Cornell-RL / tril
☆122Updated 7 months ago
Related projects: ⓘ
- ☆65Updated 2 months ago
- Lamorel is a Python library designed for RL practitioners eager to use Large Language Models (LLMs).☆186Updated 6 months ago
- Official code from the paper "Offline RL for Natural Language Generation with Implicit Language Q Learning"☆197Updated last year
- ☆131Updated 4 months ago
- ☆87Updated 2 months ago
- Intrinsic Motivation from Artificial Intelligence Feedback☆117Updated 10 months ago
- Efficient World Models with Context-Aware Tokenization. ICML 2024☆73Updated 2 months ago
- ☆198Updated last year
- We perform functional grounding of LLMs' knowledge in BabyAI-Text☆213Updated 3 weeks ago
- (ICML 2024) Alphazero-like Tree-Search can guide large language model decoding and training☆175Updated 3 months ago
- Repository for the code of the "PPL-MCTS: Constrained Textual Generation Through Discriminator-Guided Decoding" paper, NAACL'22☆62Updated last year
- Reasoning with Language Model is Planning with World Model☆137Updated last year
- Interpreting how transformers simulate agents performing RL tasks☆62Updated 10 months ago
- ☆158Updated last year
- Code for Contrastive Preference Learning (CPL)☆147Updated 6 months ago
- Research Code for "ArCHer: Training Language Model Agents via Hierarchical Multi-Turn RL"☆84Updated 5 months ago
- An extensible benchmark for evaluating large language models on planning☆248Updated 3 months ago
- [NeurIPS '23 Spotlight] Thought Cloning: Learning to Think while Acting by Imitating Human Thinking☆249Updated 2 months ago
- AdaPlanner: Language Models for Decision Making via Adaptive Planning from Feedback☆82Updated last year
- SmartPlay is a benchmark for Large Language Models (LLMs). Uses a variety of games to test various important LLM capabilities as agents. …☆115Updated 5 months ago
- ☆24Updated 2 weeks ago
- ☆117Updated 8 months ago
- Official implementation of the DECKARD Agent from the paper "Do Embodied Agents Dream of Pixelated Sheep?"☆84Updated last year
- [NeurIPS 2023] We use large language models as commonsense world model and heuristic policy within Monte-Carlo Tree Search, enabling bett…☆141Updated 3 months ago
- ScienceWorld is a text-based virtual environment centered around accomplishing tasks from the standardized elementary science curriculum.☆202Updated 2 months ago
- Code accompanying the paper Pretraining Language Models with Human Preferences☆173Updated 7 months ago
- Emergent world representations: Exploring a sequence model trained on a synthetic task☆165Updated last year
- Efficient baselines for autocurricula in JAX.☆165Updated 3 weeks ago
- A repository for transformer critique learning and generation☆84Updated 9 months ago
- ☆120Updated 2 months ago