The impact of Large Language Models (LLMs) has extended into literary domains. However, existing evaluation metrics for literature prioritize mechanical accuracy over artistic expression and tend to overrate machine translation as being superior to human translation from experienced professionals. In the long run, this bias could result in an irreversible decline in translation quality and cultural authenticity. In response to the urgent need for a specialized literary evaluation metric, we introduce LITRANSPROQA, a novel, reference-free, LLM-based question-answering framework designed for literary translation evaluation. LITRANSPROQA integrates humans in the loop to incorporate insights from professional literary translators and researchers, focusing on critical elements in literary quality assessment such as literary devices, cultural understanding, and authorial voice. Our extensive evaluation shows that while literary-finetuned XCOMET-XL yields marginal gains, LITRANSPROQA substantially outperforms current metrics, achieving up to 0.07 gain in correlation and surpassing the best state-of-the-art metrics by over 15 points in adequacy assessments. Incorporating professional translator insights as weights further improves performance, highlighting the value of translator inputs. Notably, LITRANSPROQA reaches an adequacy performance comparable to trained linguistic student evaluators, though it still falls behind experienced professional translators. LITRANSPROQA shows broad applicability to open-source models like LLaMA3.3-70b and Qwen2.5-32b, indicating its potential as an accessible and training-free tool for evaluating literary translations that require local processing due to copyright or ethical considerations.

We present ViDRiP-LLaVA, the first large multimodal model (LMM) in computational pathology that integrates three distinct image scenarios, including single patch images, automatically segmented pathology video clips, and manually segmented pathology videos. This integration closely mirrors the natural diagnostic process of pathologists. By generating detailed histological descriptions and culminating in a definitive sign-out diagnosis, ViDRiP-LLaVA bridges visual narratives with diagnostic reasoning. Central to our approach is the ViDRiP-Instruct dataset, comprising 4278 video and diagnosis-specific chain-of-thought instructional pairs sourced from educational histopathology videos on YouTube. Although high-quality data is critical for enhancing diagnostic reasoning, its creation is time-intensive and limited in volume. To overcome this challenge, we transfer knowledge from existing single-image instruction datasets to train on weakly annotated, keyframe-extracted clips, followed by fine-tuning on manually segmented videos. ViDRiP-LLaVA establishes a new benchmark in pathology video analysis and offers a promising foundation for future AI systems that support clinical decision-making through integrated visual and diagnostic reasoning. Our code, data, and model are publicly available at: https://github.com/QuIIL/ViDRiP-LLaVA.

Large reasoning models (LRMs), such as OpenAI-o1 and DeepSeek-R1, demonstrate impressive long-horizon reasoning capabilities. However, their reliance on static internal knowledge limits their performance on complex, knowledge-intensive tasks and hinders their ability to produce comprehensive research reports requiring synthesis of diverse web information. To address this, we propose WebThinker, a deep research agent that empowers LRMs to autonomously search the web, navigate among web pages, and draft reports during the reasoning process. WebThinker integrates a Deep Web Explorer module, enabling LRMs to dynamically search, navigate, and extract information from the web when encountering knowledge gaps. It also employs an Autonomous Think-Search-and-Draft strategy, allowing the model to seamlessly interleave reasoning, information gathering, and report writing in real time. To further enhance research tool utilization, we introduce an RL-based training strategy via iterative online Direct Preference Optimization (DPO). Extensive experiments on complex reasoning benchmarks (GPQA, GAIA, WebWalkerQA, HLE) and scientific report generation tasks (Glaive) demonstrate that WebThinker significantly outperforms existing methods and strong proprietary systems. Our approach enhances LRM reliability and applicability in complex scenarios, paving the way for more capable and versatile deep research systems. The code is available at https://github.com/RUC-NLPIR/WebThinker.

Achieving both accuracy and diverse reasoning remains challenging for Large Language Models (LLMs) in complex domains like mathematics. A key bottleneck is evaluating intermediate reasoning steps to guide generation without costly human annotations. To address this, we first introduce a novel Process Reward Model (PRM) trained automatically using Monte Carlo Tree Search coupled with a similarity-based data augmentation technique, effectively capturing step-level reasoning quality. Leveraging this PRM, we then adapt Generative Flow Networks (GFlowNets) to operate at the reasoning step level. Unlike traditional reinforcement learning focused on maximizing a single reward, GFlowNets naturally sample diverse, high-quality solutions proportional to their rewards, as measured by our PRM. Empirical evaluation shows strong improvements in both accuracy and solution diversity on challenging mathematical benchmarks (e.g., +2.59% absolute accuracy on MATH Level 5 for Llama3.2-3B), with effective generalization to unseen datasets (+9.4\% absolute on SAT MATH). Furthermore, we benchmark our PRM against existing open-source reward models, demonstrating superior alignment with reasoning quality and more consistent guidance for downstream generation. Our work demonstrates the potential of PRM-guided, step-level GFlowNets for developing more robust and versatile mathematical reasoning in LLMs.

Meat reduction benefits human and planetary health, but social norms keep meat central in shared meals. To date, the development of communication strategies that promote meat reduction while minimizing social costs has required the costly involvement of human participants at each stage of the process. We present work in progress on simulating multi-round dialogues on meat reduction between Generative Agents based on large language models (LLMs). We measure our main outcome using established psychological questionnaires based on the Theory of Planned Behavior and additionally investigate Social Costs. We find evidence that our preliminary simulations produce outcomes that are (i) consistent with theoretical expectations; and (ii) valid when compared to data from previous studies with human participants. Generative agent-based models are a promising tool for identifying novel communication strategies on meat reduction -- tailored to highly specific participant groups -- to then be tested in subsequent studies with human participants.

Continuous, high-frame-rate, high-resolution processing of long video streams is critical for future AI agents, yet current video-understanding LLMs struggle to scale. Offline, fixed-frame-number methods require the stream length to adapt frame rates; streaming methods constrain memory by merging or discarding tokens, losing information. We propose video-SALMONN S, a streaming audio-visual LLM that, to our knowledge, is the first to process 3-hour videos at 1 FPS and 360p resolution under a fixed memory budget. Our model introduces (i) a test-time-training (TTT) memory module that continually updates token representations to capture long-range dependencies by replacing token merging, and (ii) a prompt-dependent memory reader that selectively retrieves context-relevant content from fixed-size memory. The TTT module is optimised with a Hessian-free conjugate-gradient procedure (TTT_HF) for efficient adaptation. On long-video benchmarks (Video-MME, LVBench, VideoEvalPro), video-SALMONN S sustains high-quality understanding on multi-hour videos with 10k frames and 1M tokens. Our 8B-parameter model achieves 74.2% overall and 67.8% on the Video-MME long split, outperforming both offline and streaming baselines.

While large language models (LLMs) are increasingly used as automated heuristic designers for vehicle routing problems (VRPs), current state-of-the-art methods predominantly rely on prompting massive, general-purpose models like GPT-4. This work challenges that paradigm by demonstrating that a smaller, specialized LLM, when meticulously fine-tuned, can generate components that surpass expert-crafted heuristics within advanced solvers. We propose RFTHGS, a novel Reinforcement learning (RL) framework for Fine-Tuning a small LLM to generate high-performance crossover operators for the Hybrid Genetic Search (HGS) solver, applied to the Capacitated VRP (CVRP). Our method employs a multi-tiered, curriculum-based reward function that progressively guides the LLM to master generating first compilable, then executable, and finally, superior-performing operators that exceed human expert designs. This is coupled with an operator caching mechanism that discourages plagiarism and promotes diversity during training. Comprehensive experiments show that our fine-tuned LLM produces crossover operators which significantly outperform the expert-designed ones in HGS. The performance advantage remains consistent, generalizing from small-scale instances to large-scale problems with up to 1000 nodes. Furthermore, RFTHGS exceeds the performance of leading neuro-combinatorial baselines, prompt-based methods, and commercial LLMs such as GPT-4o and GPT-4o-mini.

The steady growth of artificial intelligence (AI) has accelerated in the recent years, facilitated by the development of sophisticated models such as large language models and foundation models. Ensuring robust and reliable power infrastructures is fundamental to take advantage of the full potential of AI. However, AI data centres are extremely hungry for power, putting the problem of their power management in the spotlight, especially with respect to their impact on environment and sustainable development. In this work, we investigate the capacity and limits of solutions based on an innovative approach for the power management of AI data centres, i.e., making part of the input power as dynamic as the power used for data-computing functions. The performance of passive and active devices are quantified and compared in terms of computational gain, energy efficiency, reduction of capital expenditure, and management costs by analysing power trends from multiple data platforms worldwide. This strategy, which identifies a paradigm shift in the AI data centre power management, has the potential to strongly improve the sustainability of AI hyperscalers, enhancing their footprint on environmental, financial, and societal fields.

Current approaches for strengthening LLM reasoning tend to introduce a training bias toward human-like reasoning trajectories. In step-wise preference optimization, in particular, dependence on human or higher-capacity model annotations for intermediate steps limits exploration of alternative, non-human-like reasoning paths and thus constrains achievable performance. Furthermore, through a small-scale pilot study, we observed that in approximately 75% of cases, the model's first erroneous step occurs after the lowest-confidence point. This suggests that guiding the model at its lowest-confidence point before an error provides more accurate supervision than locating the first explicit error. In this paper, we propose Confidence-Guided Reasoning Path Preference Optimization (CGPO), a method that leverages a confidence signal to identify points of maximal uncertainty in the model's reasoning process and applies self-generated, non-human-like reasoning-path guidance to mitigate trajectory drift. Our experiments span diverse models applied to both code and mathematical reasoning tasks. The results show that, with the same amount of training data, our method using data generated by a small model can achieve better performance in most cases compared with approaches using data generated by a strong model or human-annotated.

The rise of large language models (LLMs) has transformed healthcare by offering clinical guidance, yet their direct deployment to patients poses safety risks due to limited domain expertise. To mitigate this, we propose repositioning LLMs as clinical assistants that collaborate with experienced physicians rather than interacting with patients directly. We conduct a two-stage inspiration-feedback survey to identify real-world needs in clinical workflows. Guided by this, we construct DoctorFLAN, a large-scale Chinese medical dataset comprising 92,000 Q&A instances across 22 clinical tasks and 27 specialties. To evaluate model performance in doctor-facing applications, we introduce DoctorFLAN-test (550 single-turn Q&A items) and DotaBench (74 multi-turn conversations). Experimental results with over ten popular LLMs demonstrate that DoctorFLAN notably improves the performance of open-source LLMs in medical contexts, facilitating their alignment with physician workflows and complementing existing patient-oriented models. This work contributes a valuable resource and framework for advancing doctor-centered medical LLM development