[ICLR 2026] PSFT is a trust-region–inspired fine-tuning objective that views SFT as a policy gradient method with constant advantages, constraining policy drift to stabilize training and improve generalization.
☆38Sep 9, 2025Updated 8 months ago
Alternatives and similar repositories for PSFT
Users that are interested in PSFT are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Tooling for exact and MinHash deduplication of large-scale text datasets☆84Mar 24, 2026Updated 2 months ago
- [NAACL'25 🏆 SAC Award] Official code for "Advancing MoE Efficiency: A Collaboration-Constrained Routing (C2R) Strategy for Better Expert…☆16Feb 4, 2025Updated last year
- Open-source evaluation toolkit of large vision-language models (LVLMs), support ~100 VLMs, 30+ benchmarks☆15Feb 17, 2025Updated last year
- Towards a Unified View of Large Language Model Post-Training☆211Sep 8, 2025Updated 8 months ago
- [COLM 2025: 1st Workshop on the Application of LLM Explainability to Reasoning and Planning] Latent Chain-of-Thought? Decoding the Depth-…☆18Oct 4, 2025Updated 7 months ago
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- This is the official code repository for the paper: Towards General Continuous Memory for Vision-Language Models.☆26Jul 3, 2025Updated 10 months ago
- ☆16Jul 12, 2024Updated last year
- Entropy-Driven GRPO with Guided Error Correction for Advantage Diversity☆22Aug 28, 2025Updated 8 months ago
- The Search for Sparse, Robustness Neural Networks☆11Mar 24, 2023Updated 3 years ago
- Official implementation for Text Generation Beyond Discrete Token Sampling☆25Aug 11, 2025Updated 9 months ago
- FLoRA: A Framework for Learning Scoring Rules in Autonomous Driving Planning Systems☆13Apr 12, 2026Updated last month
- ☆12Apr 25, 2025Updated last year
- 2022 秋季学期清华大学电子系数据与算法课程 OJ 参考解答☆10Jun 18, 2023Updated 2 years ago
- Official implementation for DenseMixer: Improving MoE Post-Training with Precise Router Gradient☆67Aug 3, 2025Updated 9 months ago
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Reproduced the DFT method without using Verl. https://arxiv.org/abs/2508.05629☆23Oct 14, 2025Updated 7 months ago
- Official implementation of Latent-SFT: teaching LLMs to reason with vocabulary-space latent chains.☆48May 18, 2026Updated last week
- Learning from preferences is a common paradigm for fine-tuning language models. Yet, many algorithmic design decisions come into play. Ou…☆32Apr 20, 2024Updated 2 years ago
- decision-making processes of human drivers☆13Mar 28, 2024Updated 2 years ago
- [ICLR 2022] Denoising Likelihood Score Matching for Conditional Score-based Data Generation☆11Jan 2, 2025Updated last year
- ACL'2025: SoftCoT: Soft Chain-of-Thought for Efficient Reasoning with LLMs. and preprint: SoftCoT++: Test-Time Scaling with Soft Chain-of…☆89May 30, 2025Updated 11 months ago
- Text Adventure Learning Environment Suite - Benchmark to evaluate language models on interactive text environments.☆30May 9, 2026Updated 2 weeks ago
- ☆13Aug 18, 2022Updated 3 years ago
- [ICLRW'26] EoRA: Fine-tuning-free Compensation for Compressed LLM with Eigenspace Low-Rank Approximation☆45Apr 21, 2026Updated last month
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- An optimization-based algorithm to accurately estimate the causal effects and robustly predict under distribution shifts. It leverages th…☆14Jul 10, 2024Updated last year
- ☆21Jul 23, 2025Updated 10 months ago
- Official implementation of "Reasoning by Superposition: A Theoretical Perspective on Chain of Continuous Thought" (NeurIPS 2025)☆39Oct 8, 2025Updated 7 months ago
- Long Is More for Alignment: A Simple but Tough-to-Beat Baseline for Instruction Fine-Tuning [ICML 2024]☆21May 2, 2024Updated 2 years ago
- ☆13May 4, 2026Updated 3 weeks ago
- Code for EMNLP2022 Findings paper "An Error-Guided Correction Model for Chinese Spelling Error Correction"☆14Mar 6, 2023Updated 3 years ago
- A simple code base for Gaussian Splatting research☆21Nov 6, 2024Updated last year
- Project Page of Paper "Drive in Corridors: Enhancing the Safety of End-to-end Autonomous Driving via Corridor Learning and Planning"☆29May 8, 2025Updated last year
- Debug DeepSpeed-Chat step by step in IDE (在IDE里一步一步调试DeepSpeed-Chat)☆10Apr 17, 2023Updated 3 years ago
- End-to-end encrypted cloud storage - Proton Drive • AdSpecial offer: 40% Off Yearly / 80% Off First Month. Protect your most important files, photos, and documents from prying eyes.
- The repository of EMNLP 2023 "A Frustratingly Easy Plug-and-Play Detection-and-Reasoning Module for Chinese Spelling Check"☆20Nov 17, 2023Updated 2 years ago
- ☆47Jun 24, 2025Updated 11 months ago
- NeurIPS'24 - LLM Safety Landscape☆40Oct 21, 2025Updated 7 months ago
- ☆18Mar 15, 2021Updated 5 years ago
- ☆30May 22, 2025Updated last year
- An implementation of <Group Fisher Pruning for Practical Network Compression> based on pytorch and mmcv☆18Nov 21, 2021Updated 4 years ago
- PESLA - TORCS Deep Reinforcement Learning Agent☆10Oct 20, 2019Updated 6 years ago