论文检索

Huaming Du, Lei Yuan, Cancan Feng, Guisong Liu, Gang Kou, Carl Yang

Large Language Models (LLMs) have achieved impressive success across a range of natural language processing tasks. However, they still underperform in text classification tasks compared to fine-tuned small models. This can be linked to complexities in addressing context-dependent expressions and complex linguistic phenomena. In contrast, fine-tuned small models typically achieve high prediction accuracy but often lack explanations for predictions. Existing explanation methods that generate keywords may be less effective due to missing critical contextual information. To mitigate these challenges, we propose a novel method termed Dialectical Explanation Training (**DET**). This method introduces a new prompting strategy, Dialectical Prompting, and integrates it with Explanation-Guided Training. Dialectical Prompting uses LLMs with our designed dialectical prompt to generate explanations for possible labels. These explanations handle context-dependent expressions and complex linguistic phenomena by considering multiple perspectives and providing rich, contextually relevant information. Explanation-Guided Training employs these explanations as features for training a small model, which combines the advantages of dialectical explanations and the predictive power of fine-tuned models to improve overall accuracy and interpretability. In addition, we incorporate the theory of Evidential Deep Learning, which further enhances the model’s classification performance and quantify the uncertainty of its predictions. Extensive experiments on multiple datasets from diverse domains have demonstrated that our proposed model significantly improves accuracy and explanation quality over state-of the-art methods in text classification.

Shohei Higashiyama, Masao Utiyama

Lexical normalization research has sought to tackle the challenge of processing informal expressions in user-generated text, yet the absence of comprehensive evaluations leaves it unclear which methods excel across multiple perspectives. Focusing on unsegmented languages, we make three key contributions: (1) creating a large-scale, multi-domain Japanese normalization dataset, (2) developing normalization methods based on state-of-the-art pre-trained models, and (3) conducting experiments across multiple evaluation perspectives. Our experiments show that both encoder-only and decoder-only approaches achieve promising results in both accuracy and efficiency.

Yaozu Wu, Dongyuan Li, Yankai Chen, Renhe Jiang, Henry Peng Zou, Wei-Chieh Huang, Yangning Li, Liancheng Fang, Zhen Wang, Philip S. Yu

Autonomous Driving Systems (ADSs) are revolutionizing transportation by reducing human intervention, improving operational efficiency, and enhancing safety. Large Language Models (LLMs), known for their exceptional planning and reasoning capabilities, have been integrated into ADSs to assist with driving decision-making. However, LLM-based single-agent ADSs face three major challenges: limited perception, insufficient collaboration, and high computational demands. To address these issues, recent advancements in LLM-based multi-agent ADSs have focused on improving inter-agent communication and cooperation. This paper provides a frontier survey of LLM-based multi-agent ADSs. We begin with a background introduction to related concepts, followed by a categorization of existing LLM-based approaches based on different agent interaction modes. We then discuss agent-human interactions in scenarios where LLM-based agents engage with humans. Finally, we summarize key applications, datasets, and challenges in this field to support future research (https://github.com/Yaozuwu/LLM-based_Multi-agent_ADS).

Sungeun Hahm, Heejin Kim, Gyuseong Lee, Hyunji M. Park, Jaejin Lee

To ensure a balance between open access to justice and personal data protection, the South Korean judiciary mandates the de-identification of court judgments before they can be publicly disclosed. However, the current de-identification process is inadequate for handling court judgments at scale while adhering to strict legal requirements. Additionally, the legal definitions and categorizations of personal identifiers are vague and not well-suited for technical solutions. To tackle these challenges, we propose a de-identification framework called Thunder-DeID, which aligns with relevant laws and practices. Specifically, we (i) construct and release the first Korean legal dataset containing annotated judgments along with corresponding lists of entity mentions, (ii) introduce a systematic categorization of Personally Identifiable Information (PII), and (iii) develop an end-to-end deep neural network (DNN)-based de-identification pipeline. Our experimental results demonstrate that our model achieves state-of-the-art performance in the de-identification of court judgments.

Shin-nosuke Ishikawa, Masato Todo, Taiki Ogihara, Hirotsugu Ohba

We present a tendency of large language models (LLMs) to generate absurd patterns despite their clear inappropriateness in a simple task of identifying regularities in number series. Several approaches have been proposed to apply LLMs to complex real-world tasks, such as providing knowledge through retrieval-augmented generation and executing multi-step tasks using AI agent frameworks. However, these approaches rely on the logical consistency and self-coherence of LLMs, making it crucial to evaluate these aspects and consider potential countermeasures. To identify cases where LLMs fail to maintain logical consistency, we conducted an experiment in which LLMs were asked to explain the patterns in various integer sequences, ranging from arithmetic sequences to randomly generated integer series. While the models successfully identified correct patterns in arithmetic and geometric sequences, they frequently over-recognized patterns that were inconsistent with the given numbers when analyzing randomly generated series. This issue was observed even in multi-step reasoning models, including OpenAI o3, o4-mini, and Google Gemini 2.5 Flash Preview Thinking. This tendency to perceive non-existent patterns can be interpreted as the AI model equivalent of Idola Tribus and highlights potential limitations in their capability for applied tasks requiring logical reasoning, even when employing chain-of-thought reasoning mechanisms.

Yang Cao, Sikun Yang, Yujiu Yang, Lianyong Qi, Ming Liu

Two-step approaches combining pre-trained large language model embeddings and anomaly detectors demonstrate strong performance in text anomaly detection by leveraging rich semantic representations. However, high-dimensional dense embeddings extracted by large language models pose challenges due to substantial memory requirements and high computation time. To address this challenge, we introduce the Simplified Isolation Kernel (SIK), which maps high-dimensional dense embeddings to lower-dimensional sparse representations while preserving crucial anomaly characteristics. SIK has linear-time complexity and significantly reduces space complexity through its innovative boundary-focused feature mapping.Experiments across 7 datasets demonstrate that SIK achieves better detection performance than 11 SOTA anomaly detection algorithms while maintaining computational efficiency and low memory cost. All code and demonstrations are available at https://github.com/charles-cao/SIK.

Mohammad R. Rezaei, Reza Saadati Fard, Jayson L. Parker, Rahul G. Krishnan, Milad Lankarany

Large Language Models (LLMs) have greatly advanced medical Question Answering (QA) by leveraging vast clinical data and medical literature. However, the rapid evolution of medical knowledge and the labor-intensive process of manually updating domain-specific resources can undermine the reliability of these systems. We address this challenge with Agentic Medical Graph-RAG (AMG-RAG), a comprehensive framework that automates the construction and continuous updating of Medical Knowledge Graph (MKG), integrates reasoning, and retrieves current external evidence from the MKG for medical QA.Evaluations on the MEDQA and MEDMCQA benchmarks demonstrate the effectiveness of AMG-RAG, achieving an F1 score of 74.1% on MEDQA and an accuracy of 66.34% on MEDMCQA—surpassing both comparable models and those 10 to 100 times larger. By dynamically linking new findings and complex medical concepts, AMG-RAG not only boosts accuracy but also enhances interpretability for medical queries, which has a critical impact on delivering up-to-date, trustworthy medical insights.

Zi Yu, Shaoxiang Wang, Guozheng Li, Yu Zhang, Chi Harold Liu

Open-ended event forecasting (OEEF) seeks to predict future events from a given context without being restricted to a predefined scope or format. It plays a crucial role in domains such as risk management and financial decision making. Although large language models show potential for OEEF, existing approaches and datasets often overlook the complex relationships among events, and current research lacks comprehensive evaluation methods. To address these limitations, we propose ForestCast, a prediction pipeline that extracts forecast-relevant events from news data, organizes them into a story tree, and predicts subsequent events along each path. The pipeline comprises four stages: (1) grouping news into event nodes, (2) constructing a news story tree, (3) mining the semantic structure of the tree, and (4) predicting the next event node and evaluating prediction quality. To support this pipeline, we construct NewsForest, a dataset of 12,406 event chains, each representing a chronologically and logically linked sequence of news events. In addition, we introduce a comprehensive evaluation framework that measures both the accuracy and the quality of prediction. Experimental results demonstrate that ForestCast improves the ability of LLMs to forecast events in news data.

Cheng Yan, Feng Zhao, Ruilin Zhao, Hong Zhang

Knowledge graphs are dynamic structures that continuously evolve as new entities emerge, often accompanied by only a handful of associated triples. Current knowledge graph reasoning methods struggle in these few-shot scenarios due to their reliance on extensive structural information.To address this limitation, we introduce ENGRAM, a novel approach that enables inductive reasoning on few-shot KGs by innovatively enriching the semantics from both textual and structural perspectives. Our key innovation lies in designing a task-aware language model that activates the language model’s in-context learning ability for structured KG tasks, effectively bridging the gap between unstructured natural language and structured tasks. Unlike prior methods that inefficiently employ classification over exhaustive candidate sets, we recast knowledge graph reasoning from a generative perspective, allowing for direct computation of inference results without iterative enumeration. Additionally, we propose a distant neighborhood awareness strategy to enrich the sparse structural features of few-shot entities.Our experimental findings indicate that our method not only achieves state-of-the-art performance in few-shot scenarios. The tunable parameters of our model are approximately 1% of those in previous language model-based methods, and the inference time has been reduced to 1/10 of that required by previous methods.

Zhangming Li, Qinghao Hu, Yiqun Chen, Peisong Wang, Yifan Zhang, Jian Cheng

As the parameter size of language models becomes extremely large, fine-tuning them with limited resources has become a challenging task. Latest advancements in parameter-efficient fine-tuning (PEFT) techniques allow for adjustments to only a minor fraction of the parameters of these LLMs. Yet, most of PEFT methods may suffer from the following limitations: (1) As the rank decreases sharply, PEFT methods like LoRA and Adapter tuning will exhibit significant performance degradation in downstream tasks. (2) An accuracy gap between these methods and full fine-tuning (Full-FT) still exists. To tackle these problems, we propose a Low-Rank Direct Attention Adaptation (LoRaDA) method for efficient LLM fine-tuning. Specifically, we introduce a novel Low-rank Multi-head Attention Map Module (LMAM), which can bring negative attention to self-attention modules and learn low-rank attention weights directly, capturing the characteristics of downstream tasks. Furthermore, LMAM can serve as a plug-in to existing methods, such as LoRA and Adapter, providing state-of-the-art performance even with extreme low rank setting.Extensive experiments on various downstream tasks demonstrate the superior performance of our LoRaDA method. Specifically, LoRaDA even outperforms the full fine-tuning method by up to 2.1% on GLUE benchmark. As a plug-in, LMAM boosts the accuracy of LoRA by up to 27.7% with LLaMA-7B on Commonsense Reasoning benchmark.

Shang Ma, Tianyi Ma, Jiahao Liu, Wei Song, Zhenkai Liang, Xusheng Xiao, Yanfang Ye

Over the years, online scams have grown dramatically,with nearly 50% of global consumersencountering scam attempts each week.These scams cause not only significant financiallosses to individuals and businesses, butalso lasting psychological trauma, largely dueto scammers’ strategic employment of psychologicaltechniques (PTs) to manipulate victims.Meanwhile, scammers continually evolve theirtactics by leveraging advances in Large LanguageModels (LLMs) to generate diverse scamvariants that easily bypass existing defenses.To address this pressing problem, we introducePsyScam, a benchmark designed to systematicallycapture the PTs employed in real-worldscam reports, and investigate how LLMs canbe utilized to generate variants of scams basedon the PTs and the contexts provided by thesescams. Specifically, we collect a wide range ofscam reports and ground its annotations of employedPTs in well-established cognitive andpsychological theories. We further demonstrateLLMs’ capabilities in generating through twodownstream tasks: scam completion, and scamaugmentation. Experimental results show thatPsyScam presents significant challenges toexisting models in both detecting and generatingscam content based on the PTs used byreal-world scammers. Our code and dataset areavailable.

Thomas Berkane, Marie-Laure Charpignon, Maimuna S. Majumder

Researchers across many fields rely on web data to gain new insights and validate methods. However, assembling accurate and comprehensive datasets typically requires manual review of numerous web pages to identify and record only those data points relevant to specific research objectives. The vast and scattered nature of online information makes this process time-consuming and prone to human error. To address these challenges, we present a human-in-the-loop framework that automates web-scale data collection end-to-end using large language models (LLMs). Given a textual description of a target dataset, our framework (1) automatically formulates search engine queries, (2) navigates the web to identify relevant web pages, (3) extracts the data points of interest, and (4) performs quality control to produce a structured, research-ready dataset. Importantly, users remain in the loop throughout the process and can inspect and adjust the framework’s decisions to ensure alignment with their needs. We introduce techniques to mitigate both search engine bias and LLM hallucinations during data extraction. Experiments across three diverse data collection tasks show that our framework greatly outperforms existing methods, while a user evaluation demonstrates its practical utility. We release our code at https://github.com/tberkane/web-data-collection to help other researchers create custom datasets more efficiently.

Xinyi Chen, Yifei Yuan, Jiaang Li, Serge Belongie, Maarten de Rijke, Anders Søgaard

Language models are often said to face a symbol grounding problem. While some have argued the problem can be solved without resort to other modalities, many have speculated that grounded learning is more efficient. We explore this question in Othello, a simplified, rule-based world that offers a controlled and interpretable testbed for studying world understanding. Building on prior work, we introduce VISOTHELLO, a multi-modal model trained jointly on move sequences and board images. Using the Othello rule understanding task, we examine whether multi-modal learning provides advantages over text-only approaches. We further evaluate robustness under semantically irrelevant perturbations and analyze the consistency of cross-modal alignment. Our results suggest that multi-modal training not only improves performance and robustness but also promotes convergence toward shared internal representations across different model architectures.

Yunhai Hu, Yilun Zhao, Chen Zhao, Arman Cohan

We introduce MCTS-RAG, a novel approach that enhances the reasoning capabilities of small language models on knowledge-intensive tasks by leveraging retrieval-augmented generation (RAG) to provide relevant context and Monte Carlo Tree Search (MCTS) to refine reasoning paths. MCTS-RAG dynamically integrates retrieval and reasoning through an iterative decision-making process. Unlike standard RAG methods, which typically retrieve information independently from reasoning and thus integrate knowledge suboptimally, or conventional MCTS reasoning, which depends solely on internal model knowledge without external facts, MCTS-RAG combines structured reasoning with adaptive retrieval. This integrated approach enhances decision-making, reduces hallucinations, and ensures improved factual accuracy and response consistency. The experimental results on multiple reasoning and knowledge-intensive datasets datasets (ComplexWebQA, GPQA, and FoolMeTwice) show that our method enables small-scale LMs to achieve performance comparable to frontier LLMs like GPT-4o by effectively scaling inference-time compute, setting a new standard for reasoning in small-scale models.

Bohui Zhang, Yuan He, Lydia Pintscher, Albert Meroño-Peñuela, Elena Simperl

Schemas play a vital role in ensuring data quality and supporting usability in the Semantic Web and natural language processing. Traditionally, their creation demands substantial involvement from knowledge engineers and domain experts. Leveraging the impressive capabilities of large language models (LLMs) in tasks like ontology engineering, we explore schema generation using LLMs. To bridge the resource gap, we introduce two datasets: YAGO Schema and Wikidata EntitySchema, along with novel evaluation metrics. The LLM-based pipelines utilize local and global information from knowledge graphs (KGs) to generate schemas in Shape Expressions (ShEx). Experiments demonstrate LLMs’ strong potential in producing high-quality ShEx schemas, paving the way for scalable, automated schema generation for large KGs. Furthermore, our benchmark introduces a new challenge for structured generation, pushing the limits of LLMs on syntactically rich formalisms.

Meinan Liu, Yunfang Dong, Xixian Liao, Bonnie Webber

Previous work has shown that simple mask-filling can provide useful information about the discourse informativeness of syntactic structures. Dong et al. (2024) first adopted this approach to investigating preposing constructions. The problem with single token mask fillers was that they were, by and large, ambiguous. We address the issue by adapting the approach of Kalinsky et al. (2023) to support the prediction of multi-token connectives in masked positions. Our first experiment demonstrates that this multi-token mask-filling approach substantially outperforms the previously considered single-token approach in recognizing implicit discourse relations. Our second experiment corroborates previous findings, providing additional empirical support for the role of preposed syntactic constituents in signaling discourse coherence. Overall, our study extends existing mask-filling methods to a new discourse-level task and reinforces the linguistic hypothesis concerning the discourse informativeness of preposed structures.

Hangliang Ren

Latent-recurrent language models solve tasks by iteratively refining hidden states rather than emitting chain-of-thought tokens, yet the opacity of those hidden trajectories hinders credit assignment and limits mathematical reasoning accuracy. We propose Latent-State Supervised Reinforcement Learning (LSRL), a process-supervised variant of Guided Reward Policy Optimization (GRPO) that delivers dense rewards at every latent step. We decode each recurrent depth of a 3.5-billion-parameter Huginn model and score the partial solutions with a GPT-4.1-nano grader aligned to final-answer correctness. Using LoRA adapters, we update the policy on a single NVIDIA L40S GPU with only 500 GSM-8K training problems. Relative to the depth-8 supervised Huginn baseline, LSRL improves absolute accuracy by +4.27 points on GSM-8K and +2.06 points on MathQA. These results demonstrate that rewarding latent steps provides an efficient route to stronger mathematical reasoning in latent-recurrent language models.

Yiwei Li, Jiayi Shi, Shaoxiong Feng, Peiwen Yuan, Xinglin Wang, Yueqi Zhang, Ji Zhang, Chuyi Tan, Boyuan Pan, Yao Hu 等

We propose a novel speculative decoding method tailored for multi-sample reasoning scenarios, such as self-consistency and Best-of-N sampling. Our method exploits the intrinsic consensus of parallel generation paths to synthesize high-quality draft tokens without requiring auxiliary models or external databases. By dynamically analyzing structural patterns across parallel reasoning paths through a probabilistic aggregation mechanism, it identifies consensus token sequences that align with the decoding distribution. Evaluations on mathematical reasoning and code generation benchmarks demonstrate a substantial improvement in draft acceptance rates over baselines, while reducing the latency in draft token construction. This work establishes a paradigm shift for efficient multi-sample inference, enabling seamless integration of speculative decoding with sampling-based reasoning techniques.

Sikun Guo, Amir Hassan Shariatmadari, Peng Wang, Albert Huang, Aidong Zhang

Advancements in Large Language Models (LLMs) have opened new opportunities for scientific discovery by assisting researchers in generating novel hypotheses and ideas. In this process, a major challenge is how to optimally and efficiently utilize LLMs’ parametric knowledge obtained from their pretraining process. Inspired by Generative Adversarial Networks (GANs), we propose inference time adversarial learning (termed InfAL), implemented through multi-LLM-agent interactions, to enhance research ideation. This approach optimizes the utilization of LLMs’ parametric knowledge without requiring additional model training, making adversarial learning efficient and context-driven. To evaluate the quality of generated ideas, we propose a relative quality ranking metric as a scalable alternative to human evaluation. Our results show that InfAL significantly improves idea generation, with GPT-4o achieving a 21% increase in novelty and a 322% increase in feasibility, demonstrating its transformative potential for driving innovation in scientific research.

Yuchun Fan, Yilin Wang, Yongyu Mu, Lei Huang, Bei Li, Xiaocheng Feng, Tong Xiao, JingBo Zhu

Large vision-language models (LVLMs) have demonstrated exceptional capabilities in understanding visual information with human languages but also exhibit an imbalance in multilingual capabilities. In this work, we delve into the multilingual working pattern of LVLMs and identify a salient correlation between the multilingual understanding ability of LVLMs and language-specific neuron activations in shallow layers. Building on this insight, we introduce PLAST, a training recipe that achieves efficient multilingual enhancement for LVLMs by Precise LAnguage Specific layers fine-Tuning. PLAST first identifies layers involved in multilingual understanding by monitoring language-specific neuron activations. These layers are then precisely fine-tuned with question-translation pairs to achieve multilingual alignment. Our empirical results on MMBench and MMMB demonstrate that PLAST effectively improves the multilingual capabilities of LVLMs and achieves significant efficiency with only 14% of the parameters tuned. Further analysis reveals that PLAST facilitates the language-specific visual information engagement in shallow layers.