Add YAML metadata to repo card
Browse files
README.md
CHANGED
|
@@ -1,3 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# Purpose Agent — Self-Improving Agentic Framework via State-Value Evaluation
|
| 2 |
|
| 3 |
A lightweight, modular framework where an LLM agent improves across tasks **without weight updates** — using an RL-inspired self-reflection loop with a "Purpose Function" that evaluates intermediate state improvements.
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: purpose-agent
|
| 3 |
+
license: mit
|
| 4 |
+
language:
|
| 5 |
+
- en
|
| 6 |
+
tags:
|
| 7 |
+
- reinforcement-learning
|
| 8 |
+
- agents
|
| 9 |
+
- self-improving
|
| 10 |
+
- experience-replay
|
| 11 |
+
- llm-as-judge
|
| 12 |
+
- state-value-evaluation
|
| 13 |
+
- memory-augmented
|
| 14 |
+
- react
|
| 15 |
+
- orchestration
|
| 16 |
+
- modular
|
| 17 |
+
pipeline_tag: text-generation
|
| 18 |
+
---
|
| 19 |
+
|
| 20 |
# Purpose Agent — Self-Improving Agentic Framework via State-Value Evaluation
|
| 21 |
|
| 22 |
A lightweight, modular framework where an LLM agent improves across tasks **without weight updates** — using an RL-inspired self-reflection loop with a "Purpose Function" that evaluates intermediate state improvements.
|