Automating web navigation which aims to build a web agent that follows user instructions to complete tasks like booking flights by interacting with websites, has received increasing attention due to its practical value.Although existing web agents are mostly equipped with visual perception, planning, and memory abilities, their reasoning process are still deviate from human cognition.In this work, we study the human thought pattern to empower agent with more human-like abilities in web navigation.To tackle this problem, we propose a novel multimodal web agent framework called WebExperT, which is designed to emulate the human planning process of "thinking fast and slow" to effectively decompose complex user instructions.Furthermore, WebExperT leverages experiential learning by reflecting from failure for continuously refining planning and decision-making outcomes.Experimental results on the MIND2WEB benchmark demonstrate the superiority of WebExperT in both supervised and unsupervised settings.

How to alleviate the hallucinations of Large Language Models (LLMs) has always been the fundamental goal pursued by the LLMs research community. Looking through numerous hallucination-related studies, a mainstream category of methods is to reduce hallucinations by optimizing the knowledge representation of LLMs to change their output. Considering that the core focus of these works is the knowledge acquired by models, and knowledge has long been a central theme in human societal progress, we believe that the process of models refining knowledge can greatly benefit from the way humans learn. In our work, by imitating the human learning process, we design an Adaptive Contrastive Learning strategy. Our method flexibly constructs different positive and negative samples for contrastive learning based on LLMs' actual mastery of knowledge. This strategy helps LLMs consolidate the correct knowledge they already possess, deepen their understanding of the correct knowledge they have encountered but not fully grasped, forget the incorrect knowledge they previously learned, and honestly acknowledge the knowledge they lack. Extensive experiments and detailed analyses on widely used datasets demonstrate the effectiveness and competitiveness of our method.

Leveraging mathematical Large Language Models (LLMs) for proof generation is a fundamental topic in LLMs research. We argue that the ability of current LLMs to prove statements largely depends on whether they have encountered the relevant proof process during training. This reliance limits their deeper understanding of mathematical theorems and related concepts. Inspired by the pedagogical method of "proof by counterexamples" commonly used in human mathematics education, our work aims to enhance LLMs’ ability to conduct mathematical reasoning and proof through counterexamples. Specifically, we manually create a high-quality, university-level mathematical benchmark, COUNTERMATH, which requires LLMs to prove mathematical statements by providing counterexamples, thereby assessing their grasp of mathematical concepts. Additionally, we develop a data engineering framework to automatically obtain training data for further model improvement. Extensive experiments and detailed analyses demonstrate that COUNTERMATH is challenging, indicating that LLMs, such as OpenAI o1, have insufficient counterexample-driven proof capabilities. Moreover, our exploration into model training reveals that strengthening LLMs’ counterexample-driven conceptual reasoning abilities is crucial for improving their overall mathematical capabilities. We believe that our work offers new perspectives on the community of mathematical LLMs.

Large Language Models (LLMs) have demonstrated outstanding performance in mathematical reasoning capabilities. However, we argue that current large-scale reasoning models primarily rely on scaling up training datasets with diverse mathematical problems and long thinking chains, which raises questions about whether LLMs genuinely acquire mathematical concepts and reasoning principles or merely remember the training data. In contrast, humans tend to break down complex problems into multiple fundamental atomic capabilities. Inspired by this, we propose a new paradigm for evaluating mathematical atomic capabilities. Our work categorizes atomic abilities into two dimensions: (1) field-specific abilities across four major mathematical fields, algebra, geometry, analysis, and topology, and (2) logical abilities at different levels, including conceptual understanding, forward multi-step reasoning with formal math language, and counterexample-driven backward reasoning. We propose corresponding training and evaluation datasets for each atomic capability unit, and conduct extensive experiments about how different atomic capabilities influence others, to explore the strategies to elicit the required specific atomic capability. Evaluation and experimental results on advanced models show many interesting discoveries and inspirations about the different performances of models on various atomic capabilities and the interactions between atomic capabilities. Our findings highlight the importance of decoupling mathematical intelligence into atomic components, providing new insights into model cognition and guiding the development of training strategies toward a more efficient, transferable, and cognitively grounded paradigm of "atomic thinking".

Emotional Support Conversation (ESC) systems are pivotal in providing empathetic interactions, aiding users through negative emotional states by understanding and addressing their unique experiences. In this paper, we tackle two key challenges in ESC: enhancing contextually relevant and empathetic response generation through dynamic demonstration retrieval, and advancing cognitive understanding to grasp implicit mental states comprehensively. We introduce Dynamic Demonstration Retrieval and Cognitive-Aspect Situation Understanding (D2RCU), a novel approach that synergizes these elements to improve the quality of support provided in ESCs. By leveraging in-context learning and persona information, we introduce an innovative retrieval mechanism that selects informative and personalized demonstration pairs. We also propose a cognitive understanding module that utilizes four cognitive relationships from the ATOMIC knowledge source to deepen situational awareness of help-seekers' mental states. Our supportive decoder integrates information from diverse knowledge sources, underpinning response generation that is both empathetic and cognitively aware. The effectiveness of D2RCU is demonstrated through extensive automatic and human evaluations, revealing substantial improvements over numerous state-of-the-art models, with up to 13.79% enhancement in overall performance of ten metrics. Our codes are available for public access to facilitate further research and development.

A home automation system means monitoring and controlling various kinds of devices in the home remotely using the Internet of things (IoT). Technologies such as natural language processing techniques, user-friendly visual programming, and machine intelligence programming are already available for home automation. For such systems, the increase in the number of devices often makes users focused on the system’s ability to perform complex or composing tasks. However, some existing natural language processing systems can only perform simple tasks and cannot meet users’ needs. Thus, it is difficult for users to develop the home automation systems they need using visual programming systems because of the large amount of programming knowledge required. Meanwhile, automatic programming without user action can only write a few lines of code and implement little functionality. There are relatively few tools available for generating home automation scripting languages. To address this problem, we propose a practical method for generating executable home automation scripts using Chinese texts. Our method includes the following steps: it extracts critical information from the command sentences in Chinese; it uses first-order logic to check the validity of the extracted information; based on the validation, the correct sentences are mapped into the intermediate language scripts, which can interface with different home platforms. We conducted experiments on Home Assistant, converted intermediate scripts to Home Assistant, and collected 600 scenario descriptions. The experimental results show that the method can automatically generate executable scripts for the Home Assistant platform, and the correct rate was 93.66%.