影响力指数

88.36/100

前 0.7%

全站排名 #447

发表论文27 篇

平均评分5.7

年均产出9.0 篇/年

Ya Zhang

Professor@Shanghai Jiao Tong University·中国·OpenReview

研究方向

Generalist medical artificial intelligence · Medical image analysis · Computer vision · Machine learning · Information retrieval

6.5

Overthinking Reduction with Decoupled Rewards and Curriculum Data Scheduling

ICLR 2026Oral

三作

6.0

Improving Diffusion Models for Class-imbalanced Training Data via Capacity Manipulation

ICLR 2026Oral

5.0

Combatting Dimensional Collapse in LLM Pre-Training Data via Submodular File Selection

ICLR 2025Oral

7.5

Fine-tuning with Reserved Majority for Noise Reduction

ICLR 2025Spotlight

三作

7.3

Learning to Instruct for Visual Instruction Tuning

NeurIPS 2025Poster

6.8

RAD: Towards Trustworthy Retrieval-Augmented Multi-modal Clinical Diagnosis

NeurIPS 2025Poster

6.6

ConText: Driving In-context Learning for Text Removal and Segmentation

ICML 2025Poster

通讯

6.4

SaFiRe: Saccade-Fixation Reiteration with Mamba for Referring Image Segmentation

NeurIPS 2025Poster

6.1

MoMa: Modulating Mamba for Adapting Image Foundation Models to Video Recognition

ICML 2025Poster

5.4

G4Seg: Generation for Online Segmentation Refinement with Diffusion Models

ICLR 2025Withdrawn

4.8

Continual Task Learning through Adaptive Policy Self-Composition

ICLR 2025Withdrawn

4.5

Protecting Minorities in Diffusion Models via Capacity Allocation

合作者 (20)

Ya Zhang

Overthinking Reduction with Decoupled Rewards and Curriculum Data Scheduling

Improving Diffusion Models for Class-imbalanced Training Data via Capacity Manipulation

Wide-In, Narrow-Out: Revokable Decoding for Efficient and Effective DLLMs

The State of Reinforcement Finetuning for Transformer-based Agents

Automatic Reward Shaping from Multi-Objective Human Heuristics

Q-learning Penalized Transformer for Safe Offline Reinforcement Learning

Visual Programmability: A Guide for Code-as-Thought in Chart Understanding

Combatting Dimensional Collapse in LLM Pre-Training Data via Submodular File Selection

Fine-tuning with Reserved Majority for Noise Reduction

Learning to Instruct for Visual Instruction Tuning

RAD: Towards Trustworthy Retrieval-Augmented Multi-modal Clinical Diagnosis

ConText: Driving In-context Learning for Text Removal and Segmentation

SaFiRe: Saccade-Fixation Reiteration with Mamba for Referring Image Segmentation

MoMa: Modulating Mamba for Adapting Image Foundation Models to Video Recognition

G4Seg: Generation for Online Segmentation Refinement with Diffusion Models

Continual Task Learning through Adaptive Policy Self-Composition

Protecting Minorities in Diffusion Models via Capacity Allocation