fe1ixxu / CPO_SIMPO
This repository contains the joint use of CPO and SimPO method for better reference-free preference learning methods.
☆52Updated 7 months ago
Alternatives and similar repositories for CPO_SIMPO:
Users that are interested in CPO_SIMPO are comparing it to the libraries listed below
- [ICLR'24 spotlight] Tool-Augmented Reward Modeling☆46Updated 3 months ago
- Official repository for paper "Weak-to-Strong Extrapolation Expedites Alignment"☆73Updated 9 months ago
- Towards Systematic Measurement for Long Text Quality☆34Updated 6 months ago
- Official code for "MAmmoTH2: Scaling Instructions from the Web" [NeurIPS 2024]☆137Updated 5 months ago
- An Experiment on Dynamic NTK Scaling RoPE☆62Updated last year
- ☆98Updated 6 months ago
- Source code of "Reasons to Reject? Aligning Language Models with Judgments"☆58Updated last year
- [ICLR 2024] CLEX: Continuous Length Extrapolation for Large Language Models☆76Updated last year
- [NeurIPS 2024] Train LLMs with diverse system messages reflecting individualized preferences to generalize to unseen system messages☆44Updated 3 months ago
- Official implementation for 'Extending LLMs’ Context Window with 100 Samples'☆75Updated last year
- Official implementation of the paper "From Complex to Simple: Enhancing Multi-Constraint Complex Instruction Following Ability of Large L…☆46Updated 9 months ago
- Code for preprint "Metadata Conditioning Accelerates Language Model Pre-training (MeCo)"☆36Updated last week
- Suri: Multi-constraint instruction following for long-form text generation (EMNLP’24)☆22Updated 4 months ago
- Code implementation of synthetic continued pretraining☆97Updated 2 months ago
- [NAACL 2024 Outstanding Paper] Source code for the NAACL 2024 paper entitled "R-Tuning: Instructing Large Language Models to Say 'I Don't…☆109Updated 8 months ago
- Code and models for EMNLP 2024 paper "WPO: Enhancing RLHF with Weighted Preference Optimization"☆38Updated 6 months ago
- ☆48Updated 11 months ago
- Organize the Web: Constructing Domains Enhances Pre-Training Data Curation☆39Updated last month
- ☆68Updated last year
- We introduce ScaleQuest, a scalable, novel and cost-effective data synthesis method to unleash the reasoning capability of LLMs.☆60Updated 5 months ago
- Reformatted Alignment☆115Updated 6 months ago
- Unofficial implementation of AlpaGasus☆90Updated last year
- ☆17Updated 10 months ago
- Code and data used in the paper: "Training on Incorrect Synthetic Data via RL Scales LLM Math Reasoning Eight-Fold"☆29Updated 9 months ago
- Official repository for MATES: Model-Aware Data Selection for Efficient Pretraining with Data Influence Models [NeurIPS 2024]☆63Updated 4 months ago
- [ACL'24] Superfiltering: Weak-to-Strong Data Filtering for Fast Instruction-Tuning☆147Updated 6 months ago
- ☆23Updated 2 months ago
- [ICLR'25] Data and code for our paper "Why Does the Effective Context Length of LLMs Fall Short?"☆70Updated 4 months ago
- [ACL 2024] Self-Training with Direct Preference Optimization Improves Chain-of-Thought Reasoning☆41Updated 8 months ago
- Repository for "Propagating Knowledge Updates to LMs Through Distillation" (NeurIPS 2023).☆25Updated 7 months ago