Picture for Traian Rebedea

Traian Rebedea

National University of Science and Technology POLITEHNICA Bucharest, NVIDIA

Meta-learning how to Share Credit among Macro-Actions

Add code
Jun 16, 2025
Viaarxiv icon

MultiMatch: Multihead Consistency Regularization Matching for Semi-Supervised Text Classification

Add code
Jun 09, 2025
Viaarxiv icon

Safety Through Reasoning: An Empirical Study of Reasoning Guardrail Models

Add code
May 26, 2025
Viaarxiv icon

Aegis2.0: A Diverse AI Safety Dataset and Risks Taxonomy for Alignment of LLM Guardrails

Add code
Jan 15, 2025
Figure 1 for Aegis2.0: A Diverse AI Safety Dataset and Risks Taxonomy for Alignment of LLM Guardrails
Figure 2 for Aegis2.0: A Diverse AI Safety Dataset and Risks Taxonomy for Alignment of LLM Guardrails
Figure 3 for Aegis2.0: A Diverse AI Safety Dataset and Risks Taxonomy for Alignment of LLM Guardrails
Figure 4 for Aegis2.0: A Diverse AI Safety Dataset and Risks Taxonomy for Alignment of LLM Guardrails
Viaarxiv icon

GIT-CXR: End-to-End Transformer for Chest X-Ray Report Generation

Add code
Jan 05, 2025
Figure 1 for GIT-CXR: End-to-End Transformer for Chest X-Ray Report Generation
Figure 2 for GIT-CXR: End-to-End Transformer for Chest X-Ray Report Generation
Figure 3 for GIT-CXR: End-to-End Transformer for Chest X-Ray Report Generation
Figure 4 for GIT-CXR: End-to-End Transformer for Chest X-Ray Report Generation
Viaarxiv icon

Towards Inference-time Category-wise Safety Steering for Large Language Models

Add code
Oct 02, 2024
Figure 1 for Towards Inference-time Category-wise Safety Steering for Large Language Models
Figure 2 for Towards Inference-time Category-wise Safety Steering for Large Language Models
Figure 3 for Towards Inference-time Category-wise Safety Steering for Large Language Models
Figure 4 for Towards Inference-time Category-wise Safety Steering for Large Language Models
Viaarxiv icon

"Vorbeşti Româneşte?" A Recipe to Train Powerful Romanian LLMs with English Instructions

Add code
Jun 26, 2024
Viaarxiv icon

Unsupervised Extraction of Dialogue Policies from Conversations

Add code
Jun 21, 2024
Figure 1 for Unsupervised Extraction of Dialogue Policies from Conversations
Figure 2 for Unsupervised Extraction of Dialogue Policies from Conversations
Figure 3 for Unsupervised Extraction of Dialogue Policies from Conversations
Figure 4 for Unsupervised Extraction of Dialogue Policies from Conversations
Viaarxiv icon

OpenLLM-Ro -- Technical Report on Open-source Romanian LLMs

Add code
May 17, 2024
Figure 1 for OpenLLM-Ro -- Technical Report on Open-source Romanian LLMs
Figure 2 for OpenLLM-Ro -- Technical Report on Open-source Romanian LLMs
Figure 3 for OpenLLM-Ro -- Technical Report on Open-source Romanian LLMs
Figure 4 for OpenLLM-Ro -- Technical Report on Open-source Romanian LLMs
Viaarxiv icon

CantTalkAboutThis: Aligning Language Models to Stay on Topic in Dialogues

Add code
Apr 04, 2024
Viaarxiv icon