Pith. sign in

REVIEW 9 cited by

Parameter-efficient Multi-task Fine-tuning for Transformers via Shared Hypernetworks

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 2106.04489 v1 pith:BE23THTY submitted 2021-06-08 cs.CL

classification cs.CL
keywords tasktasksacrossadapterhypernetworksmodelmulti-taskparameter-efficient
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

State-of-the-art parameter-efficient fine-tuning methods rely on introducing adapter modules between the layers of a pretrained language model. However, such modules are trained separately for each task and thus do not enable sharing information across tasks. In this paper, we show that we can learn adapter parameters for all layers and tasks by generating them using shared hypernetworks, which condition on task, adapter position, and layer id in a transformer model. This parameter-efficient multi-task learning framework allows us to achieve the best of both worlds by sharing knowledge across tasks via hypernetworks while enabling the model to adapt to each individual task through task-specific adapters. Experiments on the well-known GLUE benchmark show improved performance in multi-task learning while adding only 0.29% parameters per task. We additionally demonstrate substantial performance improvements in few-shot domain generalization across a variety of tasks. Our code is publicly available in https://github.com/rabeehk/hyperformer.

Discussion (0). Sign in to comment.

Forward citations

Cited by 9 Pith papers

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

  1. Hypernetwork-Conditioned WENO5 Conservative-Form CNNs for One-Dimensional Conservation Laws

    math.NA 2026-05 conditional novelty 7.0 of 10

    A hypernetwork conditions a conservative-form CNN to predict WENO5 weights from mesh and initial-condition metadata, preserving conservation and generalizing across resolutions for 1D hyperbolic conservation laws.

  2. Retrieval-Augmented Personalization with Foundation Models for Wearable Stress Detection

    cs.LG 2026-06 unverdicted novelty 6.0 of 10

    Retrieval from out-of-domain foundation models enables personalization of a lightweight transformer for stress detection, yielding +3.92% accuracy and +4.76% F1 gains on WESAD without user labels.

  3. UnHype: CLIP-Guided Hypernetworks for Dynamic LoRA Unlearning

    cs.CV 2026-02 conditional novelty 6.0 of 10

    UnHype generates concept-specific LoRA unlearning weights on the fly from CLIP text embeddings by training a hypernetwork to follow the gradient of an unlearning loss, enabling single- and multi-concept erasure in dif...

  4. SOS-LoRA: Static Orthogonal-Subspace Low-Rank Adaptation with Fixed Multi-Scale Scaling

    cs.LG 2026-06 conditional novelty 5.0 of 10

    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.

  5. Deep Reprogramming Distillation for Medical Foundation Models

    cs.CV 2026-05 unverdicted novelty 5.0 of 10

    DRD introduces a reprogramming module and CKA-based distillation to enable efficient, robust adaptation of medical foundation models to downstream 2D/3D classification and segmentation tasks, outperforming prior PEFT ...

  6. (Almost) Free Modality Stitching of Foundation Models

    cs.CV 2025-07 conditional novelty 5.0 of 10

    A hypernetwork that generates connector weights for all image-text model pairs can rank pairs like grid search at about 10x lower training cost, but the best connector lags grid search by a few points.

  7. Capacity and Redundancy Trade-offs in Multi-Task Learning

    cs.LG 2026-07 conditional novelty 4.0 of 10

    A shared representation's total per-task information is bounded by capacity plus label redundancy; clustered sharing wins exactly when interference reduction exceeds the redundancy it loses.

  8. Towards Unified Multi-task EEG Analysis with Low-Rank Adaptation

    cs.LG 2026-04 unverdicted novelty 4.0 of 10

    MTEEG uses task-specific LoRA modules to jointly adapt a pre-trained EEG model across multiple tasks, outperforming single-task baselines on most metrics in evaluations on six downstream tasks.

  9. 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