Author name cluster

Yibo Li

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

2 author rows

NeurIPS Conference 2025 Conference Paper

ConfTuner: Training Large Language Models to Express Their Confidence Verbally

Yibo Li
Miao Xiong
Jiaying Wu
Bryan Hooi

Large Language Models (LLMs) are increasingly deployed in high-stakes domains such as science, law, and healthcare, where accurate expressions of uncertainty are essential for reliability and trust. However, current LLMs are often observed to generate incorrect answers with high confidence—a phenomenon known as "overconfidence". Recent efforts have focused on calibrating LLMs' verbalized confidence: i. e. , their expressions of confidence in text form, such as "I am 80% confident that. .. ". Existing approaches either rely on prompt engineering or fine-tuning with heuristically generated uncertainty estimates, both of which have limited effectiveness and generalizability. Motivated by the notion of proper scoring rules for calibration in classical machine learning models, we introduce ConfTuner, a simple and efficient fine-tuning method that introduces minimal overhead and does not require ground-truth confidence scores or proxy confidence estimates. ConfTuner relies on a new loss function, tokenized Brier score, which we theoretically prove to be a proper scoring rule, intuitively meaning that it "correctly incentivizes the model to report its true probability of being correct". ConfTuner improves calibration across diverse reasoning tasks and generalizes to black-box models such as GPT-4o. Our results further show that better-calibrated confidence enables downstream gains in self-correction and model cascade, advancing the development of trustworthy LLM systems. The code is available at https: //github. com/liushiliushi/ConfTuner.

PDF Details

NeurIPS Conference 2025 Conference Paper

MLR-Bench: Evaluating AI Agents on Open-Ended Machine Learning Research

Hui Chen
Miao Xiong
Yujie Lu
Wei Han
Ailin Deng
Yufei He
Jiaying Wu
Yibo Li

Recent advancements in AI agents have demonstrated their growing potential to drive and support scientific discovery. In this work, we introduce MLR-Bench, a comprehensive benchmark for evaluating AI agents on open-ended machine learning research. MLR-Bench includes three key components: (1) 201 research tasks sourced from NeurIPS, ICLR, and ICML workshops covering diverse ML topics; (2) MLR-Judge, an automated evaluation framework combining LLM-based reviewers with carefully designed review rubrics to assess research quality; and (3) MLR-Agent, a modular agent scaffold capable of completing research tasks through four stages: idea generation, proposal formulation, experimentation, and paper writing. Our framework supports both stepwise assessment across these distinct research stages, and end-to-end evaluation of the final research paper. We then use MLR-Bench to evaluate six frontier LLMs and an advanced coding agent, finding that while LLMs are effective at generating coherent ideas and well-structured papers, current coding agents frequently (e. g. , in 80\% of the cases) produce fabricated or invalidated experimental results—posing a major barrier to scientific reliability. We validate MLR-Judge through human evaluation, showing high agreement with expert reviewers, supporting its potential as a scalable tool for research evaluation. We open-source MLR-Bench to help the community benchmark, diagnose, and improve AI research agents toward trustworthy and transparent scientific discovery.

PDF Details

AAAI Conference 2024 Conference Paper

A Generalized Neural Diffusion Framework on Graphs

Yibo Li
Xiao Wang
Hongrui Liu
Chuan Shi

Recent studies reveal the connection between GNNs and the diffusion process, which motivates many diffusion based GNNs to be proposed. However, since these two mechanisms are closely related, one fundamental question naturally arises: Is there a general diffusion framework that can formally unify these GNNs? The answer to this question can not only deepen our understanding of the learning process of GNNs, but also may open a new door to design a broad new class of GNNs. In this paper, we propose a general diffusion equation framework with the fidelity term, which formally establishes the relationship between the diffusion process with more GNNs. Meanwhile, with this framework, we identify one characteristic of graph diffusion networks, i.e., the current neural diffusion process only corresponds to the first-order diffusion equation. However, by an experimental investigation, we show that the labels of high-order neighbors actually appear monophily property, which induces the similarity based on labels among high-order neighbors without requiring the similarity among first-order neighbors. This discovery motives to design a new high-order neighbor-aware diffusion equation, and derive a new type of graph diffusion network (HiD-Net) based on the framework. With the high-order diffusion equation, HiD-Net is more robust against attacks and works on both homophily and heterophily graphs. We not only theoretically analyze the relation between HiD-Net with high-order random walk, but also provide a theoretical convergence guarantee. Extensive experimental results well demonstrate the effectiveness of HiD-Net over state-of-the-art graph diffusion networks.

PDF Details DOI

ICML Conference 2024 Conference Paper

Less is More: on the Over-Globalizing Problem in Graph Transformers

Yujie Xing
Xiao Wang 0017
Yibo Li
Hai Huang
Chuan Shi 0001

Graph Transformer, due to its global attention mechanism, has emerged as a new tool in dealing with graph-structured data. It is well recognized that the global attention mechanism considers a wider receptive field in a fully connected graph, leading many to believe that useful information can be extracted from all the nodes. In this paper, we challenge this belief: does the globalizing property always benefit Graph Transformers? We reveal the over-globalizing problem in Graph Transformer by presenting both empirical evidence and theoretical analysis, i. e. , the current attention mechanism overly focuses on those distant nodes, while the near nodes, which actually contain most of the useful information, are relatively weakened. Then we propose a novel Bi-Level Global Graph Transformer with Collaborative Training (CoBFormer), including the inter-cluster and intra-cluster Transformers, to prevent the over-globalizing problem while keeping the ability to extract valuable information from distant nodes. Moreover, the collaborative training is proposed to improve the model’s generalization ability with a theoretical guarantee. Extensive experiments on various graphs well validate the effectiveness of our proposed CoBFormer.

Details

EAAI Journal 2023 Journal Article

Online surface defect segmentation on aluminum strip production line using a lightweight and efficient model

Zehua Lv
Yibo Li
Siying Qian
Liuqing Wu

A significant challenge in automated defect inspection (ADI) of aluminum strip surfaces is improving segmentation speed to satisfy the online inspection requirements of the production line while maintaining the precision of defect identification. This study proposes a lightweight and efficient defect segmentation model that can be applied in aluminum processing enterprises for fast and precise segmentation of aluminum strip surface defects. A novel fusion attention (FA) mechanism is first established to enhance the focus on critical characteristics along the spatial and channel dimensions. This mechanism adopts continuous dilated convolutions with appropriate dilation rates to effectively increase the range of the receptive field and improve defect localization accuracy. Subsequently, a lightweight MobileViTv2 with an embedded FA mechanism is employed as a multi-scale feature extractor to learn comprehensive representations from defect images. Next, a novel feature fusion method, named large-scale feature pyramid network (LSFPN), is introduced to enhance the focus on details within large-scale features. LSFPN establishes four progressively shallower top-down pathways with fast normalized fusion weights and incorporates lightweight aggregation nodes based on the MoblieNetv2 block. Surface images of straightened aluminum strips with five universal defects were collected, whereby a new dataset was established. The experimental outcomes demonstrate the proposed model outperforms other state-of-the-art techniques synthetically, achieving a mean Intersection over Union (mIoU) of 87. 01%, a segmentation speed of 61. 67 fps, and a model size of 16. 23 MB. This model may serve as a valuable theoretical foundation for the online segmentation of aluminum strip surface defects in embedded devices.

Details DOI