Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks24 from public data
- InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models1,936
InternVL3 incorporates variable visual position encoding (V2PE) to support extended multimodal contexts, employs advanced post-training techniques such as supervised fine-tuning (SFT) and mixed preference optimization (MPO), and adopts test-time scaling strategies alongside an optimized training infrastructure.
- Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling1,813
This work explores the relationship between model scaling and performance, systematically exploring the performance trends in vision encoders, language models, dataset sizes, and test-time configurations and hopes this model contributes to the open-source community by setting new standards for developing and applying multimodal AI systems.
- SEED-X: Multimodal Models with Unified Multi-granularity Comprehension and Generation365
This work presents a unified and versatile foundation model, namely, SEED-X, which is able to model multi-granularity visual semantics for comprehension and generation tasks and demonstrates its effectiveness in handling real-world applications across various domains after instruction tuning.
- Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization270
This work designs an automated preference data construction pipeline to create MMPR, a high-quality, large-scale multimodal reasoning preference dataset and explores integrating PO with MLLMs, developing a simple yet effective method, termed Mixed Preference Optimization (MPO), which boosts multimodal CoT performance.
- Uni-Perceiver: Pre-training Unified Architecture for Generic Perception for Zero-shot and Few-shot Tasks163
Results show that the pre-trained model without any tuning can achieve reasonable performance even on novel tasks, and the performance can be improved to a level close to state-of-the-art methods by conducting prompt tuning on 1% of downstream task data.
- 140
- Complementary Relation Contrastive Distillation110
This paper proposes a novel knowledge distillation method, namely Complementary Relation Contrastive Distillation (CRCD), to transfer the structural knowledge from the teacher to the student and estimates the mutual relation in an anchor-based way and distill the anchorstudent relation under the supervision of its corresponding anchor-teacher relation.
- Layerwise Optimization by Gradient Decomposition for Continual Learning100
This work decomposed the gradient of an old task into a part shared by all old tasks and a part specific to that task, and performs optimization for the gradients of each layer separately rather than the concatenation of all gradients as in previous works.
- Uni-Perceiver v2: A Generalist Model for Large-Scale Vision and Vision-Language Tasks75
This paper proposes Uni-Perceiver v2, which is the first generalist model capable of handling major large-scale vision and vision-language tasks with competitive performance and proposes an effective optimization technique named Task-Balanced Gradient Normalization to ensure stable multi-task learning with an unmixed sampling strategy, helpful for tasks requiring large batch-size training.
- A Deep Learning Method to Detect Foreign Objects for Inspecting Power Transmission Lines51
A deep learning method to detect invading foreign objects for power transmission line inspection is proposed, based on the regression strategy with oriented bounding boxes to accurately predict spatial location and orientation angle of foreign objects, as well as their categories in cluttered backgrounds.
- 50
- SynerGen-VL: Towards Synergistic Image Understanding and Generation with Vision Experts and Token Folding49
To address challenges identified in existing encoder-free unified MLLMs, the token folding mechanism and the vision-expert-based progressive alignment pretraining strategy are introduced, which effectively support high-resolution image understanding while reducing training complexity.
- ZeroGUI: Automating Online GUI Learning at Zero Human Cost38
ZeroGUI is proposed, a scalable, online learning framework for automating GUI Agent training at Zero human cost that integrates VLM-based automatic task generation to produce diverse training goals from the current environment state, and two-stage online reinforcement learning to continuously interact with and learn from GUI environments.
- V2PE: Improving Multimodal Long-Context Capability of Vision-Language Models with Variable Visual Position Encoding36
This work proposes Variable Visual Position Encoding (V2PE), a novel positional encoding approach that employs variable and smaller increments for visual tokens, enabling more efficient management of long multimodal sequences.
- 30
- Power-Llava: Large Language and Vision Assistant for Power Transmission Line Inspection17
Power-LLaVA, the first large language and vision assistant designed to offer professional and reliable inspection services for power transmission line by engaging in dialogues with humans, is introduced and a large-scale and high-quality dataset specialized for the inspection task is constructed.
- 13
- 11
- 3
- 1
- Transmission Lines Monitoring Based on Convolution Neural Network and Edge Computation1
Convolutional neural networks combining with an edge computation platform are used for feature extraction in this paper and comprehensively considers some features, such as complex image scenes, the difficulty in extracting hidden dangers type features, and large target size variation spans.
- –
- –
- SARS-COV-2 Induce Pulmonary Injury from Basic to Clinical Research–
This review analyzes the SARS-CoV-2 biological characteristics, in-sights the effect of alveolar epithelium and its adjacent microvascular endothelium, investigates human host cells immune response and immunothrombosis, and explains clinical manifestations of COVID-19 associated lung injury.
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.