影响力指数

74.78/100

前 1.9%

全站排名 #1,229

发表论文15 篇

平均评分5.7

年均产出5.0 篇/年

Pan Lu

Postdoc@Stanford University·美国·OpenReview

研究方向

Theorem Proving · LLM Agents · AI for Science · Tool Usage · Agentic AI · Large Language Models · Scientific Reasoning · Mathematical Reasoning · Vision-Language Models

7.3

Fractional Reasoning via Latent Steering Vectors Improves Inference Time Compute

ICLR 2026Rejected

三作

3.5

Where LLM Agents Fail And How They can Learn From Failures

ICLR 2026Withdrawn

6.5

MMSearch: Unveiling the Potential of Large Models as Multi-modal Search Engines

ICLR 2025Poster

6.3

Weak-for-Strong: Training Weak Meta-Agent to Harness Strong Executors

COLM 2025Poster

5.8

ChemAgent: Self-updating Memories in Large Language Models Improves Chemical Reasoning

ICLR 2025Poster

5.6

MRAG-Bench: Vision-Centric Evaluation for Retrieval-Augmented Multimodal Models

ICLR 2025Poster

5.2

MuirBench: A Comprehensive Benchmark for Robust Multi-image Understanding

合作者 (20)

Pan Lu

In-the-Flow Agentic System Optimization for Effective Planning and Tool Use

TaTToo: Tool-Grounded Thinking PRM for Test-Time Scaling in Tabular Reasoning

Learning Human-Perceived Fakeness in AI-Generated Videos via Multimodal LLMs

Fractional Reasoning via Latent Steering Vectors Improves Inference Time Compute

Where LLM Agents Fail And How They can Learn From Failures

MMSearch: Unveiling the Potential of Large Models as Multi-modal Search Engines

Weak-for-Strong: Training Weak Meta-Agent to Harness Strong Executors

ChemAgent: Self-updating Memories in Large Language Models Improves Chemical Reasoning

MRAG-Bench: Vision-Centric Evaluation for Retrieval-Augmented Multimodal Models

MuirBench: A Comprehensive Benchmark for Robust Multi-image Understanding