Recent VLMs improve multimodal reasoning by generating explicit chains of thought, but often overthink with redundant reasoning. Although conventional length penalties mitigate this redundancy, they rely on coarse global rewards that fail to discern intermediate process quality or adapt to task complexity. To address this dilemma, we propose Binned Relative Policy Optimization (BRPO), a training-efficient framework that enables non-trivial process-level evaluation without introducing costly process reward models. Specifically, BRPO leverages high entropy lexical markers to segment reasoning trajectories into discrete intermediate prefixes, evaluating their problem-solving potential through forced answer completions. Building on this, we introduce Bin Normalization along with a length-aware adjustment to map prefixes into process bins, dynamically calibrating rewards and regulating length constraints based on reasoning stages and visual complexity. Experiments on six multimodal math benchmarks show that our method improves both accuracy and efficiency: on Qwen2.5-VL-7B, it raises average accuracy from 47.8% to 53.6% while reducing the average response length by 39.0%.

Radiologists read CT as a sequential decision process, actively scrolling, zooming, adjusting windows, comparing with prior scans, and measuring structures, where each observation guides the next action. Current VLMs instead answer from passively sampled views, missing the multi-step reasoning that clinical reading demands. We address this by aligning the agent’s action space to the radiologists’ operations within the Picture Archiving and Communication System (PACS), enabling active, multi-step CT reading with inherently transparent reasoning for clinical decision support. In particular, we build **ACTOR**, a unified framework comprising: (1) a PACS-like interactive CT environment with five radiologist-aligned tools; (2) **ActiveCTR-Bench**, spanning 7 tasks across a Perception–Interpretation–Reasoning hierarchy (7,112 items with visual ground truth and verified vision necessity); and (3) **ActiveCTReasoner**, a 7B model trained on 108K synthesized trajectories via SFT and reinforcement learning. ActiveCTReasoner achieves 73.2% close-ended accuracy, outperforming GPT-5.4 (47.3%) remarkably. In open-ended evaluations, it scores 44.2% compared to 23.8% for the best baseline. A radiologist collaboration study provides preliminary evidence for the decision-support utility of radiologist-aligned agentic reasoning.

The rapid advancement of large language models (LLMs) has spurred significant interest in tool learning, where LLMs are augmented with external tools to tackle complex tasks. However, existing tool environments face challenges in balancing stability, scale, and realism, particularly for benchmarking purposes. To address this, we propose MirrorAPI, a novel framework that trains specialized LLMs to accurately simulate real API responses, effectively acting as “mirrors” to tool environments. Using a comprehensive dataset of request-response pairs from 7,000+ APIs, we employ supervised fine-tuning and chain-of-thought reasoning to enhance simulation fidelity. MirrorAPI achieves superior accuracy and stability compared to state-of-the-art methods, as demonstrated by its performance on the newly constructed MirrorAPI-Bench and its integration into StableToolBench.

Reasoning-based large language models have excelled in mathematics and programming, yet their potential in knowledge-intensive medical question answering remains underexplored and insufficiently validated in clinical contexts. To bridge this gap, we introduce ReasonMed, the largest medical reasoning dataset to date, comprising 370k high-quality examples distilled from 1.75 million initial reasoning paths generated by complementary LLMs and curated through a cost-efficient easy-medium-difficult (EMD) pipeline. ReasonMed is built through a multi-agent generation, verification, and refinement process, in which an Error Refiner improves reasoning paths by correcting error-prone steps identified by a verifier. Using ReasonMed, we investigate effective strategies for training medical reasoning models and find that integrating detailed CoT reasoning with concise answer summaries yields the most robust fine-tuning results. Models trained on ReasonMed set a new benchmark: ReasonMed-7B surpasses the prior best sub-10B models by 4.17% and even exceeds LLaMA3.1-70B on PubMedQA by 4.60%. When scaled to ReasonMed-14B, it remains highly competitive, underscoring consistent scaling potential. The codes and datasets are available at https://github.com/YuSun-Work/ReasonMed.

Continual pre-training (CPT) has been an important approach for adapting language models to specific domains or tasks. In this paper, we comprehensively study its key designs to balance the new abilities while retaining the original abilities, and present an effective CPT method that can greatly improve the Chinese language ability and scientific reasoning ability of LLMs. To achieve it, we design specific data mixture and curriculum strategies based on existing datasets and synthetic high-quality data. Concretely, we synthesize multidisciplinary scientific QA pairs based on related web pages to guarantee the data quality, and also devise the performance tracking and data mixture adjustment strategy to ensure the training stability. For the detailed designs, we conduct preliminary studies on a relatively small model, and summarize the findings to help optimize our CPT method. Extensive experiments on a number of evaluation benchmarks show that our approach can largely improve the performance of Llama-3 (8B), including both the general abilities (+8.81 on C-Eval and +6.31 on CMMLU) and the scientific reasoning abilities (+12.00 on MATH and +4.13 on SciEval). Our model, data, and codes are available at https://github.com/RUC-GSAI/Llama-3-SynE.

Knowledge representation learning (KRL) has been used in plenty of knowledge-driven tasks. Despite fruitfully progress, existing methods still suffer from the immaturity on tackling potentially-imperfect knowledge graphs and highly-imbalanced positive-negative instances during training, both of which would hinder the performance of KRL. In this paper, we propose Contrastive Completion Coding (C³), a novel KRL framework that is composed of two functional components: 1. Hierarchical Architecture, which integrates both low-level standalone features and high-level topology-aware features to yield robust embedding for each entity/relation. 2. Normalized Contrasitive Training, which conducts normalized one-to-many contrasitive learning to emphasize different negatives with different weights, delivering better convergence compared to conventional training losses. Extensive experiments on several benchmarks verify the efficacy of the two proposed techniques and combing them together generally achieves superior performance against state-of-the-art approaches.