benpry / why-think-step-by-step
Code and data for the paper "Why think step by step? Reasoning emerges from the locality of experience"
☆59Updated last year
Alternatives and similar repositories for why-think-step-by-step:
Users that are interested in why-think-step-by-step are comparing it to the libraries listed below
- Learning to Retrieve by Trying - Source code for Grounding by Trying: LLMs with Reinforcement Learning-Enhanced Retrieval☆32Updated 4 months ago
- ☆110Updated 2 weeks ago
- Understanding the correlation between different LLM benchmarks☆29Updated last year
- Anchored Preference Optimization and Contrastive Revisions: Addressing Underspecification in Alignment☆55Updated 6 months ago
- Q-Probe: A Lightweight Approach to Reward Maximization for Language Models☆40Updated 9 months ago
- ☆59Updated 10 months ago
- Code for RATIONALYST: Pre-training Process-Supervision for Improving Reasoning https://arxiv.org/pdf/2410.01044☆32Updated 5 months ago
- Minimal implementation of the Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models paper (ArXiv 20232401.01335)☆29Updated last year
- ☆73Updated 6 months ago
- Repository for the paper Stream of Search: Learning to Search in Language☆139Updated last month
- ☆48Updated 4 months ago
- ☆95Updated 8 months ago
- Flow of Reasoning: Training LLMs for Divergent Problem Solving with Minimal Examples☆76Updated last week
- Implementation of the paper: "AssistantBench: Can Web Agents Solve Realistic and Time-Consuming Tasks?"☆52Updated 3 months ago
- This is the official repository for Inheritune.☆109Updated last month
- Code for the arXiv preprint "The Unreasonable Effectiveness of Easy Training Data"☆46Updated last year
- ☆101Updated last month
- ☆27Updated 2 weeks ago
- ☆11Updated 7 months ago
- ☆60Updated last month
- ☆15Updated 5 months ago
- OpenCoconut implements a latent reasoning paradigm where we generate thoughts before decoding.☆167Updated last month
- Code for PHATGOOSE introduced in "Learning to Route Among Specialized Experts for Zero-Shot Generalization"☆81Updated last year
- Framework and toolkits for building and evaluating collaborative agents that can work together with humans.☆68Updated 3 weeks ago