Pith. sign in

REVIEW 7 cited by

Mixture-of-Subspaces in Low-Rank Adaptation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2406.11909 v4 pith:X2E3PX63 submitted 2024-06-16 cs.LG cs.AI

classification cs.LGcs.AI
keywords loramosloraadaptationlow-rankmethodmixermixture-of-subspacessubspaces
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

In this paper, we introduce a subspace-inspired Low-Rank Adaptation (LoRA) method, which is computationally efficient, easy to implement, and readily applicable to large language, multimodal, and diffusion models. Initially, we equivalently decompose the weights of LoRA into two subspaces, and find that simply mixing them can enhance performance. To study such a phenomenon, we revisit it through a fine-grained subspace lens, showing that such modification is equivalent to employing a fixed mixer to fuse the subspaces. To be more flexible, we jointly learn the mixer with the original LoRA weights, and term the method Mixture-of-Subspaces LoRA (MoSLoRA). MoSLoRA consistently outperforms LoRA on tasks in different modalities, including commonsense reasoning, visual instruction tuning, and subject-driven text-to-image generation, demonstrating its effectiveness and robustness. Codes are available at https://github.com/wutaiqiang/MoSLoRA.

Discussion (0). Sign in to comment.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. SOLAR: Communication-Efficient Model Adaptation via Subspace-Oriented Latent Adapter Reparametrization

    cs.LG 2026-04 unverdicted novelty 6.0 of 10

    SOLAR compresses PEFT adapters by representing task-specific updates as linear combinations of the foundation model's singular vectors with controlled perturbations, reducing communication costs while preserving perfo...

  2. Little by Little: Continual Learning via Incremental Mixture of Rank-1 Associative Memory Experts

    cs.LG 2025-06 unverdicted novelty 6.0 of 10

    MoRAM frames continual learning as incremental addition of rank-1 adapters viewed as self-activating key-value associative memory units in a mixture-of-experts setup.

  3. Little by Little: Continual Learning via Incremental Mixture of Rank-1 Associative Memory Experts

    cs.LG 2025-06 conditional novelty 6.0 of 10

    MoRAM learns continually by adding small rank-1 adapters that act as associative memory items, using input-key similarity to retrieve and mix only the relevant adapters at test time.

  4. Polynomial Expansion Rank Adaptation: Enhancing Low-Rank Fine-Tuning with High-Order Interactions

    cs.LG 2026-04 unverdicted novelty 5.0 of 10

    PERA augments low-rank adaptation by expanding factors with polynomial terms to capture higher-order interactions, yielding better benchmark performance than prior linear methods without raising rank or inference cost.

  5. SOLAR: A Self-Optimizing Open-Ended Autonomous Agent for Lifelong Learning and Continual Adaptation

    cs.AI 2026-03 unverdicted novelty 5.0 of 10

    SOLAR introduces a self-optimizing agent using meta-learning on model weights and RL-driven strategy discovery for lifelong adaptation in LLMs, claiming superior performance on reasoning tasks across domains.

  6. Regularizing Subspace Redundancy of Low-Rank Adaptation

    cs.CV 2025-07 conditional novelty 5.0 of 10

    ReSoRA adds a penalty that reduces redundancy among rank-1 subspaces of LoRA-style adapters, producing modest accuracy improvements on vision-language retrieval and visual classification.

  7. Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey

    cs.LG 2024-03 accept novelty 4.0 of 10

    A comprehensive survey of PEFT algorithms for large models, covering their performance, overhead, applications, and real-world system implementations.

Pith tools