PaperHub
影响力指数
92.12/100
前 0.4%
全站排名 #276
发表论文44 篇
平均评分4.9
年均产出14.7 篇/年

Linfeng Zhang

Assistant Professor@Shanghai Jiaotong University·中国·OpenReview
研究方向

Knowledge Distillation · Efficient Inference Methods · Diffusion Models · Dataset Distillation · Token Pruning · Efficient Multi-Modal Models

7.0
32

Let Features Decide Their Own Solvers: Hybrid Feature Caching for Diffusion Transformers

ICLR 2026Oral
通讯
6.0
20

Generation then Reconstruction: Accelerating Masked Autoregressive Models via Two-Stage Sampling

ICLR 2026Poster
通讯
6.0
15

HiCache: A Plug-in Scaled-Hermite Upgrade for Taylor-Style Cache-then-Forecast Diffusion Acceleration

ICLR 2026Poster
通讯
6.0
9

FastVMT: Eliminating Redundancy in Video Motion Transfer

ICLR 2026Poster
通讯
5.5
22

Mixing Importance with Diversity: Joint Optimization for KV Cache Compression in Large Vision-Language Models

ICLR 2026Poster
通讯
5.5
5

EVCtrl: Efficient Control Adapter for Visual Generation

ICLR 2026Withdrawn
通讯
5.5
14

Follow-Your-Shape: Shape-Aware Image Editing via Trajectory-Guided Region Control

ICLR 2026Poster
5.5
20

Accelerating Diffusion Large Language Models with SlowFast Sampling: The Three Golden Principles

ICLR 2026Poster
通讯
5.5
25

Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agents

ICLR 2026Poster
5.5
32

FreqCa: Accelerating Diffusion Models via Frequency Decomposition Caching

ICLR 2026Desk Rejected
通讯
5.3
19

dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Caching

ICLR 2026Rejected
通讯
5.0
35

Grounding and Enhancing Informativeness and Utility in Dataset Distillation

ICLR 2026Poster
通讯
5.0
34

Towards Self-Evolving Agent Benchmarks : Validatable Agent Trajectory via Test-Time Exploration

ICLR 2026Poster
5.0
23

The Devil behind the mask: An emergent safety vulnerability of Diffusion LLMs

ICLR 2026Poster
通讯
5.0
22

DD-Ranking: Rethinking the Evaluation of Dataset Distillation

ICLR 2026Rejected
4.8
27

Socratic-Zero : Bootstrapping Reasoning via Data-Free Agent Co-evolution

ICLR 2026Rejected
通讯
4.5
15

MASC: Boosting Autoregressive Image Generation with a Manifold-Aligned Semantic Clustering

ICLR 2026Desk Rejected
三作
4.5
18

SkipVAR: Accelerating Visual Autoregressive Modeling via Adaptive Frequency-Aware Skipping

ICLR 2026Rejected
通讯
4.5
22

Zero in on Faithful Anchors: High-Fidelity Visual Token Condensation for Multimodal Large Language Models

ICLR 2026Rejected
4.5
16

Z-Cache: Accelerating Diffusion Transformers via Self-Reflection

ICLR 2026Rejected
通讯
4.5
32

Winning the Pruning Gamble: A Unified Approach to Joint Sample and Token Pruning for Efficient Supervised Fine-Tuning

ICLR 2026Desk Rejected
通讯
4.5
37

Rethinking LLM Evaluation: Can We Evaluate LLMs with 200× Less Data?

ICLR 2026Poster
通讯
4.0
18

Accelerate Diffusion Transformers with Feature Momentum

ICLR 2026Rejected
通讯
4.0
11

dVLA: Diffusion Vision-Language-Action Model with Multimodal Chain-of-Thought

ICLR 2026Rejected
4.0
44

Towards Principled Dataset Distillation: A Spectral Distribution Perspective

ICLR 2026Rejected
通讯
4.0
21

AudioMarathon: A Comprehensive Benchmark for Long-Context Audio Understanding and Efficiency in Audio LLMs

ICLR 2026Withdrawn
通讯
3.5
5

Self Speculative Decoding for Diffusion Large Language Model

ICLR 2026Withdrawn
通讯
3.5
6

Understanding-in-Generation: Reinforcing Generative Capability of Unified Model via Infusing Understanding into Generation

ICLR 2026Rejected
3.5
5

Learning Robust Anymodal Segmentor with Unimodal and Cross-modal Distillation

ICLR 2026Withdrawn
3.3
4

Spot the Critical Words: Text-Guided Visual Token Pruning for Efficient Large Vision-Language Model Inference

ICLR 2026Withdrawn
3.3
4

ViCO: A Training Strategy towards Semantic Aware Dynamic High-Resolution

ICLR 2026Withdrawn
3.0
5

Flow Diverse and Efficient: Learning Momentum Flow Matching via Stochastic Velocity Field Sampling

ICLR 2026Withdrawn
3.0
15

Detecting Variant Contamination in LLMs via Variance of Generation Distribution

ICLR 2026Withdrawn
2.0
5

Self-Reflective Reinforcement Learning for Diffusion-based Image Reasoning Generation

ICLR 2026Withdrawn
8.0
23

REEF: Representation Encoding Fingerprints for Large Language Models

ICLR 2025Oral
7.3
27

Don't Just Chase “Highlighted Tokens” in MLLMs: Revisiting Visual Holistic Context Retention

NeurIPS 2025Poster
7.0
14

EfficientVLA: Training-Free Acceleration and Compression for Vision-Language-Action Models

NeurIPS 2025Poster
通讯
6.0
35

Accelerating Diffusion Transformers with Token-wise Feature Caching

ICLR 2025Poster
通讯
6.0
7

Gnothi Seauton: Empowering Faithful Self-Interpretability in Black-Box Transformers

ICLR 2025Poster
通讯
6.0
21

Efficient Multi-modal Large Language Models via Progressive Consistency Distillation

NeurIPS 2025Poster
通讯
5.5
11

RealRAG: Retrieval-augmented Realistic Image Generation via Self-reflective Contrastive Learning

ICML 2025Poster
4.2
7

Decouple-Then-Merge: Towards Better Training for Diffusion Models

ICLR 2025Withdrawn
通讯
4.2
6

DRUPI: Dataset Reduction Using Privileged Information

ICLR 2025Withdrawn
通讯
合作者 (20)