The conversational capabilities of large language models hold significant promise for enabling scalable and interactive tutoring. Prior work has primarily evaluated general instructional competence while overlooking a critical dimension: the ability to adaptively guide learners based on their evolving cognitive states. This study takes instructional guidance as a core capability and asks whether LLMs can emulate expert tutors who dynamically adjust strategies in response to learners’ needs. We introduce GuideEval, a benchmark grounded in authentic educational dialogues that evaluates pedagogical guidance through a three-phase behavioral framework: (i) Assessment, inferring learner states; (ii) Orchestration, adapting instructional strategies; and (iii) Elicitation, stimulating proper reflections. Empirical results indicate that existing LLMs often fail to provide effective adaptive scaffolding when learners experience confusion or require redirection. Complementing quantitative results, we present a detailed failure case analysis, offering intuitive insights into these limitations. We further propose a behavior-guided finetuning strategy based on behavior-prompted instructional dialogues, substantially enhancing guidance performance. By shifting the focus from isolated content correctness to learner-centered state-aware interaction, this work advocates a more dialogic paradigm for evaluating Socratic LLMs.

Long-form instructional videos require automatic chaptering for efficient organization and navigation. While long-context language models can predict chapters from textualized video representations, they become costly for content-dense videos with long transcripts, smooth topic transitions, and detailed chapter descriptions. A scalable segment-then-caption paradigm reduces this cost but suffers from boundary error propagation and fragmented cross-chapter context. We propose CausalChapter, a causal-inspired interventional dependency enhancement framework for long-video chaptering. Our method estimates the observable influence of semantic units and context segments through lightweight masking and removal interventions. For boundary localization, it detects dependency changes between adjacent temporal windows via masking-based feature reconstruction; for chapter description generation, it selects cross-chapter contexts according to their influence on generated descriptions. Experiments on long-video chaptering benchmarks show consistent improvements in boundary localization, description quality, and cross-chapter coherence.

Despite exceptional performance on static benchmarks, Large Language Models (LLMs) exhibit significant brittleness when evaluated on perturbed variants of the original questions. Current extrinsic interventions, prompt engineering or supervised fine-tuning, overlook an important intrinsic factor underlying this degradation: the distributional shifts in layer activations between original and perturbed questions. To address this, this work formulates the reasoning robustness challenge as an Unsupervised Domain Adaptation (UDA) problem and proposes the Elastic Domain Adaptation (EDA) method. Unlike rigid strategies that enforce a uniform alignment constraint, EDA dynamically aligns layer activations based on the specific layers and perturbation types. Specifically, EDA incorporates an Adaptive Layer Selection (ALS) to identify pivotal layers and employs an Elastic Shift Module (ESM) to selectively pull or push layer representations depending on the perturbation types. Extensive experiments across six LLMs and 15 perturbation types demonstrate that EDA achieves state-of-the-art performance on most perturbed benchmarks while preserving reasoning capability on original benchmarks.

Equipping Large Language Models (LLMs) with pedagogical tutoring capabilities holds significant promise for education. Existing approaches simulate tutor behaviors or preferences and use them to prompt or fine-tune LLMs for dialogue tutoring. However, such methods often fail to sustain high-quality pedagogical conversations that provide explicit stepwise scaffolding and adapt to learners’ evolving cognitive states. To address this, we propose ScaffoldLM, a planning-guided tutoring framework with an assessment-driven memory for multi-turn math dialogue tutoring. ScaffoldLM first generates a stepwise pedagogical plan from solution steps, which serves as a stable backbone for explicit scaffolding. During tutoring, the tutoring memory is updated by an assessment-driven control loop that infers the learner’s cognitive state, evaluates whether the current step target is met, and adaptively selects tutoring actions. The plan, step-level progress, inferred learner states, and dialogue history are maintained in memory to support coherent multi-turn guidance. Experiments on multi-turn math tutoring benchmarks demonstrate that ScaffoldLM substantially improves pedagogical tutoring quality over strong baselines. Code is publicly available at https://github.com/BNU-ERC-ITEA/ScaffoldLM.

Large Language Models (LLMs) have achieved remarkable performance across a wide range of mathematical benchmarks. However, concerns remain as to whether these successes reflect genuine reasoning or superficial pattern recognition. Existing evaluation methods, which typically focus either on the final answer or on the intermediate reasoning steps, reduce mathematical reasoning to a shallow input–output mapping, overlooking its inherently multi-stage and multi-dimensional cognitive nature. Inspired by P’olya’s problem-solving theory, we propose SMART, a benchmark that decomposes mathematical problem-solving into four cognitive dimensions: Semantic Understanding, Mathematical Reasoning, Arithmetic Computation, and Reflection RefinemenT, and introduces dimension-specific tasks to measure the corresponding cognitive processes of LLMs. We apply SMART to 22 state-of-the-art open- and closed-source LLMs and uncover substantial discrepancies in their capabilities across dimensions. Our findings reveal genuine weaknesses in current models and motivate a new metric, the All-Pass Score, designed to better capture true problem-solving capability.

In-context learning (ICL) can significantly enhance the complex reasoning capabilities of large language models (LLMs), with the key lying in the selection and ordering of demonstration examples. Previous methods typically relied on simple features to measure the relevance between examples. We argue that these features are not sufficient to reflect the intrinsic connections between examples. In this study, we propose a curriculum ICL strategy guided by problem-solving logic. We select demonstration examples by analyzing the problem-solving logic and order them based on curriculum learning. Specifically, we constructed a problem-solving logic instruction set based on the BREAK dataset and fine-tuned a language model to analyze the problem-solving logic of examples. Subsequently, we selected appropriate demonstration examples based on problem-solving logic and assessed their difficulty according to the number of problem-solving steps. In accordance with the principles of curriculum learning, we ordered the examples from easy to hard to serve as contextual prompts. Experimental results on multiple benchmarks indicate that our method outperforms previous ICL approaches in terms of performance and efficiency, effectively enhancing the complex reasoning capabilities of LLMs. Our project will be publicly available subsequently.

Existing reinforcement learning (RL) strategies based on outcome supervision have proven effective in enhancing the performance of large language models (LLMs) for code generation. While reinforcement learning based on process supervision shows great potential in multi-step reasoning tasks, its effectiveness in the field of code generation still lacks sufficient exploration and verification. The primary obstacle stems from the resource-intensive nature of constructing a high-quality process-supervised reward dataset, which requires substantial human expertise and computational resources. To overcome this challenge, this paper proposes a “mutation/refactoring-execution verification” strategy. Specifically, the teacher model is used to mutate and refactor the statement lines or blocks, and the execution results of the compiler are used to automatically label them, thus generating a process-supervised reward dataset. Based on this dataset, we have carried out a series of RL experiments. The experimental results show that, compared with the method relying only on outcome supervision, reinforcement learning based on process supervision performs better in handling complex code generation tasks. In addition, this paper for the first time confirms the advantages of the Direct Preference Optimization (DPO) method in the RL task of code generation based on process supervision, providing new ideas and directions for code generation research.