WARP recovers training domain mixtures from fine-tuned model weights using weight-space interpolation via model merging to generate pseudo-checkpoints and geometric features mapped to proportions.
Localize-and-stitch: Efficient model merging via sparse task arithmetic
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4representative citing papers
SAEs used for layer selection with raw task vectors outperform subspace projection and raise math reasoning accuracy on Gemma-3-4B-IT.
DiM3 is a direction- and magnitude-aware merging method that composes heterogeneous multilingual and multimodal updates in LLM backbones, outperforming baselines on 57-language benchmarks while retaining multimodal performance.
HeteroFusion fuses heterogeneous LLMs via topology-based alignment and conflict-aware denoising, outperforming merging and ensemble baselines in cross-family and multi-source settings.
citing papers explorer
-
WARP: Weight-Space Analysis for Recovering Training Data Portfolios
WARP recovers training domain mixtures from fine-tuned model weights using weight-space interpolation via model merging to generate pseudo-checkpoints and geometric features mapped to proportions.
-
Interpretability-Guided Layer Selection over Subspace Projection: SAEs as Stethoscopes, Not Scalpels, for Raw Task Vector Model Editing
SAEs used for layer selection with raw task vectors outperform subspace projection and raise math reasoning accuracy on Gemma-3-4B-IT.
-
DiM\textsuperscript{3}: Bridging Multilingual and Multimodal Models via Direction- and Magnitude-Aware Merging
DiM3 is a direction- and magnitude-aware merging method that composes heterogeneous multilingual and multimodal updates in LLM backbones, outperforming baselines on 57-language benchmarks while retaining multimodal performance.
-
Can Heterogeneous Language Models Be Fused?
HeteroFusion fuses heterogeneous LLMs via topology-based alignment and conflict-aware denoising, outperforming merging and ensemble baselines in cross-family and multi-source settings.