(ICLR '26) Safety Subspaces are Not Linearly Distinct: A Fine-Tuning Case Study
☆12Oct 4, 2025Updated last year
Alternatives and similar repositories for safety-subspaces
Users that are interested in safety-subspaces are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- This is the unofficial re-implementation of "Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-t…☆13Jul 14, 2025Updated last year
- (ICLR '26) ABBA-Adapters: Efficient and Expressive Fine-Tuning of Foundation Models☆24Sep 25, 2025Updated last year
- (TMLR J2C Certification) Fed-SB: A Silver Bullet for Extreme Communication Efficiency and Performance in (Private) Federated LoRA Fine-Tu…☆31Oct 4, 2025Updated last year
- (ACL '25 - Oral) FedEx-LoRA: Exact Aggregation for Federated and Efficient Fine-Tuning of Foundation Models☆38Oct 4, 2025Updated last year
- Github repo for NeurIPS 2024 paper "Safe LoRA: the Silver Lining of Reducing Safety Risks when Fine-tuning Large Language Models"☆32Dec 21, 2025Updated 10 months ago
- Virtual machines for every use case on DigitalOcean • AdGet dependable uptime with 99.99% SLA, simple security tools, and predictable monthly pricing with DigitalOcean's virtual machines, called Droplets.
- Initialization using Update Approximation is a Silver Bullet for Extremely Efficient Low-Rank Fine-Tuning☆52Oct 17, 2025Updated last year
- Code for the paper "AsFT: Anchoring Safety During LLM Fune-Tuning Within Narrow Safety Basin".☆38Jul 10, 2025Updated last year
- [ICML 2025] Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning☆12Jun 10, 2025Updated last year
- ☆20Jun 21, 2025Updated last year
- NeurIPS'24 - LLM Safety Landscape☆41Oct 21, 2025Updated last year
- This is the official code for the paper "Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning" (NeurIPS2024)☆30Sep 10, 2024Updated 2 years ago
- Inference-time alignment for harmlessness through cross-model guidance (ACL 2024). Code + MM-Harmful Bench.☆38Oct 2, 2024Updated 2 years ago
- [NeurIPS 2025] Panacea: Mitigating Harmful Fine-tuning for Large Language Models via Post-fine-tuning Perturbation.☆14Jul 28, 2026Updated 3 months ago
- This is the official code for the paper "Vaccine: Perturbation-aware Alignment for Large Language Models" (NeurIPS2024)☆54Jan 15, 2026Updated 9 months ago
- Wordpress hosting with auto-scaling - Free Trial Offer • AdFully Managed hosting for WordPress and WooCommerce businesses that need reliable, auto-scalable performance. Cloudways SafeUpdates now available.
- Official implementation of "EDITS: Enhancing Dataset Distillation with Implicit Textual Semantics" (Accepted by ICASSP 2026).☆12May 11, 2026Updated 5 months ago
- An implementation of SEAL: Safety-Enhanced Aligned LLM fine-tuning via bilevel data selection.☆24Feb 20, 2025Updated last year
- Fine-tuning Recipes for Apertus☆40Oct 12, 2026Updated 3 weeks ago
- A survey on harmful fine-tuning attack for large language model (ACM CSUR)☆254Sep 6, 2026Updated last month
- Toolkit to benchmark the tamper-resistance of LLMs.☆35May 15, 2026Updated 5 months ago
- [EMNLP 2025] Layer-Aware Representation Filtering: Purifying Finetuning Data to Preserve LLM Safety Alignment☆17Jul 22, 2025Updated last year
- Code for “SaLoRA: Safety-Alignment Preserved Low-Rank Adaptation(ICLR 2025)”☆30Oct 23, 2025Updated last year
- DP-FTRL from "Practical and Private (Deep) Learning without Sampling or Shuffling" for centralized training.☆39Jul 17, 2026Updated 3 months ago
- ☆11Apr 25, 2021Updated 5 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.
- This is the official code for the paper "Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturba…☆43Mar 22, 2025Updated last year
- Released Code for ACL 21 paper: DocOIE A Document-level Context-Aware Dataset for OpenIE☆15Nov 25, 2022Updated 3 years ago
- The official code repo for "Safe Delta: Consistently Preserving Safety when Fine-Tuning LLMs on Diverse Datasets" in ICML 2025.☆59Feb 12, 2026Updated 8 months ago
- ☆24Dec 8, 2024Updated last year
- ☆14Apr 22, 2024Updated 2 years ago
- Restore safety in fine-tuned language models through task arithmetic☆33Mar 28, 2024Updated 2 years ago
- ☆37Oct 10, 2024Updated 2 years ago
- ☆37Oct 27, 2026Updated last week
- Code for the paper "Self-Detoxifying Language Models via Toxification Reversal" (EMNLP 2023)☆19Oct 17, 2023Updated 3 years ago
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- Official Repository for The Paper: Safety Alignment Should Be Made More Than Just a Few Tokens Deep☆194Apr 23, 2025Updated last year
- clone from agramfort of scikit-learn: machine learning in Python☆12Mar 16, 2022Updated 4 years ago
- code space of paper "Safety Layers in Aligned Large Language Models: The Key to LLM Security" (ICLR 2025)☆27Apr 26, 2025Updated last year
- (Nanochat based) Full-bandwidth transformer, feed LLM's own deep hidden states back to itself☆16Oct 3, 2026Updated last month
- ☆18Aug 18, 2026Updated 2 months ago
- [ICLR 2026] Fewer Weights, More Problems: A Practical Attack on LLM Pruning☆14Apr 6, 2026Updated 6 months ago
- The loss landscape of Large Language Models resemble basin!☆49Jul 8, 2025Updated last year