Spaces:
Sleeping
Sleeping
Commit History
-
Update grader score precision: Change rounding from 4 to 2 decimal places
6e684d0 -
Enhance ContentModerationEnvironment: Initialize grader score at episode start and compute it consistently
dbc0ad4 -
Update benchmark display: Add clarification on score accuracy and adjust score formatting
dc54ec6 -
Enhance Content Moderation Environment: Add package exports, improve import handling, and update moderation actions
4b94b38
-
Enhance Content Moderation Environment: Update Dockerfile, add client, improve README, and refine environment handling
06dfa9c
-
updating benchmarks
182b62d -
prevent reward hacking
1b8abf6 -
view_thread and view_post each reward only once per post per episode.
f82c8e6 -
polish
f2f2b35 -
clamping scores to be strictly between 0 and 1
7caffe6
-
reset and step fixes
55bd0c2 -
feat: add very hard difficulty tier and interactive demo UI
Add new "very_hard" task pack with adversarial content and benchmark results. Introduce playable demo UI with server endpoints (/demo, /demo/reset, /demo/step) for human testing. Enhance models with reward/done fields and improve inference error handling. BREAKING CHANGE: Entry point changed from `server.app:create_app` to `server.app:app` in openenv.yaml4559388 -
refactoring, polish, and adding very hard data
bf76643
-
refactor: improve inference robustness, update documentation, and enhance Pydantic schema definitions
745c2e2
-
refactor: transition moderation environment to active-post focus with failure tracking and update model benchmarks
f8db5ac
-
update content moderation datasets and increase max_steps in environment configuration
62c8aca -
feat: enhance inference engine with structured logging, conversation history, multi-stage JSON schema fallbacks, and global timeout management.
5b9bf13
-
Initial commit: OpenEnv Content Moderation Environment for Meta PyTorch Hackathon
bab840f