Danau5tin / calculator_agent_rlLinks
Training an LLM to use a calculator with multi-turn reinforcement learning, achieving a **62% absolute increase in evaluation accuracy**.
☆53Updated 5 months ago
Alternatives and similar repositories for calculator_agent_rl
Users that are interested in calculator_agent_rl are comparing it to the libraries listed below
Sorting:
- OpenCoconut implements a latent reasoning paradigm where we generate thoughts before decoding.☆172Updated 8 months ago
- ☆68Updated 4 months ago
- ☆133Updated 6 months ago
- Train your own SOTA deductive reasoning model☆107Updated 6 months ago
- Simple GRPO scripts and configurations.☆59Updated 8 months ago
- Official repo for Learning to Reason for Long-Form Story Generation☆72Updated 5 months ago
- Archon provides a modular framework for combining different inference-time techniques and LMs with just a JSON config file.☆184Updated 6 months ago
- Optimizing Causal LMs through GRPO with weighted reward functions and automated hyperparameter tuning using Optuna☆55Updated 8 months ago
- [ACL 2024] Do Large Language Models Latently Perform Multi-Hop Reasoning?☆77Updated 6 months ago
- Repository for the paper Stream of Search: Learning to Search in Language☆151Updated 8 months ago
- ☆40Updated last year
- Entropy Based Sampling and Parallel CoT Decoding☆17Updated 11 months ago
- ☆123Updated 7 months ago
- An easy-to-understand framework for LLM samplers that rewind and revise generated tokens☆146Updated 7 months ago
- Training-Ready RL Environments + Evals☆116Updated this week
- Storing long contexts in tiny caches with self-study☆192Updated 3 weeks ago
- ☆54Updated 10 months ago
- Matrix (Multi-Agent daTa geneRation Infra and eXperimentation framework) is a versatile engine for multi-agent conversational data genera…☆94Updated this week
- Functional Benchmarks and the Reasoning Gap☆89Updated last year
- Source code for the collaborative reasoner research project at Meta FAIR.☆103Updated 5 months ago
- look how they massacred my boy☆63Updated 11 months ago
- ☆103Updated 2 weeks ago
- j1-micro (1.7B) & j1-nano (600M) are absurdly tiny but mighty reward models.☆98Updated 2 months ago
- smolLM with Entropix sampler on pytorch☆150Updated 11 months ago
- ☆88Updated last year
- Verifiers for LLM Reinforcement Learning☆74Updated 5 months ago
- Simple repository for training small reasoning models☆40Updated 7 months ago
- The Automated LLM Speedrunning Benchmark measures how well LLM agents can reproduce previous innovations and discover new ones in languag…☆99Updated 2 months ago
- QAlign is a new test-time alignment approach that improves language model performance by using Markov chain Monte Carlo methods.☆24Updated 2 weeks ago
- accompanying material for sleep-time compute paper☆115Updated 5 months ago