Update README.md
Browse files
README.md
CHANGED
|
@@ -27,6 +27,8 @@ tags:
|
|
| 27 |
|
| 28 |
# Falconsai/LightDec
|
| 29 |
|
|
|
|
|
|
|
| 30 |
**A lightweight, single-pass, typed, calibrated decision model for agentic systems.** Give it a **state** (text, code or JSON), one or more **typed questions** (`choice`, `noul` yes/no, `score` ordinal) and a closed set of options. It returns a calibrated probability for every option, from one encoder pass per question.
|
| 31 |
|
| 32 |
LightDec is the FalconDec architecture trained with **FalconDec notebook V2** on the **`standard`** data preset. That run adds agent-specific decisions (AgentTrek, Counsel, HotpotQA) to a balanced mix of 58 test tasks across 9 domains. This page is both the **model card** and the **developer guide**: how to load it, test it, and use it as a decision component in an agentic system.
|
|
@@ -971,12 +973,6 @@ Report issues, misroutes or evaluation results through the Community tab of this
|
|
| 971 |
|
| 972 |
---
|
| 973 |
|
| 974 |
-
# Model Card — Falconsai/LightDec/
|
| 975 |
-
|
| 976 |
-
**[View in Model Surgeon](https://surgeon.falcons.ai/?hub=Falconsai/LightDec)**
|
| 977 |
-
|
| 978 |
-
**[View in Model Surgeon](https://surgeon.falcons.ai/?hub=Falconsai/LightDec)**
|
| 979 |
-
|
| 980 |
This card is generated from the surgical record itself; the package's
|
| 981 |
`lineage.intoto.jsonl` is the signed source of truth (verify it free at
|
| 982 |
the Surgeon's public verifier or with the bundled `verify_attestation.py`).
|
|
|
|
| 27 |
|
| 28 |
# Falconsai/LightDec
|
| 29 |
|
| 30 |
+
**[View in Model Surgeon](https://surgeon.falcons.ai/?hub=Falconsai/LightDec)**
|
| 31 |
+
|
| 32 |
**A lightweight, single-pass, typed, calibrated decision model for agentic systems.** Give it a **state** (text, code or JSON), one or more **typed questions** (`choice`, `noul` yes/no, `score` ordinal) and a closed set of options. It returns a calibrated probability for every option, from one encoder pass per question.
|
| 33 |
|
| 34 |
LightDec is the FalconDec architecture trained with **FalconDec notebook V2** on the **`standard`** data preset. That run adds agent-specific decisions (AgentTrek, Counsel, HotpotQA) to a balanced mix of 58 test tasks across 9 domains. This page is both the **model card** and the **developer guide**: how to load it, test it, and use it as a decision component in an agentic system.
|
|
|
|
| 973 |
|
| 974 |
---
|
| 975 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 976 |
This card is generated from the surgical record itself; the package's
|
| 977 |
`lineage.intoto.jsonl` is the signed source of truth (verify it free at
|
| 978 |
the Surgeon's public verifier or with the bundled `verify_attestation.py`).
|