Pith. sign in

REVIEW 7 cited by

Model merging with SVD to tie the Knots

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 2410.19735 v1 pith:AXLEP5R2 submitted 2024-10-25 cs.CV

classification cs.CV
keywords modelsknotsloramergingmodelalignmentfinetunedfully-finetuned
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Recent model merging methods demonstrate that the parameters of fully-finetuned models specializing in distinct tasks can be combined into one model capable of solving all tasks without retraining. Yet, this success does not transfer well when merging LoRA finetuned models. We study this phenomenon and observe that the weights of LoRA finetuned models showcase a lower degree of alignment compared to their fully-finetuned counterparts. We hypothesize that improving this alignment is key to obtaining better LoRA model merges, and propose KnOTS to address this problem. KnOTS uses the SVD to jointly transform the weights of different LoRA models into an aligned space, where existing merging methods can be applied. In addition, we introduce a new benchmark that explicitly evaluates whether merged models are general models. Notably, KnOTS consistently improves LoRA merging by up to 4.3% across several vision and language benchmarks, including our new setting. We release our code at: https://github.com/gstoica27/KnOTS.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 7 Pith papers

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

  1. DA-MergeLoRA: Hypernetwork-Based LoRA Merging for Few-Shot Test-Time Domain Adaptation

    cs.CV 2026-07 conditional novelty 6.0 of 10

    A hypernetwork generates per-column merging weights to combine source LoRA modules on CLIP, achieving state-of-the-art few-shot test-time domain adaptation.

  2. Model soups need only one ingredient

    cs.LG 2026-02 conditional novelty 6.0 of 10

    A single checkpoint, edited by splitting each layer's update with SVD and reweighting the high- and low-energy parts, reaches soup-level OOD robustness without multi-model training.

  3. Unraveling LoRA Interference: Orthogonal Subspaces for Robust Model Merging

    cs.CL 2025-05 conditional novelty 6.0 of 10

    Initializing LoRA's A matrix with the smallest-eigenvalue eigenvectors of other tasks' feature covariance reduces interference when merging task-specific LoRA models, improving average merged accuracy.

  4. GRIP: In-Parameter Graph Reasoning through Fine-Tuning Large Language Models

    cs.CL 2025-11 reject novelty 5.0 of 10

    An LLM can memorize a knowledge graph into LoRA weights and answer relation/reasoning queries about it without graph context, but the evaluation partly trains on the test task.

  5. Onsager Principle-Based Domain Embedding for Thermodynamically Consistent Cahn-Hilliard Model in Arbitrary Domain

    math.NA 2025-08 unverdicted novelty 5.0 of 10

    As submitted, the paper is unverifiable: the abstract claims a thermodynamically consistent Cahn-Hilliard domain-embedding derivation, while the full text is a different paper on person re-identification.

  6. ICM-Fusion: In-Context Meta-Optimized LoRA Fusion for Multi-Task Adaptation

    cs.CV 2025-08 reject novelty 5.0 of 10

    ICM-Fusion uses a conditional VAE plus task-vector guidance to fuse multiple LoRA adapters into one model, reporting marginal average gains on vision and language benchmarks and larger gains in a few-shot long-tail setup.

  7. Tensorized Clustered LoRA Merging for Multi-Task Interference

    cs.LG 2025-08 unverdicted novelty 5.0 of 10

    Clustering training data by embedding similarity and jointly CP-decomposing LoRA adapters cuts multi-task merging interference: +1.4% on Phi-3 and +2.3% on Mistral-7B over SVD baselines.

Pith tools