影响力指数
97.93/100
前 0.1%
全站排名 #60
发表论文58
平均评分5.7
年均产出19.3 篇/年

Xiang Wang

Full Professor@University of Science and Technology of China·中国·OpenReview
研究方向

large language models · trustworthy AI · generative recommendation · graph learning · recommendation · explainable learning

7.0
18

AlphaSteer: Learning Refusal Steering with Principled Null-Space Constraint

ICLR 2026Poster
6.0
23

Reinforced Preference Optimization for Recommendation

ICLR 2026Rejected
通讯
5.5
15

Quantile Advantage Estimation: Stabilizing RLVR for LLM Reasoning

ICLR 2026Poster
5.5
20

DualEdit: Mitigating Safety Fallback in LLM Backdoor Editing via Affirmation-Refusal Regulation

ICLR 2026Poster
5.5
14

GuardAlign: Test-time Safety Alignment in Multimodal Large Language Models

ICLR 2026Poster
5.5
17

Beyond Magnitude: Leveraging Direction of RLVR Updates for LLM Reasoning

ICLR 2026Poster
5.5
20

Denoising Neural Reranker for Recommender Systems

ICLR 2026Poster
通讯
5.3
18

PEA-DPO: Perception-Enhanced Alignment Direct Preference Optimization for MLLMs Alignment

ICLR 2026Rejected
5.0
15

Mitigating Reward Hacking in LLM-based Recommendation: A Preference Optimization Approach

ICLR 2026Rejected
5.0
18

SAFER: Probing Safety in Reward Models with Sparse Autoencoder

ICLR 2026Rejected
通讯
5.0
23

AlphaAlign: Incentivizing Safety Alignment with Extremely Simplified Reinforcement Learning

ICLR 2026Poster
通讯
5.0
24

Look Back to Reason Forward: Revisitable Memory for Long-Context LLM Agents

ICLR 2026Poster
4.5
25

bi-GRPO: Bidirectional Optimization for Jailbreak Backdoor Injection on LLMs

ICLR 2026Rejected
4.5
19

Contrastive Weak-to-Strong Generalization

ICLR 2026Rejected
4.5
22

Enhancing Multi-Modal LLMs Reasoning via Difficulty-Aware Group Normalization

ICLR 2026Rejected
4.5
16

LIMI: Less is More for Agency

ICLR 2026Rejected
4.0
13

Enhancing Multimodal LLMs Reasoning via Perception Reward Modeling

ICLR 2026Rejected
4.0
13

SINGER: Leveraging Semantic Identifier Hierarchies for Generative Recommendation

ICLR 2026Rejected
3.5
20

Attention Localization Through Separator Tokens: Unlocking Long Numerical Sequence Processing in LLMs

ICLR 2026Withdrawn
通讯
3.0
22

Bridging Perception and Reasoning: Token Reweighting for RLVR in Multimodal LLMs

ICLR 2026Rejected
8.9
15

Multi-agent Architecture Search via Agentic Supernet

ICML 2025Oral
通讯
8.3
15

AnyEdit: Edit Any Knowledge Encoded in Language Models

ICML 2025Poster
8.0
32

AlphaEdit: Null-Space Constrained Knowledge Editing for Language Models

ICLR 2025Oral
7.8
7

DAMA: Data- and Model-aware Alignment of Multi-modal LLMs

ICML 2025Poster
7.3
30

Search and Refine During Think: Facilitating Knowledge Refinement for Improved Retrieval-Augmented Reasoning

NeurIPS 2025Poster
通讯
7.3
23

Language Representations Can be What Recommenders Need: Findings and Potentials

ICLR 2025Oral
7.1
31

FACE: A General Framework for Mapping Collaborative Filtering Embeddings into LLM Tokens

NeurIPS 2025Poster
7.0
34

Unified Parameter-Efficient Unlearning for LLMs

ICLR 2025Poster
7.0
8

Reinforced Lifelong Editing for Language Models

ICML 2025Poster
通讯
7.0
20

3D-GSRD: 3D Molecular Graph Auto-Encoder with Selective Re-mask Decoding

NeurIPS 2025Poster
通讯
6.8
31

On Reasoning Strength Planning in Large Reasoning Models

NeurIPS 2025Poster
6.8
27

Towards Unified and Lossless Latent Space for 3D Molecular Latent Diffusion Modeling

NeurIPS 2025Poster
通讯
6.8
27

Think before Recommendation: Autonomous Reasoning-enhanced Recommender

NeurIPS 2025Poster
通讯
6.5
34

Uncertainty-aware Guided Diffusion for Missing Data in Sequential Recommendation

ICLR 2025Rejected
6.4
28

RePO: Understanding Preference Learning Through ReLU-Based Optimization

NeurIPS 2025Poster
通讯
6.4
27

Teaching Language Models to Reason with Tools

NeurIPS 2025Poster
6.4
27

Fading to Grow: Growing Preference Ratios via Preference Fading Discrete Diffusion for Recommendation

NeurIPS 2025Poster
通讯
6.3
24

SciLitLLM: How to Adapt LLMs for Scientific Literature Understanding

ICLR 2025Poster
6.2
30

DiffGAD: A Diffusion-based Unsupervised Graph Anomaly Detector

ICLR 2025Poster
通讯
6.2
30

Towards Robust Alignment of Language Models: Distributionally Robustifying Direct Preference Optimization

ICLR 2025Poster
6.0
32

On Efficiency-Effectiveness Trade-off of Diffusion-based Recommenders

NeurIPS 2025Poster
通讯
5.5
13

Learning Bayesian Nash Equilibrium in Auction Games via Approximate Best Response

ICML 2025Poster
通讯
5.5
39

NExT-Mol: 3D Diffusion Meets 1D Language Modeling for 3D Molecule Generation

ICLR 2025Poster
5.5
15

Larger or Smaller Reward Margins to Select Preferences for LLM Alignment?

ICML 2025Poster
通讯
5.0
30

$\alpha$-DPO: Adaptive Reward Margin is What Direct Preference Optimization Needs

ICLR 2025Rejected
4.8
28

Leveraging Diffusion Transformers for Stock Factor Augmentation in Financial Markets

ICLR 2025Rejected
三作
4.3
4

Neuron-Level Sequential Editing for Large Language Models

ICLR 2025Withdrawn
通讯
4.3
22

Massively Multi-Agents Reveal That Large Language Models Can Understand Value

ICLR 2025Withdrawn
4.2
15

AlphaDPO: Adaptive Reward Margin for Direct Preference Optimization

ICML 2025Poster