Paraphrasing re-expresses meaning to enhance applications like text simplification, machine translation, and question-answering. Specific paraphrase types facilitate accurate semantic analysis and robust language models. However, existing paraphrase-type generation methods often misalign with human preferences due to reliance on automated metrics and limited human-annotated training data, obscuring crucial aspects of semantic fidelity and linguistic transformations. This study addresses this gap by leveraging a human-ranked paraphrase-type dataset and integrating Direct Preference Optimization (DPO) to align model outputs directly with human judgments. DPO-based training increases paraphrase-type generation accuracy by 3 percentage points over a supervised baseline and raises human preference ratings by 7 percentage points. A newly created human-annotated dataset supports more rigorous future evaluations. Additionally, a paraphrase-type detection model achieves F1 scores of 0.91 for addition/deletion, 0.78 for same polarity substitution, and 0.70 for punctuation changes. These findings demonstrate that preference data and DPO training produce more reliable, semantically accurate paraphrases, enabling downstream applications such as improved summarization and more robust question-answering. The PTD model surpasses automated metrics and provides a more reliable framework for evaluating paraphrase quality, advancing paraphrase-type research toward richer, user-aligned language generation and establishing a stronger foundation for future evaluations grounded in human-centric criteria.
We propose TaxaDiffusion, a taxonomy-informed training framework for diffusion models to generate fine-grained animal images with high morphological and identity accuracy. Unlike standard approaches that treat each species as an independent category, TaxaDiffusion incorporates domain knowledge that many species exhibit strong visual similarities, with distinctions often residing in subtle variations of shape, pattern, and color. To exploit these relationships, TaxaDiffusion progressively trains conditioned diffusion models across different taxonomic levels -- starting from broad classifications such as Class and Order, refining through Family and Genus, and ultimately distinguishing at the Species level. This hierarchical learning strategy first captures coarse-grained morphological traits shared by species with common ancestors, facilitating knowledge transfer before refining fine-grained differences for species-level distinction. As a result, TaxaDiffusion enables accurate generation even with limited training samples per species. Extensive experiments on three fine-grained animal datasets demonstrate that outperforms existing approaches, achieving superior fidelity in fine-grained animal image generation. Project page: https://amink8.github.io/TaxaDiffusion/
We propose Generative Low-rank language model with Semantic Search (GLoSS), a generative recommendation framework that combines large language models with dense retrieval for sequential recommendation. Unlike prior methods such as GPT4Rec, which rely on lexical matching via BM25, GLoSS uses semantic search to retrieve relevant items beyond lexical matching. For query generation, we employ 4-bit quantized LlaMA-3 models fine-tuned with low-rank adaptation (LoRA), enabling efficient training and inference on modest hardware. We evaluate GLoSS on three real-world Amazon review datasets: Beauty, Toys, and Sports, and find that it achieves state-of-the-art performance. Compared to traditional ID-based baselines, GLoSS improves Recall@5 by 33.3%, 52.8%, and 15.2%, and NDCG@5 by 30.0%, 42.6%, and 16.1%, respectively. It also outperforms LLM-based recommenders such as P5, GPT4Rec, LlamaRec and E4SRec with Recall@5 gains of 4.3%, 22.8%, and 29.5%. Additionally, user segment evaluations show that GLoSS performs particularly well for cold-start users in the Amazon Toys and Sports datasets, and benefits from longer user histories in Amazon Beauty dataset, demonstrating robustness across different levels of interaction lengths.
Task-oriented dialogue systems often face difficulties when user utterances seem semantically complete but lack necessary structural information for appropriate system action. This arises because users frequently do not fully understand their own needs, while systems require precise intent definitions. Current LLM-based agents cannot effectively distinguish between linguistically complete and contextually triggerable expressions, lacking frameworks for collaborative intent formation. We present STORM, a framework modeling asymmetric information dynamics through conversations between UserLLM (full internal access) and AgentLLM (observable behavior only). STORM produces annotated corpora capturing expression trajectories and latent cognitive transitions, enabling systematic analysis of collaborative understanding development. Our contributions include: (1) formalizing asymmetric information processing in dialogue systems; (2) modeling intent formation tracking collaborative understanding evolution; and (3) evaluation metrics measuring internal cognitive improvements alongside task performance. Experiments across four language models reveal that moderate uncertainty (40-60%) can outperform complete transparency in certain scenarios, with model-specific patterns suggesting reconsideration of optimal information completeness in human-AI collaboration. These findings contribute to understanding asymmetric reasoning dynamics and inform uncertainty-calibrated dialogue system design.
Table reasoning, encompassing tasks such as table question answering, fact verification, and text-to-SQL, requires precise understanding of structured tabular data, coupled with numerical computation and code manipulation for effective inference. Supervised fine-tuning (SFT) approaches have achieved notable success but often struggle with generalization and robustness due to biases inherent in imitative learning. We introduce Reasoning-Table, the first application of reinforcement learning (RL) to table reasoning, achieving state-of-the-art performance. Through rigorous data preprocessing, reward design, and tailored training strategies, our method leverages simple rule-based outcome rewards to outperform SFT across multiple benchmarks. Unified training across diverse tasks enables Reasoning-Table to emerge as a robust table reasoning large language model, surpassing larger proprietary models like Claude-3.7-Sonnet by 4.0% on table reasoning benchmarks. The approach also achieves excellent performance on text-to-SQL tasks, reaching 68.3% performance on the BIRD dev dataset with a 7B model. Further experiments demonstrate that Reasoning-Table enhances the model's generalization capabilities and robustness.
Efficiently compiling quantum operations remains a major bottleneck in scaling quantum computing. Today's state-of-the-art methods achieve low compilation error by combining search algorithms with gradient-based parameter optimization, but they incur long runtimes and require multiple calls to quantum hardware or expensive classical simulations, making their scaling prohibitive. Recently, machine-learning models have emerged as an alternative, though they are currently restricted to discrete gate sets. Here, we introduce a multimodal denoising diffusion model that simultaneously generates a circuit's structure and its continuous parameters for compiling a target unitary. It leverages two independent diffusion processes, one for discrete gate selection and one for parameter prediction. We benchmark the model over different experiments, analyzing the method's accuracy across varying qubit counts, circuit depths, and proportions of parameterized gates. Finally, by exploiting its rapid circuit generation, we create large datasets of circuits for particular operations and use these to extract valuable heuristics that can help us discover new insights into quantum circuit synthesis.
High-quality math datasets are crucial for advancing the reasoning abilities of large language models (LLMs). However, existing datasets often suffer from three key issues: outdated and insufficient challenging content, neglecting human-like reasoning, and limited reliability due to single-LLM generation. To address these, we introduce $\textbf{STORM-BORN}$, an ultra-challenging dataset of mathematical derivations sourced from cutting-edge academic papers, which includes dense human-like approximations and heuristic cues. To ensure the reliability and quality, we propose a novel human-in-the-loop, multi-agent data generation framework, integrating reasoning-dense filters, multi-agent collaboration, and human mathematicians' evaluations. We curated a set of 2,000 synthetic samples and deliberately selected the 100 most difficult problems. Even most advanced models like GPT-o1 solved fewer than $5\%$ of them. Fine-tuning on STORM-BORN boosts accuracy by $7.84\%$ (LLaMA3-8B) and $9.12\%$ (Qwen2.5-7B). As AI approaches mathematician-level reasoning, STORM-BORN provides both a high-difficulty benchmark and a human-like reasoning training resource. Our code and dataset are publicly available at https://github.com/lwhere/STORM-BORN.
Stage lighting plays an essential role in live music performances, influencing the engaging experience of both musicians and audiences. Given the high costs associated with hiring or training professional lighting engineers, Automatic Stage Lighting Control (ASLC) has gained increasing attention. However, most existing approaches only classify music into limited categories and map them to predefined light patterns, resulting in formulaic and monotonous outcomes that lack rationality. To address this issue, this paper presents an end-to-end solution that directly learns from experienced lighting engineers -- Skip-BART. To the best of our knowledge, this is the first work to conceptualize ASLC as a generative task rather than merely a classification problem. Our method modifies the BART model to take audio music as input and produce light hue and value (intensity) as output, incorporating a novel skip connection mechanism to enhance the relationship between music and light within the frame grid.We validate our method through both quantitative analysis and an human evaluation, demonstrating that Skip-BART outperforms conventional rule-based methods across all evaluation metrics and shows only a limited gap compared to real lighting engineers.Specifically, our method yields a p-value of 0.72 in a statistical comparison based on human evaluations with human lighting engineers, suggesting that the proposed approach closely matches human lighting engineering performance. To support further research, we have made our self-collected dataset, code, and trained model parameters available at https://github.com/RS2002/Skip-BART .
The identification of drug-target interactions (DTI) is crucial for drug discovery and repositioning, as it reveals potential uses of existing drugs, aiding in the acceleration of the drug development process and reducing associated costs. Despite the similarity information in DTI is important, most models are limited to mining direct similarity information within homogeneous graphs, overlooking the potential yet rich similarity information in heterogeneous graphs. Inspired by real-world social interaction behaviors, we propose SOC-DGL, which comprises two specialized modules: the Affinity-Driven Graph Learning (ADGL) module and the Equilibrium-Driven Graph Learning (EDGL) module. The ADGL module adopts a comprehensive social interaction strategy, leveraging an affinity-enhanced global drug-target graph to learn both global DTI and the individual similarity information of drugs and targets. In contrast, the EDGL module employs a higher-order social interaction strategy, amplifying the influence of even-hop neighbors through an even-polynomial graph filter grounded in balance theory, enabling the indirect mining of higher-order homogeneous information. This dual approach enables SOC-DGL to effectively and comprehensively capture similarity information across diverse interaction scales within the affinity matrices and drug-target association matrices, significantly enhancing the model's generalization capability and predictive accuracy in DTI tasks. To address the issue of imbalance in drug-target interaction datasets, this paper proposes an adjustable imbalance loss function that mitigates the impact of sample imbalance by adjusting the weight of negative samples and a parameter. Extensive experiments on four benchmark datasets demonstrate significant accuracy improvements achieved by SOC-DGL, particularly in scenarios involving data imbalance and unseen drugs or targets.
The recent progress of large language model agents has opened new possibilities for automating tasks through graphical user interfaces (GUIs), especially in mobile environments where intelligent interaction can greatly enhance usability. However, practical deployment of such agents remains constrained by several key challenges. Existing training data is often noisy and lack semantic diversity, which hinders the learning of precise grounding and planning. Models trained purely by imitation tend to overfit to seen interface patterns and fail to generalize in unfamiliar scenarios. Moreover, most prior work focuses on English interfaces while overlooks the growing diversity of non-English applications such as those in the Chinese mobile ecosystem. In this work, we present AgentCPM-GUI, an 8B-parameter GUI agent built for robust and efficient on-device GUI interaction. Our training pipeline includes grounding-aware pre-training to enhance perception, supervised fine-tuning on high-quality Chinese and English trajectories to imitate human-like actions, and reinforcement fine-tuning with GRPO to improve reasoning capability. We also introduce a compact action space that reduces output length and supports low-latency execution on mobile devices. AgentCPM-GUI achieves state-of-the-art performance on five public benchmarks and a new Chinese GUI benchmark called CAGUI, reaching $96.9\%$ Type-Match and $91.3\%$ Exact-Match. To facilitate reproducibility and further research, we publicly release all code, model checkpoint, and evaluation data.
Forced alignment is a common tool to align audio with orthographic and phonetic transcriptions. Most forced alignment tools provide only a single estimate of a boundary. The present project introduces a method of deriving confidence intervals for these boundaries using a neural network ensemble technique. Ten different segment classifier neural networks were previously trained, and the alignment process is repeated with each model. The alignment ensemble is then used to place the boundary at the median of the boundaries in the ensemble, and 97.85% confidence intervals are constructed using order statistics. On the Buckeye and TIMIT corpora, the ensemble boundaries show a slight improvement over using just a single model. The confidence intervals are incorporated into Praat TextGrids using a point tier, and they are also output as a table for researchers to analyze separately as diagnostics or to incorporate uncertainty into their analyses.
We introduce the $\underline{Ko}rean \underline{G}rammar \underline{E}valuation Bench\underline{M}ark (KoGEM)$, designed to assess the linguistic competence of LLMs and humans in Korean. KoGEM consists of 1.5k multiple-choice QA pairs covering five main categories and 16 subcategories. The zero-shot evaluation of 27 LLMs of various sizes and types reveals that while LLMs perform remarkably well on straightforward tasks requiring primarily definitional knowledge, they struggle with tasks that demand the integration of real-world experiential knowledge, such as phonological rules and pronunciation. Furthermore, our in-depth analysis suggests that incorporating such experiential knowledge could enhance the linguistic competence of LLMs. With KoGEM, we not only highlight the limitations of current LLMs in linguistic competence but also uncover hidden facets of LLMs in linguistic competence, paving the way for enhancing comprehensive language understanding. Our code and dataset are available at: https://github.com/SungHo3268/KoGEM.
Language is often used strategically, particularly in high-stakes, adversarial settings, yet most work on pragmatics and LLMs centers on cooperativity. This leaves a gap in systematic understanding of non-cooperative discourse. To address this, we introduce CoBRA (Cooperation-Breach Response Assessment), along with three interpretable metrics -- Benefit at Turn (BaT), Penalty at Turn (PaT), and Normalized Relative Benefit at Turn (NRBaT) -- to quantify the perceived strategic effects of discourse moves. We also present CHARM, an annotated dataset of real courtroom cross-examinations, to demonstrate the framework's effectiveness. Using these tools, we evaluate a range of LLMs and show that LLMs generally exhibit limited pragmatic understanding of strategic language. While model size shows an increase in performance on our metrics, reasoning ability does not help and largely hurts, introducing overcomplication and internal confusion.
Mispronunciation detection (MD) models are the cornerstones of many language learning applications. Unfortunately, most systems are built for English and other major languages, while low-resourced language varieties, such as Finland Swedish (FS), lack such tools. In this paper, we introduce our MD model for FS, trained on 89 hours of first language (L1) speakers' spontaneous speech and tested on 33 minutes of L2 transcribed read-aloud speech. We trained a multilingual wav2vec 2.0 model with entropy regularization, followed by temperature scaling and top-k normalization after the inference to better adapt it for MD. The main novelty of our method lies in its simplicity, requiring minimal L2 data. The process is also language-independent, making it suitable for other low-resource languages. Our proposed algorithm allows us to balance Recall (43.2%) and Precision (29.8%), compared with the baseline model's Recall (77.5%) and Precision (17.6%).
The ultimate goal of verification is to guarantee the safety of deployed neural networks. Here, we claim that all the state-of-the-art verifiers we are aware of fail to reach this goal. Our key insight is that theoretical soundness (bounding the full-precision output while computing with floating point) does not imply practical soundness (bounding the floating point output in a potentially stochastic environment). We prove this observation for the approaches that are currently used to achieve provable theoretical soundness, such as interval analysis and its variants. We also argue that achieving practical soundness is significantly harder computationally. We support our claims empirically as well by evaluating several well-known verification methods. To mislead the verifiers, we create adversarial networks that detect and exploit features of the deployment environment, such as the order and precision of floating point operations. We demonstrate that all the tested verifiers are vulnerable to our new deployment-specific attacks, which proves that they are not practically sound.
Large language models (LLMs) have demonstrated exceptional performance across a wide range of natural language tasks. However, selecting the optimal LLM to respond to a user query often necessitates a delicate balance between performance and cost. While powerful models deliver better results, they come at a high cost, whereas smaller models are more cost-effective but less capable. To address this trade-off, we propose IRT-Router, a multi-LLM routing framework that efficiently routes user queries to the most suitable LLM. Inspired by Item Response Theory (IRT), a psychological measurement methodology, IRT-Router explicitly models the relationship between LLM capabilities and user query attributes. This not only enables accurate prediction of response performance but also provides interpretable insights, such as LLM abilities and query difficulty. Additionally, we design an online query warm-up technique based on semantic similarity, further enhancing the online generalization capability of IRT-Router. Extensive experiments on 20 LLMs and 12 datasets demonstrate that IRT-Router outperforms most baseline methods in terms of effectiveness and interpretability. Its superior performance in cold-start scenarios further confirms the reliability and practicality of IRT-Router in real-world applications. Code is available at https://github.com/Mercidaiha/IRT-Router.
This paper presents LEMONADE, a large-scale conflict event dataset comprising 39,786 events across 20 languages and 171 countries, with extensive coverage of region-specific entities. LEMONADE is based on a partially reannotated subset of the Armed Conflict Location & Event Data (ACLED), which has documented global conflict events for over a decade. To address the challenge of aggregating multilingual sources for global event analysis, we introduce abstractive event extraction (AEE) and its subtask, abstractive entity linking (AEL). Unlike conventional span-based event extraction, our approach detects event arguments and entities through holistic document understanding and normalizes them across the multilingual dataset. We evaluate various large language models (LLMs) on these tasks, adapt existing zero-shot event extraction systems, and benchmark supervised models. Additionally, we introduce ZEST, a novel zero-shot retrieval-based system for AEL. Our best zero-shot system achieves an end-to-end F1 score of 58.3%, with LLMs outperforming specialized event extraction models such as GoLLIE. For entity linking, ZEST achieves an F1 score of 45.7%, significantly surpassing OneNet, a state-of-the-art zero-shot baseline that achieves only 23.7%. However, these zero-shot results lag behind the best supervised systems by 20.1% and 37.0% in the end-to-end and AEL tasks, respectively, highlighting the need for further research.
The accelerating development of general medical artificial intelligence (GMAI), powered by multimodal large language models (MLLMs), offers transformative potential for addressing persistent healthcare challenges, including workforce deficits and escalating costs. The parallel development of systematic evaluation benchmarks emerges as a critical imperative to enable performance assessment and provide technological guidance. Meanwhile, as an invaluable knowledge source, the potential of medical textbooks for benchmark development remains underexploited. Here, we present MedBookVQA, a systematic and comprehensive multimodal benchmark derived from open-access medical textbooks. To curate this benchmark, we propose a standardized pipeline for automated extraction of medical figures while contextually aligning them with corresponding medical narratives. Based on this curated data, we generate 5,000 clinically relevant questions spanning modality recognition, disease classification, anatomical identification, symptom diagnosis, and surgical procedures. A multi-tier annotation system categorizes queries through hierarchical taxonomies encompassing medical imaging modalities (42 categories), body anatomies (125 structures), and clinical specialties (31 departments), enabling nuanced analysis across medical subdomains. We evaluate a wide array of MLLMs, including proprietary, open-sourced, medical, and reasoning models, revealing significant performance disparities across task types and model categories. Our findings highlight critical capability gaps in current GMAI systems while establishing textbook-derived multimodal benchmarks as essential evaluation tools. MedBookVQA establishes textbook-derived benchmarking as a critical paradigm for advancing clinical AI, exposing limitations in GMAI systems while providing anatomically structured performance metrics across specialties.
Large language models (LLMs) are trained on extensive datasets that encapsulate substantial world knowledge. However, their outputs often include confidently stated inaccuracies. Earlier works suggest that LLMs encode truthfulness as a distinct linear feature, termed the "truth direction", which can classify truthfulness reliably. We address several open questions about the truth direction: (i) whether LLMs universally exhibit consistent truth directions; (ii) whether sophisticated probing techniques are necessary to identify truth directions; and (iii) how the truth direction generalizes across diverse contexts. Our findings reveal that not all LLMs exhibit consistent truth directions, with stronger representations observed in more capable models, particularly in the context of logical negation. Additionally, we demonstrate that truthfulness probes trained on declarative atomic statements can generalize effectively to logical transformations, question-answering tasks, in-context learning, and external knowledge sources. Finally, we explore the practical application of truthfulness probes in selective question-answering, illustrating their potential to improve user trust in LLM outputs. These results advance our understanding of truth directions and provide new insights into the internal representations of LLM beliefs. Our code is public at https://github.com/colored-dye/truthfulness_probe_generalization
Retrieval-Augmented Generation (RAG) enhances recency and factuality in answers. However, existing evaluations rarely test how well these systems cope with real-world noise, conflicting between internal and external retrieved contexts, or fast-changing facts. We introduce Retrieval-Aware Robustness Evaluation (RARE), a unified framework and large-scale benchmark that jointly stress-tests query and document perturbations over dynamic, time-sensitive corpora. One of the central features of RARE is a knowledge-graph-driven synthesis pipeline (RARE-Get) that automatically extracts single and multi-hop relations from the customized corpus and generates multi-level question sets without manual intervention. Leveraging this pipeline, we construct a dataset (RARE-Set) spanning 400 expert-level time-sensitive finance, economics, and policy documents and 48,322 questions whose distribution evolves as the underlying sources change. To quantify resilience, we formalize retrieval-conditioned robustness metrics (RARE-Met) that capture a model's ability to remain correct or recover when queries, documents, or real-world retrieval results are systematically altered. Our results show that RAG systems exhibit surprising vulnerability to perturbations, with document robustness consistently being the weakest point regardless of generator size or architecture. RAG systems consistently show lower robustness on multi-hop queries than single-hop queries across all domains.