Skip to content

Commit 6fc4f25

Browse files
avokclaude
andcommitted
update default chunking methods in README and SVG diagrams
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
1 parent d51b71b commit 6fc4f25

3 files changed

Lines changed: 7 additions & 7 deletions

File tree

README.md

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -66,9 +66,9 @@ All metrics are implemented in [`metrics.py`](src/adaptive_chunking/metrics.py)
6666

6767
| Method | Description |
6868
|--------|-------------|
69-
| **Recursive** | Splits on a hierarchy of separators (paragraphs → sentences → characters) with configurable merge modes and overlap |
70-
| **Sentence** | Groups consecutive sentences (NLTK, SpaCy, or Stanza) into fixed-size chunks |
71-
| **Semantic** | Detects topic shifts via embedding similarity between consecutive sentences |
69+
| **Recursive (merge-to-size)** | Splits on a hierarchy of separators then merges small chunks up to a target size |
70+
| **Recursive (merge-small-only)** | Same recursive split but only merges undersized chunks, preserving natural boundaries |
71+
| **Page** | Splits on page breaks |
7272
| **LLM Regex** | Asks an LLM to generate document-specific regex split patterns |
7373

7474
The recursive splitter lives in [`splitters.py`](src/adaptive_chunking/splitters.py); the others are in [`paper/splitters.py`](src/adaptive_chunking/paper/splitters.py). You can register any callable that takes text and returns a list of chunks.

docs/architecture.svg

Lines changed: 2 additions & 2 deletions
Loading

docs/architecture_v2.svg

Lines changed: 2 additions & 2 deletions
Loading

0 commit comments

Comments
 (0)