REVIEW 7 cited by
HydraLoRA: An Asymmetric LoRA Architecture for Efficient Fine-Tuning
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
read the original abstract
Adapting Large Language Models (LLMs) to new tasks through fine-tuning has been made more efficient by the introduction of Parameter-Efficient Fine-Tuning (PEFT) techniques, such as LoRA. However, these methods often underperform compared to full fine-tuning, particularly in scenarios involving complex datasets. This issue becomes even more pronounced in complex domains, highlighting the need for improved PEFT approaches that can achieve better performance. Through a series of experiments, we have uncovered two critical insights that shed light on the training and parameter inefficiency of LoRA. Building on these insights, we have developed HydraLoRA, a LoRA framework with an asymmetric structure that eliminates the need for domain expertise. Our experiments demonstrate that HydraLoRA outperforms other PEFT approaches, even those that rely on domain knowledge during the training and inference phases.
Forward citations
Cited by 7 Pith papers
-
CoMoL: Efficient Mixture of LoRA Experts via Dynamic Core Space Merging
CoMoL represents every LoRA expert as a shared-basis core matrix and merges token-selected experts in that core space, reaching standard LoRA parameter counts while outperforming MoE-LoRA baselines on math and code.
-
FedTreeLoRA: Reconciling Statistical and Functional Heterogeneity in Federated LoRA Fine-Tuning
A federated LoRA fine-tuning method that builds a client-similarity tree and adapts aggregation depth layer-by-layer outperforms flat or global aggregation baselines on NLU and NLG tasks.
-
PLoP: Precise LoRA Placement for Efficient Finetuning of Large Models
PLoP selects LoRA adapter placement by ranking normalized feature norms and placing adapters on the lowest-scoring module types, using only forward passes.
-
SOS-LoRA: Static Orthogonal-Subspace Low-Rank Adaptation with Fixed Multi-Scale Scaling
A LoRA update split into several fixed, differently-scaled low-rank experts with orthogonal input directions improves fine-tuning accuracy at the same parameter count.
-
Advancing Compositional LLM Reasoning with Structured Task Relations in Interactive Multimodal Communications
ContextLoRA partitions a single LLM's LoRA parameters by task and trains them in dependency-graph order, while ContextGear adds pipeline parallelism to cut training time, together outperforming multi-task LoRA baselin...
-
FLoE: Fisher-Based Layer Selection for Efficient Sparse Adaptation of Low-Rank Experts
FLoE uses Fisher information to pick the transformer layers that matter and a Bayesian optimizer to set LoRA rank, cutting trainable parameters while keeping or improving accuracy.
-
Two Is Better Than One: Rotations Scale LoRAs
RadarGate rotates LoRA outputs before mixing them, expanding the set of reachable outputs beyond plain weighted sums.
Discussion (0). Sign in to comment.