Large Language models have demonstrated promising performance in research ideation across scientific domains. Hypothesis development, the process of generating a highly specific declarative statement connecting a research idea with empirical validation, has received relatively less attention. Existing approaches trivially deploy retrieval augmentation and focus only on the quality of the final output ignoring the underlying reasoning process behind ideation. We present $\texttt{HypER}$ ($\textbf{Hyp}$othesis Generation with $\textbf{E}$xplanation and $\textbf{R}$easoning), a small language model (SLM) trained for literature-guided reasoning and evidence-based hypothesis generation. $\texttt{HypER}$ is trained in a multi-task setting to discriminate between valid and invalid scientific reasoning chains in presence of controlled distractions. We find that $\texttt{HypER}$ outperformes the base model, distinguishing valid from invalid reasoning chains (+22\% average absolute F1), generates better evidence-grounded hypotheses (0.327 vs. 0.305 base model) with high feasibility and impact as judged by human experts ($>$3.5 on 5-point Likert scale).

Direct Alignment Algorithms (DAAs), such as Direct Preference Optimization (DPO) and Simple Preference Optimization (SimPO), have emerged as efficient alternatives to Reinforcement Learning from Human Feedback (RLHF) algorithms for aligning large language models (LLMs) with human preferences. However, DAAs suffer from a fundamental limitation we identify as the "reward-generation gap" -- a misalignment between optimization objectives during training and actual generation performance during inference. In this paper, we find a contributor to the reward-generation gap is the mismatch between the inherent importance of prefix tokens during the LLM generation process and how this importance is reflected in the implicit reward functions of DAAs. To bridge the gap, we adopt a token-level MDP perspective of DAAs to analyze its limitations and introduce a simple yet effective approach called Prefix-Oriented Equal-length Training (POET), which truncates both preferred and dispreferred responses to match the shorter one's length. Training with \mname, where both responses in each sample are truncated to equal length, resulting in diverse truncated lengths across samples, the optimization of DAAs objective is implicitly constrained to converge across all timesteps of token-level MDP, thus paying more attention to prefix tokens than the standard DAAs. We conduct experiments with DPO and SimPO, two representative DAAs, demonstrating that POET improves over their standard implementations, achieving up to 15.6 points in AlpacaEval 2 and overall improvements across downstream tasks. Our results highlight the importance of addressing the misalignment between reward optimization and generation performance in DAAs.

While Large Reasoning Models (LRMs) have demonstrated success in complex reasoning tasks through long chain-of-thought (CoT) reasoning, their inference often involves excessively verbose reasoning traces, resulting in substantial inefficiency. To address this, we propose Distilled Reasoning Pruning (DRP), a hybrid framework that combines inference-time pruning with tuning-based distillation, two widely used strategies for efficient reasoning. DRP uses a teacher model to perform skill-aware step decomposition and content pruning, and then distills the pruned reasoning paths into a student model, enabling it to reason both efficiently and accurately. Across several challenging mathematical reasoning datasets, we find that models trained with DRP achieve substantial improvements in token efficiency without sacrificing accuracy. Specifically, DRP reduces average token usage on GSM8K from 917 to 328 while improving accuracy from 91.7% to 94.1%, and achieves a 43% token reduction on AIME with no performance drop. Further analysis shows that aligning the reasoning structure of training CoTs with the student's reasoning capacity is critical for effective knowledge transfer and performance gains.

Large reasoning models, such as OpenAI o1 or DeepSeek R1, have demonstrated remarkable performance on reasoning tasks but often incur a long reasoning path with significant memory and time costs. Existing methods primarily aim to shorten reasoning paths by introducing additional training data and stages. In this paper, we propose three critical reward designs integrated directly into the reinforcement learning process of large reasoning models, which reduce the response length without extra training stages. Experiments on four settings show that our method significantly decreases response length while maintaining or even improving performance. Specifically, in a logic reasoning setting, we achieve a 40% reduction in response length averaged by steps alongside a 14% gain in performance. For math problems, we reduce response length averaged by steps by 33% while preserving performance.

Large Language Models (LLMs) have demonstrated significant promise for various applications in healthcare. However, their effectiveness in the Arabic medical domain remains unexplored due to the lack of high-quality domain-specific datasets and benchmarks. This study introduces MedArabiQ, a new benchmark dataset consisting of seven Arabic medical tasks, covering multiple specialties and including multiple-choice questions, fill-in-the-blank questions, and patient-doctor questions and answers. We first constructed the dataset using past medical exams as well as publicly available datasets. We conducted an extensive evaluation with eight state-of-the-art open-access and proprietary high-resource LLMs, including GPT-4, Deepseek v3, and Gemini 1.5. Our findings highlight the need for the creation of new high-quality benchmarks that span different languages to ensure fair deployment and scalability of LLMs in healthcare. By establishing this benchmark and releasing the dataset, we provide a foundation for future research aimed at evaluating and enhancing the multilingual capabilities of LLMs for the equitable use of generative AI in healthcare. Data Availability In this article, we present a new benchmark dataset, MedArabiQ, designed to evaluate the performance of LLMs on Arabic medical tasks.

Large language models (LLMs) are commonly trained on multi-domain datasets, where domain sampling strategies significantly impact model performance due to varying domain importance across downstream tasks. Existing approaches for optimizing domain-level sampling strategies struggle with maintaining intra-domain consistency and accurately measuring domain impact. In this paper, we present Domain Impact-aware Data Sampling (DIDS). To ensure intra-domain consistency, a gradient clustering algorithm is proposed to group training data based on their learning effects, where a proxy language model and dimensionality reduction are employed to reduce computational overhead. To accurately measure domain impact, we develop a Fisher Information Matrix (FIM) guided metric that quantifies how domain-specific parameter updates affect the model's output distributions on downstream tasks, with theoretical guarantees. Furthermore, to determine optimal sampling ratios, DIDS combines both the FIM-guided domain impact assessment and loss learning trajectories that indicate domain-specific potential, while accounting for diminishing marginal returns. Extensive experiments demonstrate that DIDS achieves 3.4% higher average performance while maintaining comparable training efficiency. The code is available at https://github.com/shiweijiezero/DIDS.

Plan-guided summarization attempts to reduce hallucinations in small language models (SLMs) by grounding generated summaries to the source text, typically by targeting fine-grained details such as dates or named entities. In this work, we investigate whether plan-based approaches in SLMs improve summarization in long document, narrative tasks. Narrative texts' length and complexity often mean they are difficult to summarize faithfully. We analyze existing plan-guided solutions targeting fine-grained details, and also propose our own higher-level, narrative-based plan formulation. Our results show that neither approach significantly improves on a baseline without planning in either summary quality or faithfulness. Human evaluation reveals that while plan-guided approaches are often well grounded to their plan, plans are equally likely to contain hallucinations compared to summaries. As a result, the plan-guided summaries are just as unfaithful as those from models without planning. Our work serves as a cautionary tale to plan-guided approaches to summarization, especially for long, complex domains such as narrative texts. Code available at https://github.com/amazon-science/plan-guided-summarization

Although LLM-based agents have attracted significant attention in domains such as software engineering and machine learning research, their role in advancing combinatorial optimization (CO) remains relatively underexplored. This gap underscores the need for a deeper understanding of their potential in tackling structured, constraint-intensive problems -- a pursuit currently limited by the absence of comprehensive benchmarks for systematic investigation. To address this, we introduce CO-Bench, a benchmark suite featuring 36 real-world CO problems drawn from a broad range of domains and complexity levels. CO-Bench includes structured problem formulations and curated data to support rigorous investigation of LLM agents. We evaluate multiple agentic frameworks against established human-designed algorithms, revealing the strengths and limitations of existing LLM agents and identifying promising directions for future research. CO-Bench is publicly available at https://github.com/sunnweiwei/CO-Bench.

We introduce MultiBLiMP 1.0, a massively multilingual benchmark of linguistic minimal pairs, covering 101 languages and 2 types of subject-verb agreement, containing more than 128,000 minimal pairs. Our minimal pairs are created using a fully automated pipeline, leveraging the large-scale linguistic resources of Universal Dependencies and UniMorph. MultiBLiMP 1.0 evaluates abilities of LLMs at an unprecedented multilingual scale, and highlights the shortcomings of the current state-of-the-art in modelling low-resource languages.

The boss of the firm behind ChatGPT and the UK technology secretary discussed a multibillion-pound deal to give the entire country premium access to the AI tool, the Guardian has learned. Sam Altman, a co-founder of OpenAI, talked to Peter Kyle about a potential agreement to give UK residents access to its advanced product. According to two sources with direct knowledge of the meeting, the idea was floated as part of a broader discussion in San Francisco about opportunities for collaboration between OpenAI and the UK. Those close to the discussion say Kyle never really took the idea seriously, not least because it could have cost as much as 2bn. OpenAI offers free and subscription versions of ChatGPT.