Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks20 from public data
- Efficient Image Super-Resolution Using Pixel Attention474
This work designs a lightweight convolutional neural network for image super resolution with a newly proposed pixel attention scheme that could achieve similar performance as the lightweight networks - SRResNet and CARN, but with only 272K parameters.
- ClassSR: A General Framework to Accelerate Super-Resolution Networks by Data Characteristic223
A new solution pipeline that combines classification and SR in a unified framework that can help most existing methods save up to 50% FLOPs on DIV8K datasets and introduces a new classification method with two losses – Class-Loss and Average- Loss to produce the classification results.
- AIM 2020 Challenge on Efficient Super-Resolution: Methods and Results101
The AIM 2020 challenge on efficient single image super-resolution was to super-resolve an input image with a magnification factor x4 based on a set of prior examples of low and corresponding high resolution images with focus on the proposed solutions and results.
- Temporally consistent video colorization with deep feature propagation and self-regularization learning67
A novel temporally consistent video colorization (TCVC) framework that effectively propagates frame-level deep features in a bidirectional way to enhance the temporal consistency of colorization and introduces a self-regularization learning (SRL) scheme to minimize the differences in predictions obtained using different time steps.
- Evaluating the Generalization Ability of Super-Resolution Networks34
This work makes the first attempt to propose a Generalization Assessment Index for Super-Resolution (SR) networks, namely SRGA, which exploits the statistical characteristics of the internal features of deep networks to measure the generalization ability.
- Very Lightweight Photo Retouching Network With Conditional Sequential Modulation34
An extremely lightweight framework – Conditional Sequential Retouching Network (CSRNet) that achieves state-of-the-art performance on the benchmark MIT-Adobe FiveK dataset quantitively and qualitatively and can be easily extended to learn local enhancement effects.
- SCT: A Simple Baseline for Parameter-Efficient Fine-Tuning via Salient Channels32
Experiments outperform full fine-tuning on 18 out of 19 tasks in the VTAB-1K benchmark by adding only 0.11M parameters of the ViT-B, demonstrating the proposed tuning technique’s strong capability and effectiveness in the low-data regime.
- 21
- 12
- Revisit Parameter-Efficient Transfer Learning: A Two-Stage Paradigm7
A novel two-stage paradigm, where the pre-trained model is first aligned to the target distribution and the task-relevant information is leveraged for effective adaptation, achieves state-of-the-art performance on the average accuracy of 19 downstream tasks.
- 5
- 5
- LLaVA-CMoE: Towards Continual Mixture of Experts for Large Vision-Language Models4
This work proposes LLaVA-CMoE, a continual learning framework for LLMs that requires no replay data of previous tasks and ensures both parameter efficiency and robust knowledge retention, and introduces a Probe-Guided Knowledge Extension mechanism, enabling adaptive and minimal parameter expansion tailored to task complexity.
- Sci-PRM: A Tool Aware Process Reward Model for Scientific Reasoning Verification2
Experiments demonstrate that Sci-PRM significantly enhances foundation models in two key aspects: it enables effective test-time scaling via Best-of-N selection; and when integrated into Reinforcement Learning, it serves as a dense reward signal that mitigates the critical issue of advantage disappearance, allowing the model to break through existing performance ceilings.
- 2
- 1
- –
- ReCrit: Transition-Aware Reinforcement Learning for Scientific Critic Reasoning–
ReCrit is proposed, a transition-aware reinforcement learning framework that decomposes Initial-to-Critic behavior into four quadrants: Correction, Sycophancy, Robustness, and Boundary, which rewards correction and robustness, penalizes sycophancy, and treats persistent errors as weak boundary signals.
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.