goncalorafaria / qalignLinks
QAlign is a new test-time alignment approach that improves language model performance by using Markov chain Monte Carlo methods.
☆26Updated last month
Alternatives and similar repositories for qalign
Users that are interested in qalign are comparing it to the libraries listed below
Sorting:
- Optimizing Causal LMs through GRPO with weighted reward functions and automated hyperparameter tuning using Optuna☆59Updated 3 months ago
- ☆55Updated last year
- ☆68Updated 8 months ago
- Verifiers for LLM Reinforcement Learning☆80Updated 9 months ago
- Train your own SOTA deductive reasoning model☆107Updated 10 months ago
- Simple GRPO scripts and configurations.☆59Updated 11 months ago
- [ACL 2024] Do Large Language Models Latently Perform Multi-Hop Reasoning?☆88Updated 10 months ago
- A framework for pitting LLMs against each other in an evolving library of games ⚔☆34Updated 9 months ago
- Implementation of the paper: "AssistantBench: Can Web Agents Solve Realistic and Time-Consuming Tasks?"☆69Updated last year
- A framework for pitting LLMs against each other in an evolving library of games ⚔☆34Updated 9 months ago
- Matrix (Multi-Agent daTa geneRation Infra and eXperimentation framework) is a versatile engine for multi-agent conversational data genera…☆260Updated last week
- Code for our paper PAPILLON: PrivAcy Preservation from Internet-based and Local Language MOdel ENsembles☆61Updated 8 months ago
- ☆91Updated last month
- OpenCoconut implements a latent reasoning paradigm where we generate thoughts before decoding.☆175Updated last year
- Anchored Preference Optimization and Contrastive Revisions: Addressing Underspecification in Alignment☆61Updated last year
- ☆95Updated last week
- j1-micro (1.7B) & j1-nano (600M) are absurdly tiny but mighty reward models.☆101Updated 6 months ago
- ☆123Updated 11 months ago
- Source code for the collaborative reasoner research project at Meta FAIR.☆112Updated 9 months ago
- Archon provides a modular framework for combining different inference-time techniques and LMs with just a JSON config file.☆189Updated 10 months ago
- Simple repository for training small reasoning models☆48Updated 11 months ago
- Training an LLM to use a calculator with multi-turn reinforcement learning, achieving a **62% absolute increase in evaluation accuracy**.☆65Updated 8 months ago
- ☆59Updated last year
- Official repository for "BLEUBERI: BLEU is a surprisingly effective reward for instruction following"☆31Updated 7 months ago
- Small, simple agent task environments for training and evaluation☆19Updated last year
- ☆61Updated 7 months ago
- Learning to Retrieve by Trying - Source code for Grounding by Trying: LLMs with Reinforcement Learning-Enhanced Retrieval☆51Updated last year
- Universal Reasoning Model☆121Updated 2 weeks ago
- ☆25Updated 8 months ago
- Python package for generating datasets to evaluate reasoning and retrieval of large language models☆19Updated 4 months ago