diff --git a/docs-site/index.html b/docs-site/index.html index 03f82b2..5268d99 100644 --- a/docs-site/index.html +++ b/docs-site/index.html @@ -5,8 +5,8 @@ CogniFold — A brain-inspired always-on agent memory - - + + @@ -17,178 +17,190 @@ - - + + - + - - - -
-
+
+
- -
-
- OPENNORVE · NORVELABS - APACHE-2.0 · PYTHON · MCP-COMPATIBLE -
+
+ OpenNorve · NorveLabs + Apache-2.0 · Python · MCP-compatible +
-
+ +
- -
-
BRAIN-INSPIRED AGENT MEMORY
-

CogniFold

-

An always-on agent memory that folds events into self-emerging cognitive structure — and proactively crystallizes them into intent.

-
- - - GITHUB - - - arXiv - 2605.13438 - + +
+
Brain-inspired agent memory
+

CogniFold

+

An always-on agent memory that folds events into self-emerging cognitive structure — and proactively crystallizes them into intent.

+ +
+
+
+ 19 + memory systems mapped
-
-
19memory systems mapped
-
~53%memory-system coverage
-
4.6×event → concept compression
-
0.614proactivity score
+
+ ~53% + memory-system coverage
-
- - -
-
DESIGN PHILOSOPHY
-

Memory consolidates — it doesn’t search.

-

Human memory doesn’t look things up — it settles. Borrowing the brain’s complementary learning systems, CogniFold writes events down fast, replays what recurs into durable concepts, and turns that understanding into goals and next steps. Nothing is fetched on demand: structure emerges, and the agent gradually learns what matters to you and what’s worth acting on.

-
-
COMPLEMENTARY LEARNING SYSTEMS
-
HIPPOCAMPUSWrites new events down fast, the moment they happen.
-
NEOCORTEXSettles what keeps recurring into durable concepts.
-
PREFRONTALTurns that understanding into goals and next steps.
+
+ 4.6× + event → concept compression +
+
+ 0.614 + proactivity score
+
- -
-
-
RESULTS
-
CogniFold’s own figures
+ +
+
Design philosophy
+

Memory consolidates — it doesn't search.

+

Human memory doesn't look things up — it settles. Borrowing the brain's complementary learning systems, CogniFold writes events down fast, replays what recurs into durable concepts, and turns that understanding into goals and next steps. Nothing is fetched on demand: structure emerges, and the agent gradually learns what matters to you and what's worth acting on.

+
+
Complementary learning systems
+
+
+ Hippocampus + Writes new events down fast, the moment they happen. +
+
+ Neocortex + Settles what keeps recurring into durable concepts. +
+
+ Prefrontal + Turns that understanding into goals and next steps.
-
+
+ +
+
+ Results + CogniFold's own figures +
+
-
- -
-
+
- -
-
-
Systems · circuits
-
Where memory lives
-
-
-
-
-
Communications · subspaces
-
How memory moves
-
-
-
-
+ +
- -
-
THE FOLDING LOOP
-

One map, three node types.

+ +
+
+
Systems · circuits
+
Where memory lives
+
+
+
+
Communications · subspaces
+
How memory moves
+
+
+
- -
- CogniFold brain: event, concept and intent over a complementary-learning-systems graph -
+ +
+
The folding loop
+

One map, three node types.

+
+ CogniFold brain: event, concept and intent over a complementary-learning-systems graph +
+
+
+
+
Eventα
+
Hippocampal
+
Verbatim episodes captured the moment they happen — what occurred, when, and to whom.
- -
-
-
EVENTα
-
Hippocampal
-
Verbatim episodes captured the moment they happen — what occurred, when, and to whom.
-
-
→
-
-
CONCEPTβ
-
Neocortical
-
Repeated events fold into grounded, generalized knowledge with explainable provenance.
-
-
→
-
-
INTENTγ
-
Prefrontal
-
Concepts and time anchors crystallize into proactive intent — remembering to act, by topology.
-
+
→
+
+
Conceptβ
+
Neocortical
+
Repeated events fold into grounded, generalized knowledge with explainable provenance.
-
Time is the fourth axis, threading through every layer · 4.6× compression · 0.614 proactivity
- -
-
Memory-system coverage~53%
-
-
-
-
- COVERED 7 - PARTIAL 6 - PLANNED 6 -
+
→
+
+
Intentγ
+
Prefrontal
+
Concepts and time anchors crystallize into proactive intent — remembering to act, by topology.
+
Time is the fourth axis, threading through every layer · 4.6× compression · 0.614 proactivity
+
- -
-
Behavioral · phenomenological
-
What memory does
-
-
+ +
+
+
Behavioral · phenomenological
+
What memory does
+
-
- -
- COGNIFOLD · APACHE-2.0 · arXiv:2605.13438 · OPENNORVE / NORVELABS -
+
+ + +
+
+ Memory-system coverage + ~53% +
+
+
+
+
+
+
+ Covered 7 + Partial 6 + Planned 6 +
+
+ +
CogniFold · Apache-2.0 · arXiv:2605.13438 · OpenNorve / NorveLabs
@@ -196,20 +208,20 @@

${esc(r.maps)}` : ''; + const maps = r.maps ? `${esc(r.maps)}` : ''; // Systems · circuits rows (those carrying a `fig`) drive the panel image + bottom-left // caption; every other row keeps the floating cursor tooltip. let attrs = ''; if ('fig' in r) attrs = ` data-cap="${esc(r.e || '')}" data-fig="${esc(r.fig || '')}"`; else if (r.e) attrs = ` data-tip="${esc(r.e)}"`; - return `
${esc(r.name)}${maps}
`; + return `
${esc(r.name)}${maps}
`; } function fill(sel, rows){ const n=document.querySelector(sel); if(n) n.innerHTML = rows.map(rowHTML).join(''); } @@ -264,36 +276,35 @@

` -
+
-
${b.k}
-
${b.t}
+
${esc(b.k)}
+
${esc(b.t)}
-
${b.v}${b.u?`${b.u}`:''}
+
${esc(b.v)}${b.u?`${esc(b.u)}`:''}
`).join(''); // ---- hover tooltips on memory-system rows (plain-language explanations) ---- (function(){ const tip = document.createElement('div'); - tip.style.cssText = 'position:fixed; z-index:99; max-width:300px; padding:9px 13px; border-radius:10px; background:#1d1d1f; color:#f4f2ee; font-family:Inter,system-ui,sans-serif; font-size:12.5px; line-height:1.5; font-weight:300; box-shadow:0 8px 30px rgba(0,0,0,.25); pointer-events:none; opacity:0; transform:translateY(2px); transition:opacity .12s, transform .12s;'; + tip.style.cssText = 'position:fixed; z-index:99; max-width:300px; padding:9px 13px; border-radius:12px; background:#1d1d1f; color:#f5f5f7; font-size:12.5px; line-height:1.5; box-shadow:0 8px 30px rgba(0,0,0,.25); pointer-events:none; opacity:0; transform:translateY(2px); transition:opacity .12s, transform .12s;'; document.body.appendChild(tip); function move(e){ const pad = 16, r = tip.getBoundingClientRect(); @@ -345,20 +356,6 @@