Ensuring the safety of large language models (LLMs) is critical for responsible deployment, yet existing evaluations often prioritize performance over identifying failure modes. We introduce Phare, a multilingual diagnostic framework to probe and evaluate LLM behavior across three critical dimensions: hallucination and reliability, social biases, and harmful content generation. Our evaluation of 17 state-of-the-art LLMs reveals patterns of systematic vulnerabilities across all safety dimensions, including sycophancy, prompt sensitivity, and stereotype reproduction. By highlighting these specific failure modes rather than simply ranking models, Phare provides researchers and practitioners with actionable insights to build more robust, aligned, and trustworthy language systems.

The goal of achieving Artificial General Intelligence (AGI) is to imitate humans and surpass them. Models such as OpenAI's o1, o3, and DeepSeek's R1 have demonstrated that large language models (LLMs) with human-like reasoning capabilities exhibit exceptional performance and are being gradually integrated into multimodal large language models (MLLMs). However, whether these models possess capabilities comparable to humans in handling reasoning tasks remains unclear at present. In this paper, we propose Human-Aligned Bench, a benchmark for fine-grained alignment of multimodal reasoning with human performance. Specifically, we collected 9,794 multimodal questions that solely rely on contextual reasoning, including bilingual (Chinese and English) multimodal questions and pure text-based questions, encompassing four question types: visual reasoning, definition judgment, analogical reasoning, and logical judgment. More importantly, each question is accompanied by human success rates and options that humans are prone to choosing incorrectly. Extensive experiments on the Human-Aligned Bench reveal notable differences between the performance of current MLLMs in multimodal reasoning and human performance. The findings on our benchmark provide insights into the development of the next-generation models.

--Electronic Health Records (EHRs) are digital records of patient information, often containing unstructured clinical text. Named Entity Recognition (NER) is essential in EHRs for extracting key medical entities like problems, tests, and treatments to support downstream clinical applications. This paper explores prompt-based medical entity recognition using large language models (LLMs), specifically GPT -4o and DeepSeek-R1, guided by various prompt engineering techniques, including zero-shot, few-shot, and an ensemble approach. Among all strategies, GPT -4o with prompt ensemble achieved the highest classification performance with an F1-score of 0.95 and recall of 0.98, outperforming DeepSeek-R1 on the task. The ensemble method improved reliability by aggregating outputs through embedding-based similarity and majority voting. Electronic Health Records (EHRs) are digital systems that store and manage patient information across clinical encounters. Initially designed to support administrative tasks like billing and scheduling, EHRs have now become essential tools in healthcare delivery and research. Over the past two decades, adoption of EHRs has grown rapidly, especially in the United States [1].

We present AM-Thinking-v1, a 32B dense language model that advances the frontier of reasoning, embodying the collaborative spirit of open-source innovation. Outperforming DeepSeek-R1 and rivaling leading Mixture-of-Experts (MoE) models like Qwen3-235B-A22B and Seed1.5-Thinking, AM-Thinking-v1 achieves impressive scores of 85.3 on AIME 2024, 74.4 on AIME 2025, and 70.3 on LiveCodeBench, showcasing state-of-the-art mathematical and coding capabilities among open-source models of similar scale. Built entirely from the open-source Qwen2.5-32B base model and publicly available queries, AM-Thinking-v1 leverages a meticulously crafted post-training pipeline - combining supervised fine-tuning and reinforcement learning - to deliver exceptional reasoning capabilities. This work demonstrates that the open-source community can achieve high performance at the 32B scale, a practical sweet spot for deployment and fine-tuning. By striking a balance between top-tier performance and real-world usability, we hope AM-Thinking-v1 inspires further collaborative efforts to harness mid-scale models, pushing reasoning boundaries while keeping accessibility at the core of innovation. We have open-sourced our model on \href{https://huggingface.co/a-m-team/AM-Thinking-v1}{Hugging Face}.

Recent studies have uncovered a troubling vulnerability in the fine-tuning stage of large language models (LLMs): even fine-tuning on entirely benign datasets can lead to a significant increase in the harmfulness of LLM outputs. Building on this finding, our red teaming study takes this threat one step further by developing a more effective attack. Specifically, we analyze and identify samples within benign datasets that contribute most to safety degradation, then fine-tune LLMs exclusively on these samples. We approach this problem from an outlier detection perspective and propose Self-Inf-N, to detect and extract outliers for fine-tuning. Our findings reveal that fine-tuning LLMs on 100 outlier samples selected by Self-Inf-N in the benign datasets severely compromises LLM safety alignment. Extensive experiments across seven mainstream LLMs demonstrate that our attack exhibits high transferability across different architectures and remains effective in practical scenarios. Alarmingly, our results indicate that most existing mitigation strategies fail to defend against this attack, underscoring the urgent need for more robust alignment safeguards. Codes are available at https://github.com/GuanZihan/Benign-Samples-Matter.

Despite growing interest in automated hate speech detection, most existing approaches overlook the linguistic diversity of online content. Multilingual instruction-tuned large language models such as LLaMA, Aya, Qwen, and BloomZ offer promising capabilities across languages, but their effectiveness in identifying hate speech through zero-shot and few-shot prompting remains underexplored. This work evaluates LLM prompting-based detection across eight non-English languages, utilizing several prompting techniques and comparing them to fine-tuned encoder models. We show that while zero-shot and few-shot prompting lag behind fine-tuned encoder models on most of the real-world evaluation sets, they achieve better generalization on functional tests for hate speech detection. Our study also reveals that prompt design plays a critical role, with each language often requiring customized prompting techniques to maximize performance.

While contemporary speech separation technologies adeptly process lengthy mixed audio waveforms, they are frequently challenged by the intricacies of real-world environments, including noisy and reverberant settings, which can result in artifacts or distortions in the separated speech. To overcome these limitations, we introduce SepALM, a pioneering approach that employs audio language models (ALMs) to rectify and re-synthesize speech within the text domain following preliminary separation. SepALM comprises four core components: a separator, a corrector, a synthesizer, and an aligner. By integrating an ALM-based end-to-end error correction mechanism, we mitigate the risk of error accumulation and circumvent the optimization hurdles typically encountered in conventional methods that amalgamate automatic speech recognition (ASR) with large language models (LLMs). Additionally, we have developed Chain-of-Thought (CoT) prompting and knowledge distillation techniques to facilitate the reasoning and training processes of the ALM. Our experiments substantiate that SepALM not only elevates the precision of speech separation but also markedly bolsters adaptability in novel acoustic environments.

Tree-search-based reasoning methods have significantly enhanced the reasoning capability of large language models (LLMs) by facilitating the exploration of multiple intermediate reasoning steps, i.e., thoughts. However, these methods suffer from substantial inference latency, as they have to generate numerous reasoning thoughts, severely limiting LLM applicability. To address this challenge, we propose a novel Speculative Search (SpecSearch) framework that significantly accelerates LLM reasoning by optimizing thought generation. Specifically, SpecSearch utilizes a small model to strategically collaborate with a large model at both thought and token levels, efficiently generating high-quality reasoning thoughts. The major pillar of SpecSearch is a novel quality-preserving rejection mechanism, which effectively filters out thoughts whose quality falls below that of the large model's outputs. Moreover, we show that SpecSearch preserves comparable reasoning quality to the large model. Experiments on both the Qwen and Llama models demonstrate that SpecSearch significantly outperforms state-of-the-art approaches, achieving up to 2.12$\times$ speedup with comparable reasoning quality.

As large language models (LLMs) continue to advance in capabilities, it is essential to assess how they perform on established benchmarks. In this study, we present a suite of experiments to assess the performance of modern LLMs (ranging from 3B to 90B+ parameters) on CaseHOLD, a legal benchmark dataset for identifying case holdings. Our experiments demonstrate scaling effects - performance on this task improves with model size, with more capable models like GPT4o and AmazonNovaPro achieving macro F1 scores of 0.744 and 0.720 respectively. These scores are competitive with the best published results on this dataset, and do not require any technically sophisticated model training, fine-tuning or few-shot prompting. To ensure that these strong results are not due to memorization of judicial opinions contained in the training data, we develop and utilize a novel citation anonymization test that preserves semantic meaning while ensuring case names and citations are fictitious. Models maintain strong performance under these conditions (macro F1 of 0.728), suggesting the performance is not due to rote memorization. These findings demonstrate both the promise and current limitations of LLMs for legal tasks with important implications for the development and measurement of automated legal analytics and legal benchmarks.

We challenge the dominant focus on neural scaling laws and advocate for a paradigm shift toward downscaling in the development of large language models (LLMs). While scaling laws have provided critical insights into performance improvements through increasing model and dataset size, we emphasize the significant limitations of this approach, particularly in terms of computational inefficiency, environmental impact, and deployment constraints. To address these challenges, we propose a holistic framework for downscaling LLMs that seeks to maintain performance while drastically reducing resource demands. This paper outlines practical strategies for transitioning away from traditional scaling paradigms, advocating for a more sustainable, efficient, and accessible approach to LLM development.