Spaces:
Running
Running
Update README.md
Browse files
README.md
CHANGED
|
@@ -48,7 +48,7 @@ Discovers that LLM agents exhibit a computational analog of the human sunk-cost
|
|
| 48 |
### Paper 3 — Cognitive Agentic Diversity *(In Progress)*
|
| 49 |
|
| 50 |
**Cognitive Agentic Diversity Beats Monolithic Scaling: Asymmetric Reviewer-Coder Ensembles in Autonomous ML Engineering Loops**
|
| 51 |
-
*Sanskar Jajoo,
|
| 52 |
|
| 53 |
Demonstrates that ensembles of small, role-differentiated models with diverse weight priors outperform single large models "thinking harder." Key finding: a 7B+8B+6.7B MoE-Vote ensemble achieves **67% accuracy** on reasoning puzzles vs **25% for single-agent** — a +167% improvement with zero added parameters, only diversity of weights.
|
| 54 |
|
|
|
|
| 48 |
### Paper 3 — Cognitive Agentic Diversity *(In Progress)*
|
| 49 |
|
| 50 |
**Cognitive Agentic Diversity Beats Monolithic Scaling: Asymmetric Reviewer-Coder Ensembles in Autonomous ML Engineering Loops**
|
| 51 |
+
*Sanskar Jajoo, Neuralchemy Labs, 2026*
|
| 52 |
|
| 53 |
Demonstrates that ensembles of small, role-differentiated models with diverse weight priors outperform single large models "thinking harder." Key finding: a 7B+8B+6.7B MoE-Vote ensemble achieves **67% accuracy** on reasoning puzzles vs **25% for single-agent** — a +167% improvement with zero added parameters, only diversity of weights.
|
| 54 |
|