While Large Language Models (LLMs) have demonstrated significant potential in Tool-Integrated Reasoning (TIR), existing training paradigms face significant limitations: Zero-RL suffers from inefficient exploration and mode degradation due to a lack of prior guidance, while SFT-then-RL is limited by high data costs and capability plateaus caused by low-entropy collapse. To address these challenges, we propose E3-TIR (Enhanced Experience Exploitation), a warm-up paradigm for the early stages of agent training. Specifically, we formulate training as the dynamic integration of three experience types: Expert Prefixes, Expert Guided, and Self-Exploration. By executing diverse branching exploration around expert “anchors’’ and employing a mix policy optimization mechanism, we effectively mitigate distribution shifts and resolve optimization conflicts arising from shared prefixes. Our method dynamically adapts the model’s knowledge boundaries, effectively balancing exploration diversity with training efficiency. Experimental results demonstrate that E3-TIR achieves a 6% performance improvement over traditional paradigms on tool-use tasks, while requiring less than 10% of the synthetic data. Furthermore, in terms of ROI—a comprehensive metric integrating performance, data cost, and training efficiency—we achieve a 1.46 gain compared to baselines.

Vision-language models (VLMs) are increasingly deployed in multi-turn settings where users may describe visual content with incorrect assumptions. Yet existing evaluations rarely isolate how models respond when the same visually grounded false premise persists across dialogue turns. We introduce FPCO-Dialog, a benchmark for evaluating correction and cooperation behavior in VLMs under repeated false premises. FPCO-Dialog contains 1,080 images and 10,800 question turns, stratified by visual complexity, object category, and false-premise class, and uses a 10-turn protocol in which a correct dialogue prefix is followed by repeated false-premise referring expressions. We evaluate 20 commercial and open-source VLMs with a model-agnostic protocol and CorrTP@K, a correction-rate metric over false-premise turns, scored by two independent detectors. FPCO-Dialog reveals substantial and persistent cross-model differences in aggregate correction tendency, model-specific turn-wise dynamics, and systematic variation across false-premise types under the benchmark’s substitution distribution. The dataset, evaluation protocol, model outputs, detector labels, and code are available at https://github.com/lab-klc/FPCO-Dialog.

Large-scale content moderation requires models that can apply complex, evolving policies under limited expert supervision. We present an empirical study of reinforcement learning (RL) for policy-grounded moderation across three production-derived tasks. Using established RL and inference primitives, we develop three integrity-specific designs: a policy-grounded full-trace reward, rationale-marginalized Monte Carlo scoring, and policy-structured reflection. Each design addresses an observed failure mode, including reasoning-length collapse and polarized confidence estimates. Across the three tasks, RL-Only systems trained on 200–836 labeled examples remain competitive with supervised fine-tuning (SFT) systems trained on thousands of examples, corresponding to up to approximately one order of magnitude greater label efficiency at the evaluated operating points. Performance generally improves with additional data and training compute before saturating, although larger rollout counts exhibit variability. The strongest SFT→RL systems attain point-estimate performance comparable to policy-trained full-time employee (FTE) reviewers across the three tasks.

Phonetic Cloaking Replacement (PCR), defined as the deliberate use of homophonic or near-homophonic variants to hide toxic intent, has become a major obstacle to Chinese content moderation. While this problem is well-recognized, existing evaluations predominantly rely on rule-based, synthetic perturbations that ignore the creativity of real users. We organize PCR into a four-way surface-form taxonomy and compile PCR-ToxiCN, a dataset of 500 naturally occurring, phonetically cloaked offensive posts gathered from the RedNote platform. Benchmarking state-of-the-art LLMs on this dataset exposes a serious weakness: the best model reaches only an F1-score of 0.672, and zero-shot chain-of-thought prompting pushes performance even lower. Guided by error analysis, we revisit a Pinyin-based prompting strategy that earlier studies judged ineffective and show that it recovers much of the lost accuracy. This study offers the first comprehensive taxonomy of Chinese PCR, a realistic benchmark that reveals current detectors’ limits, and a lightweight mitigation technique that advances research on robust toxicity detection.

Mental manipulation severely undermines mental wellness by covertly and negatively distorting decision-making. While there is an increasing interest in mental health care within the natural language processing community, progress in tackling manipulation remains limited due to the complexity of detecting subtle, covert tactics in conversations. In this paper, we propose Intent-Aware Prompting (IAP), a novel approach for detecting mental manipulations using large language models (LLMs), providing a deeper understanding of manipulative tactics by capturing the underlying intents of participants. Experimental results on the MentalManip dataset demonstrate superior effectiveness of IAP against other advanced prompting strategies. Notably, our approach substantially reduces false negatives, helping detect more instances of mental manipulation with minimal misjudgment of positive cases. The code of this paper is available at https://github.com/Anton-Jiayuan-MA/Manip-IAP.

Social media data is recognized for its usefulness in the early detection of mental disorders; however, there is a lack of research focused on modeling individuals’ longitudinal mental health dynamics. Moreover, fine-tuning large language models (LLMs) on large-scale, annotated datasets presents challenges due to privacy concerns and the difficulties on data collection and annotation. In this paper, we propose a novel approach for modeling mental health dynamics using hybrid LLMs, where we first apply both classification-based and generation-based models to identify adaptive and maladaptive evidence from individual posts. This evidence is then used to predict well-being scores and generate post-level and timeline-level summaries. Experimental results on the CLPsych 2025 shared task demonstrate the effectiveness of our method, with the generative-based model showing a marked advantage in evidence identification.