HackerReposA FIELD GUIDE TO OPEN SOURCE
ProSponsor the shelf♧ 0
← Back to top all-time
ffr1
unbounded-context-attention
Python2026–2026
ffr1 / unbounded-context-attention

unbounded-context-attention

An attention architecture for unbounded context with provable per-token influence and 100% recall via cryptographic hash retrieval

2Distinct people
2Mentions
0Explicit recommenders
3GitHub stars
REPOSITORY AT A GLANCE

What to know before exploring

What it is

An attention architecture for unbounded context with provable per-token influence and 100% recall via cryptographic hash retrieval

GitHub

Where it is used

No documented product use has been verified yet.

What people are saying

No recommendation phrase was found in the indexed comments.

No classified warning is attached to this page. Read the discussions before treating silence as approval.

Also in these discussions

No other repository is repeatedly linked in the same discussions.

This week on Hacker News

Nov 7–Nov 14 UTC · Direct repository links and separately labeled related project stories. Indexed through 11/14/2026, 3:56:07 PM UTC.

No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.

Mentions over time

HN mentions by hour2 dated mentions

Observed mentions: 2026-05-06 00:18:49 UTC to 2026-05-06 23:54:27 UTC.

012026-05-06 00:00:00 UTC to 2026-05-06 01:00:00 UTC (exclusive): 1 mentions; 1 submissions, 0 commentsMay 6, 00:002026-05-06 01:00:00 UTC to 2026-05-06 02:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 02:00:00 UTC to 2026-05-06 03:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 03:00:00 UTC to 2026-05-06 04:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 04:00:00 UTC to 2026-05-06 05:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 05:00:00 UTC to 2026-05-06 06:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 06:00:00 UTC to 2026-05-06 07:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 07:00:00 UTC to 2026-05-06 08:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 08:00:00 UTC to 2026-05-06 09:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 09:00:00 UTC to 2026-05-06 10:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 10:00:00 UTC to 2026-05-06 11:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 11:00:00 UTC to 2026-05-06 12:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 commentsMay 6, 11:002026-05-06 12:00:00 UTC to 2026-05-06 13:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 13:00:00 UTC to 2026-05-06 14:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 14:00:00 UTC to 2026-05-06 15:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 15:00:00 UTC to 2026-05-06 16:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 16:00:00 UTC to 2026-05-06 17:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 17:00:00 UTC to 2026-05-06 18:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 18:00:00 UTC to 2026-05-06 19:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 19:00:00 UTC to 2026-05-06 20:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 20:00:00 UTC to 2026-05-06 21:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 21:00:00 UTC to 2026-05-06 22:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 22:00:00 UTC to 2026-05-06 23:00:00 UTC (exclusive): 0 mentions; 0 submissions, 0 comments2026-05-06 23:00:00 UTC to 2026-05-07 00:00:00 UTC (exclusive): 1 mentions; 0 submissions, 1 commentsMay 6, 23:00

Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.

View mention counts and date ranges
UTC ranges include their start and exclude their end. Select a period to filter the source evidence.
PeriodStart (UTC)End (exclusive, UTC)MentionsSubmissionsComments
May 6, 00:002026-05-06 00:00:00 UTC2026-05-06 01:00:00 UTC110
May 6, 01:002026-05-06 01:00:00 UTC2026-05-06 02:00:00 UTC000
May 6, 02:002026-05-06 02:00:00 UTC2026-05-06 03:00:00 UTC000
May 6, 03:002026-05-06 03:00:00 UTC2026-05-06 04:00:00 UTC000
May 6, 04:002026-05-06 04:00:00 UTC2026-05-06 05:00:00 UTC000
May 6, 05:002026-05-06 05:00:00 UTC2026-05-06 06:00:00 UTC000
May 6, 06:002026-05-06 06:00:00 UTC2026-05-06 07:00:00 UTC000
May 6, 07:002026-05-06 07:00:00 UTC2026-05-06 08:00:00 UTC000
May 6, 08:002026-05-06 08:00:00 UTC2026-05-06 09:00:00 UTC000
May 6, 09:002026-05-06 09:00:00 UTC2026-05-06 10:00:00 UTC000
May 6, 10:002026-05-06 10:00:00 UTC2026-05-06 11:00:00 UTC000
May 6, 11:002026-05-06 11:00:00 UTC2026-05-06 12:00:00 UTC000
May 6, 12:002026-05-06 12:00:00 UTC2026-05-06 13:00:00 UTC000
May 6, 13:002026-05-06 13:00:00 UTC2026-05-06 14:00:00 UTC000
May 6, 14:002026-05-06 14:00:00 UTC2026-05-06 15:00:00 UTC000
May 6, 15:002026-05-06 15:00:00 UTC2026-05-06 16:00:00 UTC000
May 6, 16:002026-05-06 16:00:00 UTC2026-05-06 17:00:00 UTC000
May 6, 17:002026-05-06 17:00:00 UTC2026-05-06 18:00:00 UTC000
May 6, 18:002026-05-06 18:00:00 UTC2026-05-06 19:00:00 UTC000
May 6, 19:002026-05-06 19:00:00 UTC2026-05-06 20:00:00 UTC000
May 6, 20:002026-05-06 20:00:00 UTC2026-05-06 21:00:00 UTC000
May 6, 21:002026-05-06 21:00:00 UTC2026-05-06 22:00:00 UTC000
May 6, 22:002026-05-06 22:00:00 UTC2026-05-06 23:00:00 UTC000
May 6, 23:002026-05-06 23:00:00 UTC2026-05-07 00:00:00 UTC101

What Hacker News says

2 sources
Related articles; re: rounds and layers: - "Why are neural networks and cryptographic ciphers so similar? (2025)" https://reiner.org/neural-net-ciphers .. https://news.ycombinator.com/item?id=47983467 - "Show HN: ResonanceNet – Proof-of-Training Blockchain" https://news.ycombinator.com/item?id=47386901 - "Cryptographic hashing as a transformer attention head" https://github.com/ffr1/unbounded-context-attention - 'Implementation for NDSS'2025 paper: "TensorCrypt: Repurposing Neural Networks for Efficient Cryptographic Computation"' https://github.com/OSUSecLab/TensorCrypt : > More specifically, with a program translation framework that converts traditional cryptographic algorithms into NN models, our proof-of-concept implementations in TensorFlow demonstrate substantial performance improvements: encryption speeds for AES, Chacha20, and
Read the rest
Salsa20 show increases of up to 4.09×, 5.44×, and 5.06×, respectively, compared to existing GPU-based cryptographic solutions written by human experts.
In Why are neural networks and cryptographic ciphers so similar? (2025)

MAKE YOURSELF AT HOME

A few shortcuts

/
Search the collection
?
Open this guide
Esc
Close dialogs and navigation

Your saved repositories and interests stay in this browser.