DRATS derives a minimax objective from a feasibility formulation of MTRL to adaptively sample tasks with the largest return gaps, leading to better worst-task performance on MetaWorld benchmarks.
Hashimoto , title =
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
representative citing papers
VoiceBench is the first benchmark for multi-faceted evaluation of LLM voice assistants using real and synthetic spoken instructions with speaker, environmental, and content variations.
ODRPO decomposes discrete rewards into ordinal binary indicators to create robust, variance-aware advantage estimators for noisy RLAIF in LLM alignment.
citing papers explorer
-
Distributionally Robust Multi-Task Reinforcement Learning via Adaptive Task Sampling
DRATS derives a minimax objective from a feasibility formulation of MTRL to adaptively sample tasks with the largest return gaps, leading to better worst-task performance on MetaWorld benchmarks.
-
VoiceBench: Benchmarking LLM-Based Voice Assistants
VoiceBench is the first benchmark for multi-faceted evaluation of LLM voice assistants using real and synthetic spoken instructions with speaker, environmental, and content variations.
-
ODRPO: Ordinal Decompositions of Discrete Rewards for Robust Policy Optimization
ODRPO decomposes discrete rewards into ordinal binary indicators to create robust, variance-aware advantage estimators for noisy RLAIF in LLM alignment.
- Token-weighted Direct Preference Optimization with Attention