SSR-Merge merges LoRAs via subspace construction, inverse correlation decorrelation, and directional steering, shown to match the OLS solution with a streaming implementation that outperforms prior merging methods.
hub
Mixture of lora experts.arXiv preprint arXiv:2404.13628, 2024a
17 Pith papers cite this work. Polarity classification is still indexing.
17
Pith papers citing it
hub tools
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
LoRA-Key creates a standalone user-specific Watermark LoRA trained with a latent watermark prior and GOP, attachable via training-free superposition to protect LoRA ownership while preserving quality.
ScaleEarth conditions remote sensing VLMs on continuous GSD via CS-HLoRA and a visual GSD predictor, creating a closed training loop with GeoScale-VQA to achieve SOTA on Earth observation benchmarks.
Training-free Riemannian fusion merges orthogonal style and concept adapters for diffusion models via geodesic approximation on GS matrices plus spectra restoration.
UQ4CT integrates functional-level uncertainty calibration into mixture-of-experts LoRA fine-tuning via a dedicated loss, cutting expected calibration error by over 25% on multiple-choice and generative QA tasks.
LLMs recover dominant binomial orders from corpora but align less closely with exact preference distributions, with preference strength partially encoded in middle-to-late layers and manipulable via steering.
Sleep-time Knowledge Seeding plus Dreaming lets LLMs expand capacity, distill fragile in-context memories into stable parameters, and self-improve without human labels.
Knowledge conflicts in hypernetwork LLM adaptation stem from constant adapter margins losing to frequency-dependent pretrained margins; selective layer boosting and conflict-aware triggering raise deep-conflict accuracy to 71-72.5% on Gemma-2B and Mistral-7B.
Red-Bandit adapts online to LLM failure modes by dynamically selecting among RL-trained LoRA attack-style experts via a bandit policy, reporting SOTA ASR@10 on AdvBench with lower-perplexity prompts.
LoRA-Mixer routes modular LoRA experts into attention projection matrices with an adaptive Routing Specialization Loss to improve multi-task performance while using fewer trainable parameters than prior LoRA-MoE methods.
BiRG-LoRA achieves 69.31% macro-average accuracy across CMB, CMExam, MedQA, and MedMCQA, outperforming MoELoRA by 0.89 points with 28.1% fewer trainable parameters under a matched Qwen3-8B protocol.
Mixture-of-Control adaptively combines local and global control states in transformer fine-tuning by treating per-block states as experts in a sparse MoE setup to improve cross-block communication while keeping memory and compute costs comparable to prior state-based methods.
TriageRA-CCF combines source-side confidence, coverage, and counterfactual signals to supervise an adaptive LoRA rank router, reporting modest average accuracy gains over LoRA/DoRA/MoELoRA baselines on two 8B models under matched training.
CRMA adds a spectrally bounded residual adapter backbone to modular continual fine-tuning of LLMs, achieving near-zero loss drift and positive backward transfer on Mistral-7B across domains.
FediLoRA is a lightweight federated LoRA aggregation method that jointly mitigates missing modalities and heterogeneous ranks in collaborative fine-tuning of foundation models.
A comprehensive survey of PEFT algorithms for large models, covering their performance, overhead, applications, and real-world system implementations.
An overview revisits LoRA variants by categorizing advances in architectural design, efficient optimization, and applications while linking them to classical signal processing tools for principled fine-tuning.
citing papers explorer
-
SSR-Merge: Subspace Signal Routing for Training-Free LoRA Merging in Diffusion Models
SSR-Merge merges LoRAs via subspace construction, inverse correlation decorrelation, and directional steering, shown to match the OLS solution with a streaming implementation that outperforms prior merging methods.
-
LoRA-Key: User-Centric LoRA Watermarking for Text-to-Image Diffusion Models
LoRA-Key creates a standalone user-specific Watermark LoRA trained with a latent watermark prior and GOP, attachable via training-free superposition to protect LoRA ownership while preserving quality.
-
Beyond GSD-as-Token: Continuous Scale Conditioning for Remote Sensing VLMs
ScaleEarth conditions remote sensing VLMs on continuous GSD via CS-HLoRA and a visual GSD predictor, creating a closed training loop with GeoScale-VQA to achieve SOTA on Earth observation benchmarks.
-
OrthoFuse: Training-free Riemannian Fusion of Orthogonal Style-Concept Adapters for Diffusion Models
Training-free Riemannian fusion merges orthogonal style and concept adapters for diffusion models via geodesic approximation on GS matrices plus spectra restoration.
-
Functional-level Uncertainty Quantification for Calibrated Fine-tuning on LLMs
UQ4CT integrates functional-level uncertainty calibration into mixture-of-experts LoRA fine-tuning via a dedicated loss, cutting expected calibration error by over 25% on multiple-choice and generative QA tasks.
-
Behavioral and Representational Evidence of Binomial Ordering Preferences in Large Language Models
LLMs recover dominant binomial orders from corpora but align less closely with exact preference distributions, with preference strength partially encoded in middle-to-late layers and manipulable via steering.
-
Language Models Need Sleep: Learning to Self-Modify and Consolidate Memories
Sleep-time Knowledge Seeding plus Dreaming lets LLMs expand capacity, distill fragile in-context memories into stable parameters, and self-improve without human labels.
-
The Override Gap: A Magnitude Account of Knowledge Conflict Failure in Hypernetwork-Based Instant LLM Adaptation
Knowledge conflicts in hypernetwork LLM adaptation stem from constant adapter margins losing to frequency-dependent pretrained margins; selective layer boosting and conflict-aware triggering raise deep-conflict accuracy to 71-72.5% on Gemma-2B and Mistral-7B.
-
Red-Bandit: Test-Time Adaptation for LLM Red-Teaming via Bandit-Guided LoRA Experts
Red-Bandit adapts online to LLM failure modes by dynamically selecting among RL-trained LoRA attack-style experts via a bandit policy, reporting SOTA ASR@10 on AdvBench with lower-perplexity prompts.
-
LoRA-Mixer: Coordinate Modular LoRA Experts Through Serial Attention Routing
LoRA-Mixer routes modular LoRA experts into attention projection matrices with an adaptive Routing Specialization Loss to improve multi-task performance while using fewer trainable parameters than prior LoRA-MoE methods.
-
Clinically Structured Rank-Gated LoRA for Cross-Benchmark Medical Question Answering
BiRG-LoRA achieves 69.31% macro-average accuracy across CMB, CMExam, MedQA, and MedMCQA, outperforming MoELoRA by 0.89 points with 28.1% fewer trainable parameters under a matched Qwen3-8B protocol.
-
Mixture-of-Control: State-Aware Fine-Tuning for Transformer-based Models
Mixture-of-Control adaptively combines local and global control states in transformer fine-tuning by treating per-block states as experts in a sparse MoE setup to improve cross-block communication while keeping memory and compute costs comparable to prior state-based methods.
-
TriageRA-CCF: Source-Side Clinical Confidence and Coverage Signals for Adaptive Rank Budgeting in Medical LLMs
TriageRA-CCF combines source-side confidence, coverage, and counterfactual signals to supervise an adaptive LoRA rank router, reporting modest average accuracy gains over LoRA/DoRA/MoELoRA baselines on two 8B models under matched training.
-
CRMA: A Spectrally-Bounded Backbone for Modular Continual Fine-Tuning of LLMs
CRMA adds a spectrally bounded residual adapter backbone to modular continual fine-tuning of LLMs, achieving near-zero loss drift and positive backward transfer on Mistral-7B across domains.
-
FediLoRA: Practical Federated Fine-Tuning of Foundation Models Under Missing-Modality Constraints
FediLoRA is a lightweight federated LoRA aggregation method that jointly mitigates missing modalities and heterogeneous ranks in collaborative fine-tuning of foundation models.
-
Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey
A comprehensive survey of PEFT algorithms for large models, covering their performance, overhead, applications, and real-world system implementations.
-
Low-Rank Adaptation Redux for Large Models
An overview revisits LoRA variants by categorizing advances in architectural design, efficient optimization, and applications while linking them to classical signal processing tools for principled fine-tuning.