Yijin Guo

WenZhe Li, Qihang Yan, Liang Chen et al.

Scientific images function as critical evidence in research communication, yet their integrity faces unprecedented threats from AI-generated content that introduces subtle but consequential errors. Existing evaluation paradigms prove inadequate: perceptual quality metrics poorly correlate with scientific validity, while language models lack domain-specific verification capabilities. To address this gap, we propose the \textbf{S}cientific \textbf{I}mage \textbf{U}tility and \textbf{U}pgradability \textbf{A}ssessment (\textbf{SIU$^2$A}) framework, which introduces two complementary dimensions for scientific image evaluation. \textbf{Utility} encompasses \textit{error detection} (identifying scientific inaccuracies) and \textit{correction feasibility} (assessing whether errors can be reliably repaired). \textbf{Upgradability} measures the quality of correction. We categorize scientific image corruption into four fundamental types: Detail Distortion, Incompleteness, False Content, and Entity Confusion. Based on this taxonomy, we construct SIU$^2$A-Benchmark, a dataset with expert annotations for error identification and repair. The framework implements a two-stage evaluation protocol: the \textit{Utility} stage evaluates error detection capability and repair instruction generation, while the \textit{Upgradability} stage assesses whether corrections faithfully restore scientific validity without compromising existing accurate information. Experiments reveal that current multimodal systems exhibit significant limitations in both scientific error assessment and faithful correction, exposing a fundamental gap between visual perception and scientific usability.

Ye Shen, Dun Pei, Yiqiu Guo et al.

Despite recent advances in understanding and leveraging long-range conversational memory, existing benchmarks still lack systematic evaluation of large language models(LLMs) across diverse memory dimensions, particularly in multi-session settings. In this work, we propose EvolMem, a new benchmark for assessing multi-session memory capabilities of LLMs and agent systems. EvolMem is grounded in cognitive psychology and encompasses both declarative and non-declarative memory, further decomposed into multiple fine-grained abilities. To construct the benchmark, we introduce a hybrid data synthesis framework that consists of topic-initiated generation and narrative-inspired transformations. This framework enables scalable generation of multi-session conversations with controllable complexity, accompanied by sample-specific evaluation guidelines. Extensive evaluation reveals that no LLM consistently outperforms others across all memory dimensions. Moreover, agent memory mechanisms do not necessarily enhance LLMs' capabilities and often exhibit notable efficiency limitations. Data and code will be released at https://github.com/shenye7436/EvolMem.

Xiaoxiao Wang, Chunxiao Li, Junying Wang et al.

As comprehensive large model evaluation becomes prohibitively expensive, predicting model performance from limited observations has become essential. However, existing statistical methods struggle with pattern shifts, data sparsity, and lack of explanation, while pure LLM methods remain unreliable. We propose STAR, a framework that bridges data-driven STatistical expectations with knowledge-driven Agentic Reasoning. STAR leverages specialized retrievers to gather external knowledge and embeds semantic features into Constrained Probabilistic Matrix Factorization (CPMF) to generate statistical expectations with uncertainty. A reasoning module guided by Expectation Violation Theory (EVT) then refines predictions through intra-family analysis, cross-model comparison, and credibility-aware aggregation, producing adjustments with traceable explanations. Extensive experiments show that STAR consistently outperforms all baselines on both score-based and rank-based metrics, delivering a 14.46% gain in total score over the strongest statistical method under extreme sparsity, with only 1--2 observed scores per test model.

Yijin Guo, Kaiyuan Ji, Xiaorong Zhu et al.

Currently, nearly all evaluations of foundation models focus on objective metrics, emphasizing quiz performance to define model capabilities. While this model-centric approach enables rapid performance assessment, it fails to reflect authentic human experiences. To address this gap, we propose a Human-Centric subjective Evaluation (HCE) framework, focusing on three core dimensions: problem-solving ability, information quality, and interaction experience. Through experiments involving Deepseek R1, OpenAI o3 mini, Grok 3, and Gemini 2.5, we conduct over 540 participant-driven evaluations, where humans and models collaborate on open-ended research tasks, yielding a comprehensive subjective dataset. This dataset captures diverse user feedback across multiple disciplines, revealing distinct model strengths and adaptability. Our findings highlight Grok 3's superior performance, followed by Deepseek R1 and Gemini 2.5, with OpenAI o3 mini lagging behind. By offering a novel framework and a rich dataset, this study not only enhances subjective evaluation methodologies but also lays the foundation for standardized, automated assessments, advancing LLM development for research and practical scenarios. Our dataset link is https://github.com/yijinguo/Human-Centric-Evaluation.

Junying Wang, Zicheng Zhang, Yijin Guo et al.

As foundation models grow rapidly in capability and deployment, evaluating their scientific understanding becomes increasingly critical. Existing science benchmarks have made progress towards broad Range, wide Reach, and high Rigor, yet they often face two major challenges: data leakage risks that compromise benchmarking validity, and evaluation inefficiency due to large-scale testing. To address these issues, we introduce the Ever-Evolving Science Exam (EESE), a dynamic benchmark designed to reliably assess scientific capabilities in foundation models. Our approach consists of two components: 1) a non-public EESE-Pool with over 100K expertly constructed science instances (question-answer pairs) across 5 disciplines and 500+ subfields, built through a multi-stage pipeline ensuring Range, Reach, and Rigor, 2) a periodically updated 500-instance subset EESE, sampled and validated to enable leakage-resilient, low-overhead evaluations. Experiments on 32 open- and closed-source models demonstrate that EESE effectively differentiates the strengths and weaknesses of models in scientific fields and cognitive dimensions. Overall, EESE provides a robust, scalable, and forward-compatible solution for science benchmark design, offering a realistic measure of how well foundation models handle science questions. The project page is at: https://github.com/aiben-ch/EESE.

Junying Wang, Wenzhe Li, Yalun Wu et al.

Affordance theory suggests that environments inherently provide action possibilities shaping perception and behavior. While Multimodal Large Language Models (MLLMs) achieve strong performance in vision-language tasks, their ability to perceive affordance, which is crucial for intuitive and safe interactions, remains underexplored. To address this, we introduce **A4Bench**, a novel benchmark designed to evaluate the affordance perception abilities of MLLMs across two dimensions: 1) Constitutive Affordance, assessing understanding of inherent object properties through 1,282 questionanswer pairs spanning nine sub-disciplines, and 2) Transformative Affordance, probing dynamic and contextual nuances (e.g., misleading, time-dependent, cultural, or individual-specific affordance) with 718 challenging question-answer pairs. We evaluate 17 MLLMs (nine proprietary and eight open-source) and compare them to human performance. Results show that proprietary models generally outperform open-source ones, yet all models perform far below humans, especially in transformative affordance. Furthermore, even top-performing models, such as Gemini-2.0-Pro (18.05% overall exact match accuracy), significantly lag behind human performance (best: 85.34%, worst: 81.25%). These findings highlight critical gaps in environmental understanding of MLLMs and provide a foundation for advancing AI systems toward more robust, context-aware interactions.

Kaiyuan Ji, Yijin Guo, Zicheng Zhang et al.

With the increasing use of large language models (LLMs) in medical decision-support, it is essential to evaluate not only their final answers but also the reliability of their reasoning. Two key risks are Chain-of-Thought (CoT) faithfulness -- whether reasoning aligns with responses and medical facts -- and sycophancy, where models follow misleading cues over correctness. Existing benchmarks often collapse such vulnerabilities into single accuracy scores. To address this, we introduce MedOmni-45 Degrees, a benchmark and workflow designed to quantify safety-performance trade-offs under manipulative hint conditions. It contains 1,804 reasoning-focused medical questions across six specialties and three task types, including 500 from MedMCQA. Each question is paired with seven manipulative hint types and a no-hint baseline, producing about 27K inputs. We evaluate seven LLMs spanning open- vs. closed-source, general-purpose vs. medical, and base vs. reasoning-enhanced models, totaling over 189K inferences. Three metrics -- Accuracy, CoT-Faithfulness, and Anti-Sycophancy -- are combined into a composite score visualized with a 45 Degrees plot. Results show a consistent safety-performance trade-off, with no model surpassing the diagonal. The open-source QwQ-32B performs closest (43.81 Degrees), balancing safety and accuracy but not leading in both. MedOmni-45 Degrees thus provides a focused benchmark for exposing reasoning vulnerabilities in medical LLMs and guiding safer model development.

Farong Wen, Yijin Guo, Junying Wang et al.

The rapid development of Multimodal Large Language Models (MLLM) has led to a wide range of MLLM applications, and a number of benchmark datasets have sprung up in order to assess MLLM abilities. However, full-coverage Q&A testing on large-scale data is resource-intensive and time-consuming. To address this issue, we propose the MLLM Interview (MITV) strategy, which aims to quickly obtain MLLM performance metrics by quizzing fewer question. First, First, we constructed the interview dataset, which was built on an existing MLLM assessment dataset, by adding difficulty labels based on the performance of some typical MLLMs in this dataset. Second, we propose an MLLM Interview strategy, which obtains an initial performance situation of the large model by quizzing a small number of topics and then continuously tries to test the model's limits. Through extensive experiments, the result shows that the MITV strategy proposed in this paper performs well on MLLM benchmark datasets, and it is able to obtain the model evaluation capability faster through a small number of questions and answers.

Qi Jia, Xiujie Song, Zicheng Zhang et al.

Existing benchmarks for large language models (LLMs) predominantely focus on assessing their capabilities through verifiable tasks. Such objective and static benchmarks offer limited utility for practical LLM selection, making it difficult for users to find suitable models for their individual needs. To bridge this gap, we present the first User-Centric Subjective Leaderboard (USL), which provides a preference-driven, dynamic ranking of LLMs across diverse real-world scenarios. Our work is built upon a thorough investigation of real human preference data, involving more than 10K subjective queries. Our investigation reveals significant diversity and contradictions in human preferences, which limit the effectiveness of state-of-the-art reward models. To address this, we introduce Customizable Reward Models (CRMs). With only 4B parameters, our CRM surpasses the performance of leading models such as GPT-4.1 and Gemini-2.5-pro, showing exceptional generalization capabilities across new topics and criteria. The USL, powered by CRMs, exhibits strong negative correlations to contradictory preferences.

Junying Wang, Zicheng Zhang, Ye Shen et al.

High-quality, multi-modal benchmarks are crucial for advancing scientific reasoning in large models yet their manual creation is costly and unscalable. To address this bottleneck, we explore the potential for transforming Text-Only QA Pairs (TQAs) into high-quality Multi-Modal QA Pairs (MMQAs), which include three parts: 1) Task Definition \& Evaluation Rubric: We develop a TQA-to-MMQA framework and establish a comprehensive, multi-dimensional MMQA quality rubric that provides principles for the transformation. 2) Benchmark Construction: Then we construct two extensive benchmarks to rigorously evaluate state-of-the-art generation \& understanding models on the distinct tasks of MMQA generation \& MMQA quality evaluation. 3) Preliminary Solution: We develop an agentic system (Q-Mirror), which operationalizes our framework by integrating MMQA generation and evaluation into a closed loop for iterative refinement. Our experiments show that while state-of-the-art models can generate MMQAs, their outputs still leave substantial gaps, underscoring the need for reliable evaluation. We further demonstrate that top-tier understanding models align closely with human judgment in MMQA quality assessment. Leveraging both insights, the Q-Mirror agent raises average scores from 78.90 to 85.22 and pass rates from 72\% to 95\%, offering a practical path to large-scale scientific benchmarks.

Yijin Guo, Zicheng Zhang, Ye Shen et al.

Existing evaluations of foundation models, including recent human-centric approaches, fail to capture what truly matters: user's experience during interaction. Current methods treat evaluation as a matter of output correctness alone, overlooking that user satisfaction emerges from the interplay between response quality and interaction, which limits their ability to account for the mechanisms underlying user experience. To address this gap, we introduce QoNext, the first framework that adapts Quality of Experience (QoE) principles from networking and multimedia to the assessment of foundation models. QoNext identifies experiential factors that shape user experience and incorporates them into controlled experiments, where human ratings are collected under varied configurations. From these studies we construct a QoE-oriented database and train predictive models that estimate perceived user experience from measurable system parameters. Our results demonstrate that QoNext not only enables proactive and fine-grained evaluation but also provides actionable guidance for productized services of optimizing foundation models in practice.

Ye Shen, Junying Wang, Farong Wen et al.

The rapid progress of Multi-Modal Large Language Models (MLLMs) has spurred the creation of numerous benchmarks. However, conventional full-coverage Question-Answering evaluations suffer from high redundancy and low efficiency. Inspired by human interview processes, we propose a multi-to-one interview paradigm for efficient MLLM evaluation. Our framework consists of (i) a two-stage interview strategy with pre-interview and formal interview phases, (ii) dynamic adjustment of interviewer weights to ensure fairness, and (iii) an adaptive mechanism for question difficulty-level chosen. Experiments on different benchmarks show that the proposed paradigm achieves significantly higher correlation with full-coverage results than random sampling, with improvements of up to 17.6% in PLCC and 16.7% in SRCC, while reducing the number of required questions. These findings demonstrate that the proposed paradigm provides a reliable and efficient alternative for large-scale MLLM benchmarking.