docs: honest prior-art positioning in README - #24
Merged
Conversation
ANF is a supporting internal tool, not a novel format or published research. Acknowledge prior art (TOON/TRON/TSLN) and the independent agentic benchmarks (arXiv 2605.29676, 2606.01326) showing compact notations give modest, accuracy-costly gains. State plainly that ANF's savings come from view extraction, not notation. Signed-off-by: Shreyansh Sancheti <43677304+shreyanshjain7174@users.noreply.github.com>
There was a problem hiding this comment.
Pull request overview
Adds an explicit “Prior art and honest positioning” section to the README to clarify that ANF is an internal, supporting tool focused on view extraction rather than a novel serialization format, and to reference related prior work and benchmarks.
Changes:
- Introduces a new README section describing prior art and positioning.
- Adds benchmark citations and clarifies where token savings come from (view extraction vs. notation).
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| ## Prior art and honest positioning | ||
|
|
||
| ANF is not a novel serialization format, and this is not published research. The | ||
| token-efficient-format space is already crowded: TOON, TRON, and TSLN cover |
Comment on lines
+77
to
+80
| approach. The "Notation Matters" agentic benchmark (arXiv 2605.29676) finds | ||
| compact notations save roughly 18-27% inside real agent loops and cost accuracy. | ||
| State-in-context minification (arXiv 2606.01326) reports about 42% with a 12pp | ||
| accuracy drop. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds a short 'Prior art and honest positioning' section. States plainly that ANF is a supporting internal tool, not a novel format or published research, cites TOON/TRON/TSLN and the independent benchmarks (arXiv 2605.29676, 2606.01326), and clarifies that the token savings come from view extraction, not notation. docs check passes.