preference-lab / server

Commit History

refactor: apply production readiness recommendations including dataset caching, XSS protection, pure schemas, and JSON decoding logic.
5ee1380

Sibam commited on

fix: update Guided Demo output to show Anthropic dataset instead of synthetic placeholder
8fbc201

Sibam commited on

feat: add primary descriptive headline and ensure entirely emoji-free UI
c3314b1

Sibam commited on

style: remove emojis from the web interface for a professional look
350b447

Sibam commited on

feat: transform custom UI into an intelligent Agent Learning dashboard with progress metrics and reasoning
c3d75c0

Sibam commited on

fix: update Gradio polling logic to use Timer instead of deprecated every kwarg
e71b4ea

Sibam commited on

fix: remove unsupported width/height kwargs from Gradio LinePlot
a27c1c3

Sibam commited on

feat: download real datasets during docker build and mount live reward graph in Gradio UI
e0b4973

Sibam commited on

fix: clamp grader rewards to strictly (0, 1) to pass OpenEnv validation bounds
f3f7bc4

Sibam commited on

feat: update inference, models, and server logic
447a565

Sibam commited on

final: submission ready
a4c268d

Sibam commited on

fix: update API_BASE_URL to router.huggingface.co, openenv validate passes
14c1b69

Sibam commited on

fix: conform to OpenEnv base interface contract
7574c9a

Sibam commited on

Enhancement
dada51b

Sibam commited on

PreferenceLab OpenEnv environment for RLHF preference simulation
cdf485e

Sibam commited on