Reinforcement Learning for LLMs LLM Reasoning and Architecture

Can isolating task-specific parameters prevent multi-task fine-tuning interference?

Explores whether identifying and protecting task-specific parameter regions can prevent the performance degradation that occurs when fine-tuning models on multiple tasks simultaneously. This matters because it could enable safe multi-task adaptation without sacrificing individual task performance.

Note · 2026-02-22 · sourced from Training Fine Tuning
How do you build domain expertise into general AI models? How should we allocate compute budget at inference time? How should researchers navigate LLM reasoning research?

Core Parameter Isolation Fine-Tuning (CPI-FT) starts from the hypothesis that the "seesaw effect" in multi-task fine-tuning — where improving one task degrades others — arises from parameter heterogeneity: distinct capabilities rely on specific, potentially overlapping parameter subsets, and uniform updates cause destructive interference.

The framework has five steps:

  1. Independent fine-tuning per task to identify each task's "core parameter region" (parameters with largest update magnitudes)
  2. Task clustering by core region overlap — tasks sharing similar core regions are grouped for joint training
  3. Parameter transplantation — core parameters from individually fine-tuned models are directly transplanted into a unified backbone
  4. SLERP fusion — non-core parameters are merged via Spherical Linear Interpolation, enabling geometry-aware blending that avoids abrupt transitions
  5. Pipeline SFT with core regions frozen to prevent catastrophic forgetting

The key finding: full multi-task SFT (uniform parameter updates across all tasks) consistently achieves the lowest performance across all tasks and configurations. Even heuristic multi-stage approaches (training tasks in sequence) only partially mitigate interference. CPI-FT consistently outperforms both — revealing that temporal task scheduling alone is insufficient without explicit structural parameter isolation.

This extends the sparse subnetwork finding from RL. Does reinforcement learning update only a small fraction of parameters? showed that RL naturally converges on sparse parameter subsets. CPI-FT shows that SFT exhibits the same structure — task-relevant changes are concentrated in specific regions — and that explicitly identifying and protecting these regions enables better multi-task adaptation.

The connection to catastrophic forgetting in safety alignment is direct: "Can model developers allow users to fine-tune their aligned models on custom datasets while retaining safety?" CPI-FT suggests yes — by freezing the core parameter regions associated with safety alignment during subsequent task-specific fine-tuning.


Source: Training Fine Tuning

Related concepts in this collection

Concept map
15 direct connections · 137 in 2-hop network ·dense cluster

Click a node to walk · click center to open · click Open full network for a force-directed map

your link semantically near linked from elsewhere
Original note title

core parameter isolation prevents multi-task fine-tuning interference by identifying task-specific regions and merging non-core parameters geometrically