ScienceBoard: Evaluating Multimodal Autonomous Agents in Realistic Scientific Workflows

Large Language Models (LLMs) have extended their impact beyond Natural Language Processing, substantially fostering the development of interdisciplinary research. Recently, various LLM-based agents have been developed to assist scientific discovery progress across multiple aspects and domains. Among these, computer-using agents, capable of interacting with operating systems as humans do, are paving the way to automated scientific problem-solving and addressing routines in researchers' workflows. Recognizing the transformative potential of these agents, we introduce ScienceBoard, which encompasses two complementary contributions: (i) a realistic, multi-domain environment featuring dynamic and visually rich scientific workflows with integrated professional software, where agents can autonomously interact via different interfaces to accelerate complex research tasks and experiments; and (ii) a challenging benchmark of 169 high-quality, rigorously validated real-world tasks curated by humans, spanning scientific-discovery workflows in domains such as biochemistry, astronomy, and geoinformatics. Extensive evaluations of agents with state-of-the-art backbones (e.g., GPT-4o, Claude 3.7, UI-TARS) show that, despite some promising results, they still fall short of reliably assisting scientists in complex workflows, achieving only a 15% overall success rate. In-depth analysis further provides valuable insights for addressing current agent limitations and more effective design principles, paving the way to build more capable agents for scientific discovery. Our code, environment, and benchmark are at https://qiushisun.github.io/ScienceBoard-Home/.

Do Large Language Models Excel in Complex Logical Reasoning with Formal Language?

Large Language Models (LLMs) have been shown to achieve breakthrough performance on complex logical reasoning tasks. Nevertheless, most existing research focuses on employing formal language to guide LLMs to derive reliable reasoning paths, while systematic evaluations of these capabilities are still limited. In this paper, we aim to conduct a comprehensive evaluation of LLMs across various logical reasoning problems utilizing formal languages. From the perspective of three dimensions, i.e., spectrum of LLMs, taxonomy of tasks, and format of trajectories, our key findings are: 1) Thinking models significantly outperform Instruct models, especially when formal language is employed; 2) All LLMs exhibit limitations in inductive reasoning capability, irrespective of whether they use a formal language; 3) Data with PoT format achieves the best generalization performance across other languages. Additionally, we also curate the formal-relative training data to further enhance the small language models, and the experimental results indicate that a simple rejected fine-tuning method can better enable LLMs to generalize across formal languages and achieve the best overall performance. Our codes and reports are available at https://github.com/jiangjin1999/FormalEval.

CL-CaGAN: Capsule differential adversarial continuous learning for cross-domain hyperspectral anomaly detection

Anomaly detection (AD) has attracted remarkable attention in hyperspectral image (HSI) processing fields, and most existing deep learning (DL)-based algorithms indicate dramatic potential for detecting anomaly samples through specific training process under current scenario. However, the limited prior information and the catastrophic forgetting problem indicate crucial challenges for existing DL structure in open scenarios cross-domain detection. In order to improve the detection performance, a novel continual learning-based capsule differential generative adversarial network (CL-CaGAN) is proposed to elevate the cross-scenario learning performance for facilitating the real application of DL-based structure in hyperspectral AD (HAD) task. First, a modified capsule structure with adversarial learning network is constructed to estimate the background distribution for surmounting the deficiency of prior information. To mitigate the catastrophic forgetting phenomenon, clustering-based sample replay strategy and a designed extra self-distillation regularization are integrated for merging the history and future knowledge in continual AD task, while the discriminative learning ability from previous detection scenario to current scenario is retained by the elaborately designed structure with continual learning (CL) strategy. In addition, the differentiable enhancement is enforced to augment the generation performance of the training data. This further stabilizes the training process with better convergence and efficiently consolidates the reconstruction ability of background samples. To verify the effectiveness of our proposed CL-CaGAN, we conduct experiments on several real HSIs, and the results indicate that the proposed CL-CaGAN demonstrates higher detection performance and continuous learning capacity for mitigating the catastrophic forgetting under cross-domain scenarios.

CL-BioGAN: Biologically-Inspired Cross-Domain Continual Learning for Hyperspectral Anomaly Detection

Memory stability and learning flexibility in continual learning (CL) is a core challenge for cross-scene Hyperspectral Anomaly Detection (HAD) task. Biological neural networks can actively forget history knowledge that conflicts with the learning of new experiences by regulating learning-triggered synaptic expansion and synaptic convergence. Inspired by this phenomenon, we propose a novel Biologically-Inspired Continual Learning Generative Adversarial Network (CL-BioGAN) for augmenting continuous distribution fitting ability for cross-domain HAD task, where Continual Learning Bio-inspired Loss (CL-Bio Loss) and self-attention Generative Adversarial Network (BioGAN) are incorporated to realize forgetting history knowledge as well as involving replay strategy in the proposed BioGAN. Specifically, a novel Bio-Inspired Loss composed with an Active Forgetting Loss (AF Loss) and a CL loss is designed to realize parameters releasing and enhancing between new task and history tasks from a Bayesian perspective. Meanwhile, BioGAN loss with L2-Norm enhances self-attention (SA) to further balance the stability and flexibility for better fitting background distribution for open scenario HAD (OHAD) tasks. Experiment results underscore that the proposed CL-BioGAN can achieve more robust and satisfying accuracy for cross-domain HAD with fewer parameters and computation cost. This dual contribution not only elevates CL performance but also offers new insights into neural adaptation mechanisms in OHAD task.

Prejudge-Before-Think: Enhancing Large Language Models at Test-Time by Process Prejudge Reasoning

In this paper, we introduce a new \emph{process prejudge} strategy in LLM reasoning to demonstrate that bootstrapping with process prejudge allows the LLM to adaptively anticipate the errors encountered when advancing the subsequent reasoning steps, similar to people sometimes pausing to think about what mistakes may occur and how to avoid them, rather than relying solely on trial and error. Specifically, we define a prejudge node in the rationale, which represents a reasoning step, with at least one step that follows the prejudge node that has no paths toward the correct answer. To synthesize the prejudge reasoning process, we present an automated reasoning framework with a dynamic tree-searching strategy. This framework requires only one LLM to perform answer judging, response critiquing, prejudge generation, and thought completion. Furthermore, we develop a two-phase training mechanism with supervised fine-tuning (SFT) and reinforcement learning (RL) to further enhance the reasoning capabilities of LLMs. Experimental results from competition-level complex reasoning demonstrate that our method can teach the model to prejudge before thinking and significantly enhance the reasoning ability of LLMs. Code and data is released at https://github.com/wjn1996/Prejudge-Before-Think.

The Role of Visual Modality in Multimodal Mathematical Reasoning: Challenges and Insights

Recent research has increasingly focused on multimodal mathematical reasoning, particularly emphasizing the creation of relevant datasets and benchmarks. Despite this, the role of visual information in reasoning has been underexplored. Our findings show that existing multimodal mathematical models minimally leverage visual information, and model performance remains largely unaffected by changes to or removal of images in the dataset. We attribute this to the dominance of textual information and answer options that inadvertently guide the model to correct answers. To improve evaluation methods, we introduce the HC-M3D dataset, specifically designed to require image reliance for problem-solving and to challenge models with similar, yet distinct, images that change the correct answer. In testing leading models, their failure to detect these subtle visual differences suggests limitations in current visual perception capabilities. Additionally, we observe that the common approach of improving general VQA capabilities by combining various types of image encoders does not contribute to math reasoning performance. This finding also presents a challenge to enhancing visual reliance during math reasoning. Our benchmark and code would be available at \href{https://github.com/Yufang-Liu/visual_modality_role}{https://github.com/Yufang-Liu/visual\_modality\_role}.

LangYa: Revolutionizing Cross-Spatiotemporal Ocean Forecasting

Ocean forecasting is crucial for both scientific research and societal benefits. Currently, the most accurate forecasting systems are global ocean forecasting systems (GOFSs), which represent the ocean state variables (OSVs) as discrete grids and solve partial differential equations (PDEs) governing the transitions of oceanic state variables using numerical methods. However, GOFSs processes are computationally expensive and prone to cumulative errors. Recently, large artificial intelligence (AI)-based models significantly boosted forecasting speed and accuracy. Unfortunately, building a large AI ocean forecasting system that can be considered cross-spatiotemporal and air-sea coupled forecasts remains a significant challenge. Here, we introduce LangYa, a cross-spatiotemporal and air-sea coupled ocean forecasting system. Results demonstrate that the time embedding module in LangYa enables a single model to make forecasts with lead times ranging from 1 to 7 days. The air-sea coupled module effectively simulates air-sea interactions. The ocean self-attention module improves network stability and accelerates convergence during training, and the adaptive thermocline loss function improves the accuracy of thermocline forecasting. Compared to existing numerical and AI-based ocean forecasting systems, LangYa uses 27 years of global ocean data from the Global Ocean Reanalysis and Simulation version 12 (GLORYS12) for training and achieves more reliable deterministic forecasting results for OSVs. LangYa forecasting system provides global ocean researchers with access to a powerful software tool for accurate ocean forecasting and opens a new paradigm for ocean science.

HEIE: MLLM-Based Hierarchical Explainable AIGC Image Implausibility Evaluator

AIGC images are prevalent across various fields, yet they frequently suffer from quality issues like artifacts and unnatural textures. Specialized models aim to predict defect region heatmaps but face two primary challenges: (1) lack of explainability, failing to provide reasons and analyses for subtle defects, and (2) inability to leverage common sense and logical reasoning, leading to poor generalization. Multimodal large language models (MLLMs) promise better comprehension and reasoning but face their own challenges: (1) difficulty in fine-grained defect localization due to the limitations in capturing tiny details; and (2) constraints in providing pixel-wise outputs necessary for precise heatmap generation. To address these challenges, we propose HEIE: a novel MLLM-Based Hierarchical Explainable image Implausibility Evaluator. We introduce the CoT-Driven Explainable Trinity Evaluator, which integrates heatmaps, scores, and explanation outputs, using CoT to decompose complex tasks into subtasks of increasing difficulty and enhance interpretability. Our Adaptive Hierarchical Implausibility Mapper synergizes low-level image features with high-level mapper tokens from LLMs, enabling precise local-to-global hierarchical heatmap predictions through an uncertainty-based adaptive token approach. Moreover, we propose a new dataset: Expl-AIGI-Eval, designed to facilitate interpretable implausibility evaluation of AIGC images. Our method demonstrates state-of-the-art performance through extensive experiments.

OCEAN: Offline Chain-of-thought Evaluation and Alignment in Large Language Models

Offline evaluation of LLMs is crucial in understanding their capacities, though current methods remain underexplored in existing research. In this work, we focus on the offline evaluation of the chain-of-thought capabilities and show how to optimize LLMs based on the proposed evaluation method. To enable offline feedback with rich knowledge and reasoning paths, we use knowledge graphs (e.g., Wikidata5m) to provide feedback on the generated chain of thoughts. Due to the heterogeneity between LLM reasoning and KG structures, direct interaction and feedback from KGs on LLM behavior are challenging, as they require accurate entity linking and grounding of LLM-generated chains of thought in the KG. To address the above challenge, we propose an offline chain-of-thought evaluation framework, OCEAN, which models chain-of-thought reasoning in LLMs as an MDP and evaluate the policy's alignment with KG preference modeling. To overcome the reasoning heterogeneity and grounding problems, we leverage on-policy KG exploration and RL to model a KG policy that generates token-level likelihood distributions for LLM-generated chain-of-thought reasoning paths, simulating KG reasoning preference. Then we incorporate the knowledge-graph feedback on the validity and alignment of the generated reasoning paths into inverse propensity scores and propose KG-IPS estimator. Theoretically, we prove the unbiasedness of the proposed KG-IPS estimator and provide a lower bound on its variance. With the off-policy evaluated value function, we can directly enable off-policy optimization to further enhance chain-of-thought alignment. Our empirical study shows that OCEAN can be efficiently optimized for generating chain-of-thought reasoning paths with higher estimated values without affecting LLMs' general abilities in downstream tasks or their internal knowledge.

Augmenting the Veracity and Explanations of Complex Fact Checking via Iterative Self-Revision with LLMs

Explanation generation plays a more pivotal role than fact verification in producing interpretable results and facilitating comprehensive fact-checking, which has recently garnered considerable attention. However, previous studies on explanation generation has shown several limitations, such as being confined to English scenarios, involving overly complex inference processes, and not fully unleashing the potential of the mutual feedback between veracity labels and explanation texts. To address these issues, we construct two complex fact-checking datasets in the Chinese scenarios: CHEF-EG and TrendFact. These datasets involve complex facts in areas such as health, politics, and society, presenting significant challenges for fact verification methods. In response to these challenges, we propose a unified framework called FactISR (Augmenting Fact-Checking via Iterative Self-Revision) to perform mutual feedback between veracity and explanations by leveraging the capabilities of large language models(LLMs). FactISR uses a single model to address tasks such as fact verification and explanation generation. Its self-revision mechanism can further revision the consistency between veracity labels, explanation texts, and evidence, as well as eliminate irrelevant noise. We conducted extensive experiments with baselines and FactISR on the proposed datasets. The experimental results demonstrate the effectiveness of our method.