Heykuki News
Top
New
Best
Ask
Show
Jobs
Toggle theme
Login
Top
New
Best
Ask
Show
Jobs
1.
▲
Show HN: Long-horizon LLM coherence benchmark (500 cycles)
(zenodo.org)
2 points
teugent
5 months ago
discuss
2.
▲
Show HN: Sigma Runtime – model-agnostic identity control for LLMs
(github.com/sigmastratum)
2 points
teugent
5 months ago
discuss
3.
▲
Show HN: Sigma v0.4.7: 200 conversational cycles, zero drift, epistemic honesty
(github.com/sigmastratum)
2 points
teugent
5 months ago
discuss
4.
▲
Show HN: Sigma Runtime – A Cognitive Runtime Standard for LLMs
(wiki.sigmastratum.org)
1 point
teugent
6 months ago
1 comment
5.
▲
Show HN: Sigma Runtime – Maintaining 100% Fact Integrity over 120 LLM Cycles
(github.com/sigmastratum)
1 point
teugent
4 months ago
discuss
6.
▲
Show HN: 500-cycle runtime test for long-horizon LLM coherence
(zenodo.org)
1 point
teugent
4 months ago
discuss
7.
▲
Show HN: Sigma Runtime – 550-cycle identity stability benchmark on GPT-5.2
(github.com/sigmastratum)
2 points
teugent
5 months ago
discuss
8.
▲
Most LLM cost isn't compute – it's identity drift (110-cycle GPT-4o benchmark)
(github.com/sigmastratum)
1 point
teugent
6 months ago
1 comment
9.
▲
Show HN: Sigma Runtime ERI (v0.1) – Open benchmark for attractor-based LLMs
(github.com/sigmastratum)
1 point
teugent
6 months ago
discuss
10.
▲
Show HN: Sigma Runtime ERI – 800-line open cognitive runtime for LLM continuity
(github.com/sigmastratum)
1 point
teugent
6 months ago
discuss