Despite strong performance in medical question answering, large language models remain limited in clinical cold-start settings, where little patient information is initially available and key evidence must be acquired through dialogue. This requires models to decide what to ask under a limited turn budget, balancing broad history taking with targeted clarification of high-risk cues. To address the aforementioned challenges, we propose **MedCIA**, a two-stage framework for improving clinical information acquisition. First, it reconstructs patient records into structured inquiry trajectories, enabling supervised learning of expert-guided information-seeking intents and the breadth-depth trade-off in clinical questioning. Second, it refines the policy with Adaptive Tree Policy Optimization (ATPO), an uncertainty-aware reinforcement learning procedure that allocates rollout computation to ambiguous dialogue states in a profile-anchored patient simulator. Furthermore, we curate a multi-turn dataset (MedIntake) for active clinical intake with progressive patient information disclosure and diverse inquiry strategies. Experiments on an EHR-derived benchmark show that **MedCIA** improves importance-weighted coverage, early acquisition, and cost control under controlled simulation.

While large language models (LLMs) show immense potential in specialized domains, their adaptation is often hindered by limited access to domain-specific data and the high computational cost of model training. To address these challenges, we propose a lightweight knowledge injection framework. Instead of sharing raw data directly, our approach fine-tunes a small model and transfers its knowledge vectors to the target LLM for knowledge injection, thereby reducing data exposure and computational overhead. To mitigate the knowledge loss incurred during the injection process, we further introduce an iterative knowledge calibration strategy. We theoretically bound the error of the transformation matrix and prove the stability and linear convergence of the calibration process. Comprehensive experiments demonstrate that our method significantly outperforms parameter-efficient baselines and achieves performance comparable to full-parameter fine-tuning, while consuming only 28% of the GPU memory and 3% of the raw training data. Code is available at https://github.com/Yuhanleeee/LiSKI.

Large language models (LLMs) are currently applied to scientific paper evaluation by assigning an absolute score to each paper independently. However, since score scales vary across conferences, time periods, and evaluation criteria, models trained on absolute scores are prone to fitting narrow, context-specific rules rather than developing robust scholarly judgment. To overcome this limitation, we propose shifting paper evaluation from isolated scoring to collaborative ranking. In particular, we design a Comparison-Native framework for Paper Evaluation (CNPE), integrating comparison into both data construction and model learning. We first propose a graph-based similarity ranking algorithm to facilitate the sampling of more informative and discriminative paper pairs from a collection. We then enhance relative quality judgment through supervised fine-tuning and reinforcement learning with comparison-based rewards. At inference, the model performs pairwise comparisons over sampled paper pairs and aggregates these preference signals into a global relative quality ranking. Experimental results demonstrate that our framework achieves an average relative improvement of 21.8% over the strong baseline DeepReview-14B, while exhibiting robust generalization to five previously unseen datasets. Our code is available at https://github.com/ECNU-Text-Computing/ComparisonReview.

Complex reasoning over tabular data is crucial in real-world data analysis, yet large language models (LLMs) often underperform due to complex queries, noisy data, and limited numerical capabilities. To address these issues, we propose TabDSR, a three-agent framework consisting of: (1) a query decomposer that breaks down complex questions, (2) a table sanitizer that cleans and filters noisy tables, and (3) a program-of-thoughts (PoT)-based reasoner that generates executable code to derive the final answer from the sanitized table. To ensure unbiased evaluation and mitigate data leakage, we introduce a new dataset, CalTab151, specifically designed for complex numerical reasoning over tables. Experimental results demonstrate that TabDSR consistently outperforms existing methods, achieving state-of-the-art (SOTA) performance with 8.79%, 6.08%, and 19.87% accuracy improvement on TAT-QA, TableBench, and CalTab151, respectively. Moreover, our framework integrates seamlessly with mainstream LLMs, providing a robust solution for complex tabular numerical reasoning. These findings highlight the effectiveness of our framework in enhancing LLM performance for complex tabular numerical reasoning.

Evaluating large language models (LLMs) is fundamental, particularly in the context of practical applications. Conventional evaluation methods, typically designed primarily for LLM development, yield numerical scores that ignore the user experience. Therefore, our study shifts the focus from model-centered to human-centered evaluation in the context of AI-powered writing assistance applications. Our proposed metric, termed “Revision Distance,” utilizes LLMs to suggest revision edits that mimic the human writing process. It is determined by counting the revision edits generated by LLMs. Benefiting from the generated revision edit details, our metric can provide a self-explained text evaluation result in a human-understandable manner beyond the context-independent score. Our results show that for the easy-writing task, “Revision Distance” is consistent with established metrics (ROUGE, Bert-score, and GPT-score), but offers more insightful, detailed feedback and better distinguishes between texts. Moreover, in the context of challenging academic writing tasks, our metric still delivers reliable evaluations where other metrics tend to struggle. Furthermore, our metric also holds significant potential for scenarios lacking reference texts.