io

https://docs.langchain.com/langsmith/evaluate-rag-tutorial

docs.langchain.com

https://arxiv.org/abs/2005.11401

arxiv.org

https://weaviate.io/blog/late-chunking

weaviate.io

https://arxiv.org/abs/2401.18059

arxiv.org

https://arxiv.org/abs/2406.15319

arxiv.org

https://aclanthology.org/2024.tacl-1.9/

aclanthology.org

https://arxiv.org/abs/2406.00456

arxiv.org

https://aclanthology.org/2025.findings-acl.422/

aclanthology.org

https://arxiv.org/abs/2603.25333

arxiv.org