影响力指数

85.93/100

前 0.9%

全站排名 #552

发表论文38 篇

平均评分5.3

年均产出12.7 篇/年

Qifan Wang

Principal Researcher@Meta AI·美国·OpenReview

研究方向

deep learning · graph embedding · GNNs · data mining · representation learning · hashing · natural language processing · information retrieval

6.0

Productive LLM Hallucinations: Conditions, Mechanisms, and Benefits

ICLR 2026Rejected

三作

5.5

TokenSeek: Memory Efficient Fine Tuning via Instance-Aware Token Ditching

ICLR 2026Poster

二作

4.0

Meaningless Tokens, Meaningful Gains: How Activation Shifts Enhance LLM Reasoning

ICLR 2026Rejected

4.0

See, Think, Hallucinate: Interpreting Reasoning and Hallucinations Beyond the First Hop in Vision-Language Models

ICLR 2026Rejected

4.0

IDInit: A Universal and Stable Initialization Method for Neural Network Training

ICLR 2025Poster

6.3

Probabilistic Token Alignment for Large Language Model Fusion

ICLR 2025Rejected

5.8

Modality-Specialized Synergizers for Interleaved Vision-Language Generalists

ICLR 2025Poster

5.8

Re-Imagining Multimodal Instruction Tuning: A Representation View

ICLR 2025Poster

5.7

RoRA-VLM: Robust Retrieval-Augmented Vision Language Models

ICLR 2025Rejected

5.3

Inertial Confinement Fusion Forecasting via Large Language Models

ICLR 2025Rejected

5.0

Hallucinating LLM Could Be Creative

ICLR 2025Rejected

4.8

PDETime: Rethinking Long-term Multivariate Time Series Forecasting from the Perspective of Partial Differential Equations

ICLR 2025Rejected

4.3

Grounded-VideoLLM: Sharpening Fine-grained Temporal Grounding in Video Large Language Models

ICLR 2025Withdrawn

3.5

LLM as GNN: Graph Vocabulary Learning for Graph Foundation Model

合作者 (20)

Qifan Wang

Productive LLM Hallucinations: Conditions, Mechanisms, and Benefits

TokenSeek: Memory Efficient Fine Tuning via Instance-Aware Token Ditching

Meaningless Tokens, Meaningful Gains: How Activation Shifts Enhance LLM Reasoning

See, Think, Hallucinate: Interpreting Reasoning and Hallucinations Beyond the First Hop in Vision-Language Models

ReMix: Reinforcement Routing for Mixtures of LoRAs in LLM Finetuning

DAG-MoE: From Simple Mixture to Structural Aggregation in Mixture-of-Experts

Demystifying Hybrid Thinking: Can LLMs Truly Switch Between Think and No-Think?

Beyond Next-Token Prediction: Diffusion vs. Autoregressive Reasoning in LLMs

Advantage-Guided Distillation for Preference Alignment in Small Language Models

AR-RAG: Autoregressive Retrieval Augmentation for Image Generation

Probabilistic Token Alignment for Large Language Model Fusion

Visual Agents as Fast and Slow Thinkers

All You Need is One: Capsule Prompt Tuning with a Single Vector

IDInit: A Universal and Stable Initialization Method for Neural Network Training

Probabilistic Token Alignment for Large Language Model Fusion

Modality-Specialized Synergizers for Interleaved Vision-Language Generalists

Re-Imagining Multimodal Instruction Tuning: A Representation View

RoRA-VLM: Robust Retrieval-Augmented Vision Language Models

Inertial Confinement Fusion Forecasting via Large Language Models

Hallucinating LLM Could Be Creative

PDETime: Rethinking Long-term Multivariate Time Series Forecasting from the Perspective of Partial Differential Equations

Grounded-VideoLLM: Sharpening Fine-grained Temporal Grounding in Video Large Language Models

LLM as GNN: Graph Vocabulary Learning for Graph Foundation Model