Popular repositories Loading

1. wilddoc-robust-vqa wilddoc-robust-vqa Public

   Robust multimodal document understanding pipeline (retrieval + VQA) on WildDoc with EM/F1 and degradation-based robustness evaluation.

   Python 2

2. rag-evidence-attack-lab rag-evidence-attack-lab Public

   Scientific QA robustness evaluation pipeline for evidence-missing RAG scenarios on PeerQA, with EM/F1 reliability analysis.

   Python 1

3. maze-rl-walker maze-rl-walker Public

   Forked from BrettTheHobbit/474-Coverage-GridWorld

   Robust PPO-based gridworld coverage agent with custom observations, BFS features, enemy prediction, and reward shaping for dynamic maze navigation.

   Python