Official repository of "Localizing Task Information for Improved Model Merging and Compression" [ICML 2024]
☆53Dec 22, 2025Updated 3 months ago
Alternatives and similar repositories for tall_masks
Users that are interested in tall_masks are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Official repository of "LiNeS: Post-training Layer Scaling Prevents Forgetting and Enhances Model Merging"☆32Nov 4, 2024Updated last year
- Code for paper "Parameter Efficient Multi-task Model Fusion with Partial Linearization"☆25Sep 13, 2024Updated last year
- Official codebase for AdaRank: Adaptive Rank Pruning for Enhanced Model Merging (ICLR 2026)☆19Jan 26, 2026Updated 2 months ago
- Representation Surgery for Multi-Task Model Merging. ICML, 2024.☆47Oct 10, 2024Updated last year
- Task Singular Vectors: Reducing Task Interference in Model Merging. Merge models avoiding task interference through separable models.☆53Dec 15, 2025Updated 3 months ago
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- [NeurIPS 2024] For paper Parameter Competition Balancing for Model Merging☆48Oct 11, 2024Updated last year
- ☆212Feb 3, 2024Updated 2 years ago
- [ICML 2025] No Task Left Behind: Isotropic Model Merging with Common and Task-Specific Subspaces (official repository)☆41Aug 7, 2025Updated 7 months ago
- AdaMerging: Adaptive Model Merging for Multi-Task Learning. ICLR, 2024.☆103Oct 28, 2024Updated last year
- [NeurIPS 2024 Spotlight] EMR-Merging: Tuning-Free High-Performance Model Merging☆77Mar 1, 2025Updated last year
- MAP: Low-compute Model Merging with Amortized Pareto Fronts via Quadratic Approximation☆14Sep 2, 2024Updated last year
- ☆33Jul 8, 2024Updated last year
- ☆19Jun 10, 2024Updated last year
- Editing Models with Task Arithmetic☆534Jan 11, 2024Updated 2 years ago
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- [NeurIPS 2025] MergeBench: A Benchmark for Merging Domain-Specialized LLMs