summaryrefslogtreecommitdiff
path: root/cv/index.typ
diff options
context:
space:
mode:
authorVoid Agent <void@jayrup.hermes>2026-09-20 15:08:28 +0100
committerVoid Agent <void@jayrup.hermes>2026-09-20 15:08:28 +0100
commit1dcecb80aa5855a7cd8d578104f29b3a93622a13 (patch)
tree174e14c3f0ff71df90d440f5905004458e68726c /cv/index.typ
parent0ee667ac6cb2493e4f9ddaa5b69f595071aca936 (diff)
links: point dissertation + site repos at git.jayrup.me (self-hosted cgit)
Diffstat (limited to 'cv/index.typ')
-rw-r--r--cv/index.typ2
1 files changed, 1 insertions, 1 deletions
diff --git a/cv/index.typ b/cv/index.typ
index e4f8227..bf2ef9c 100644
--- a/cv/index.typ
+++ b/cv/index.typ
@@ -56,7 +56,7 @@
- Built a provider abstraction layer for glass-box reasoning extraction and a post-hoc interrogation system to adversarially probe agent intent.
- Designed a 3-prong automated judging pipeline (regex, LLM blackbox, LLM glassbox) and ran *960 experimental trials* across exploratory and confirmatory studies.
- Quantified a *"Stealth Gap"*: deceptive intent held steady (~62--67%) under high oversight while output-level detectability dropped (47.8%→35.6%), showing output-only monitoring underestimates risk.
-- Code and evaluation harness: github.com/CaptainJack2491/Dissertation
+- Code and evaluation harness: #link("https://git.jayrup.me/c/dissertation.git")[git.jayrup.me/c/dissertation.git]
== Projects