RE-MOVE: An Adaptive Policy Design Approach for Dynamic Environments via Language-Based Feedback
Paper • 2303.07622 • Published
Chakraborty
souradip24
AI & ML interests
Reinforcement Learning, Machine Learning, NLP
Recent Activity
updated a model 29 days ago
souradip24/dpo-merged-vllm-r4-r3 published a model 29 days ago
souradip24/dpo-merged-vllm-r4-r3 updated a model about 1 month ago
souradip24/dpo-llama-3.2-3b-set2-samp500-pref100Organizations
authored 9
papers 11 months ago
Rethinking Adversarial Policies: A Generalized Attack Formulation and Provable Defense in RL
Paper • 2305.17342 • Published
PARL: A Unified Framework for Policy Alignment in Reinforcement Learning
Paper • 2308.02585 • Published
MaxMin-RLHF: Towards Equitable Alignment of Large Language Models with Diverse Human Preferences
Paper • 2402.08925 • Published • 1
Is poisoning a real threat to LLM alignment? Maybe more so than you think
Paper • 2406.12091 • Published
Transfer Q Star: Principled Decoding for LLM Alignment
Paper • 2405.20495 • Published
Immune: Improving Safety Against Jailbreaks in Multi-modal LLMs via Inference-Time Alignment
Paper • 2411.18688 • Published
Align-Pro: A Principled Approach to Prompt Optimization for LLM Alignment
Paper • 2501.03486 • Published
BalancedDPO: Adaptive Multi-Metric Alignment
Paper • 2503.12575 • Published