editing models via task vectors

**Editing Models via Task Vectors** is a **model modification framework that decomposes fine-tuned model knowledge into portable, composable vectors** — enabling transfer, removal, and combination of learned behaviors by manipulating these vectors in weight space. **Key Operations** - **Extraction**: $ au = heta_{fine} - heta_{pre}$ (extract what fine-tuning learned). - **Transfer**: Apply $ au$ from model $A$ to model $B$: $ heta_B' = heta_B + au_A$. - **Forgetting**: $ heta' = heta_{fine} - lambda au$ (partially undo fine-tuning for selective forgetting). - **Analogy**: If $ au_{EN ightarrow FR}$ maps English→French, apply it to other models for similar translation ability. **Why It Matters** - **Modular ML**: Neural network capabilities become modular, composable units. - **Efficient Transfer**: Transfer specific capabilities without full fine-tuning. - **Debiasing**: Remove biased behavior by subtracting the corresponding task vector. **Editing via Task Vectors** is **modular surgery for neural networks** — extracting, transplanting, and removing capabilities as portable weight-space operations.