Multilingual models often treat language diversity as a problem of data imbalance, overlooking structural variation. We introduce the Morphological Index (MoI), a typologically grounded metric that quantifies how strongly a language relies on surface morphology for noun classification. Building on MoI, we propose MoI-MoE, a Mixture of Experts model that routes inputs based on morphological structure. Evaluated on 10 Bantu language, a large, morphologically rich and underrepresented family, MoI-MoE outperforms strong baselines, improving Swahili accuracy by 14 points on noun class recognition while maintaining performance on morphology-rich languages like Zulu. These findings highlight typological structure as a practical and interpretable signal for multilingual model adaptation.
论文检索
输入标题、作者或关键词，从 7,876 篇学术成果中精准定位
The scientific research paradigm is undergoing a profound transformation owing to the development of Artificial Intelligence (AI). Recent works demonstrate that various AI-assisted research methods can largely improve research efficiency by improving data analysis, accelerating computation, and fostering novel idea generation. To further move towards the ultimate goal (i.e., automatic scientific research), in this paper, we introduce Dolphin, a closed-loop LLM-driven framework to enhance the automation level of scientific research. Dolphin first generates novel ideas based on feedback from previous experiments and relevant papers ranked by the topic and task attributes. Then, the generated ideas can be implemented using a code template refined and debugged with the designed exception-traceback-guided local code structure. Finally, Dolphin automatically analyzes the results of each idea and feeds the results back to the next round of idea generation. Experiments are conducted on the benchmark datasets of different topics and a subset of MLE-bench. Results show that Dolphin can continuously improve the performance of the input topic in a loop. We highlight that Dolphin can automatically propose methods that are comparable to the state-of-the-art in some tasks such as 3D point classification.
Often when we interact with a first-person account of events, we consider whether or not the narrator, the primary speaker of the text, is reliable. In this paper, we propose using computational methods to identify unreliable narrators, i.e. those who unintentionally misrepresent information. Borrowing literary theory from narratology to define different types of unreliable narrators based on a variety of textual phenomena, we present TUNa, a human-annotated dataset of narratives from multiple domains, including blog posts, subreddit posts, hotel reviews, and works of literature. We define classification tasks for intra-narrational, inter-narrational, and inter-textual unreliabilities and analyze the performance of popular open-weight and proprietary LLMs for each. We propose learning from literature to perform unreliable narrator classification on real-world text data. To this end, we experiment with few-shot, fine-tuning, and curriculum learning settings. Our results show that this task is very challenging, and there is potential for using LLMs to identify unreliable narrators. We release our expert-annotated dataset and code at https://github.com/adbrei/unreliable-narrators and invite future research in this area.
We present a new adaptation method MaCP, Minimal yet Mighty adaptive Cosine Projection, that achieves exceptional performance while requiring minimal parameters and memory for fine-tuning large foundation models.Its general idea is to exploit the superior energy compaction and decorrelation properties of cosine projection to improve both model efficiency and accuracy.Specifically, it projects the weight change from the low-rank adaptation into the discrete cosine space.Then, the weight change is partitioned over different levels of the discrete cosine spectrum, and each partition’s most critical frequency components are selected.Extensive experiments demonstrate the effectiveness of MaCP across a wide range of single-modality tasks, including natural language understanding, natural language generation, text summarization, as well as multi-modality tasks such as image classification and video understanding. MaCP consistently delivers superior accuracy, significantly reduced computational complexity, and lower memory requirements compared to existing alternatives.
This paper introduces Generalized Attention Flow (GAF), a novel feature attribution method for Transformer-based models to address the limitations of current approaches. By extending Attention Flow and replacing attention weights with the generalized Information Tensor, GAF integrates attention weights, their gradients, the maximum flow problem, and the barrier method to enhance the performance of feature attributions. The proposed method exhibits key theoretical properties and mitigates the shortcomings of prior techniques that rely solely on simple aggregation of attention weights. Our comprehensive benchmarking on sequence classification tasks demonstrates that a specific variant of GAF consistently outperforms state-of-the-art feature attribution methods in most evaluation settings, providing a more reliable interpretation of Transformer model outputs.
Vertical text input is commonly encountered in various real-world applications, such as mathematical computations and word-based Sudoku puzzles. While current large language models (LLMs) have excelled in natural language tasks, they remain vulnerable to variations in text formatting.Recent research demonstrates that modifying input formats, such as vertically aligning words for encoder-based models, can substantially lower accuracy in text classification tasks. While easily understood by humans, these inputs can significantly mislead models, posing a potential risk of bypassing detection in real-world scenarios involving harmful or sensitive information. With the expanding application of LLMs, a crucial question arises: Do decoder-based LLMs exhibit similar vulnerabilities to vertically formatted text input? In this paper, we investigate the impact of vertical text input on the performance of various LLMs across multiple text classification datasets and analyze the underlying causes. Our findings are as follows: (i) Vertical text input significantly degrades the accuracy of LLMs in text classification tasks. (ii) Chain of Thought (CoT) reasoning does not help LLMs recognize vertical input or mitigate its vulnerability, but few-shot learning with careful analysis does. (iii) We explore the underlying cause of the vulnerability by analyzing the inherent issues in tokenization and attention matrices.
Reinforcement learning from human feedback (RLHF) has emerged as the primary method for aligning large language models (LLMs) with human preferences. While it enables LLMs to achieve human-level alignment, it often incurs significant computational and financial costs due to its reliance on training external reward models or human-labeled preferences. In this work, we propose Implicit Preference Optimization (IPO), an alternative approach that leverages generative LLMs as preference classifiers, thereby reducing the dependence on external human feedback or reward models to obtain preferences. We conduct a comprehensive evaluation on the preference classification ability of LLMs using RewardBench, assessing models across different sizes, architectures, and training levels to validate our hypothesis. Furthermore, we investigate the self-improvement capabilities of LLMs by generating multiple responses for a given instruction and employing the model itself as a preference classifier for Direct Preference Optimization (DPO)-based training. Our findings demonstrate that models trained through IPO achieve performance comparable to those utilizing state-of-the-art reward models for obtaining preferences.
Discourse understanding is essential for many NLP tasks, yet most existing work remains constrained by framework-dependent discourse representations. This work investigates whether large language models (LLMs) capture discourse knowledge that generalizes across languages and frameworks. We address this question along two dimensions: (1) developing a unified discourse relation label set to facilitate cross-lingual and cross-framework discourse analysis, and (2) probing LLMs to assess whether they encode generalizable discourse abstractions. Using multilingual discourse relation classification as a testbed, we examine a comprehensive set of 23 LLMs of varying sizes and multilingual capabilities. Our results show that LLMs, especially those with multilingual training corpora, can generalize discourse information across languages and frameworks. Further layer-wise analyses reveal that language generalization at the discourse level is most salient in the intermediate layers. Lastly, our error analysis provides an account of challenging relation classes.
A well-known issue with Retrieval Augmented Generation (RAG) is that retrieved passages that are irrelevant to the query sometimes distract the answer-generating LLM, causing it to provide an incorrect response. In this paper, we shed light on this core issue and formulate the distracting effect of a passage w.r.t. a query (and an LLM). We provide a quantifiable measure of the distracting effect of a passage and demonstrate its robustness across LLMs. Our research introduces novel methods for identifying and using hard distracting passages to improve RAG systems. By fine-tuning LLMs with these carefully selected distracting passages, we achieve up to a 7.5% increase in answering accuracy compared to counterparts fine-tuned on conventional RAG datasets. Our contribution is two-fold: first, we move beyond the simple binary classification of irrelevant passages as either completely unrelated vs. distracting, and second, we develop and analyze multiple methods for finding hard distracting passages. To our knowledge, no other research has provided such a comprehensive framework for identifying and utilizing hard distracting passages.
Generative large language models are crucial in natural language processing, but they are vulnerable to backdoor attacks, where subtle triggers compromise their behavior. Although backdoor attacks against LLMs are constantly emerging, existing benchmarks remain limited in terms of sufficient coverage of attack, metric system integrity, backdoor attack alignment. And existing pre-trained backdoor attacks are idealized in practice due to resource access constraints. Therefore we establish \textit{ELBA-Bench}, a comprehensive and unified framework that allows attackers to inject backdoor through parameter efficient fine-tuning (\textit{e.g.,} LoRA) or without fine-tuning techniques (\textit{e.g.,} In-context-learning). \textit{ELBA-Bench} provides over 1300 experiments encompassing the implementations of 12 attack methods, 18 datasets, and 12 LLMs. Extensive experiments provide new invaluable findings into the strengths and limitations of various attack strategies. For instance, PEFT attack consistently outperform without fine-tuning approaches in classification tasks while showing strong cross-dataset generalization with optimized triggers boosting robustness; Task-relevant backdoor optimization techniques or attack prompts along with clean and adversarial demonstrations can enhance backdoor attack success while preserving model performance on clean samples. Additionally, we introduce a universal toolbox designed for standardized backdoor attack research at https://github.com/NWPUliuxx/ELBA_Bench, with the goal of propelling further progress in this vital area.
Can You Really Trust Code Copilot? Evaluating Large Language Models from a Code Security Perspective
PDF ↗Code security and usability are both essential for various coding assistant applications driven by large language models (LLMs). Current code security benchmarks focus solely on single evaluation task and paradigm, such as code completion and generation, lacking comprehensive assessment across dimensions like secure code generation, vulnerability repair and discrimination. In this paper, we first propose CoV-Eval, a multi-task benchmark covering various tasks such as code completion, vulnerability repair, vulnerability detection and classification, for comprehensive evaluation of LLM code security. Besides, we developed VC-Judge, an improved judgment model that aligns closely with human experts and can review LLM-generated programs for vulnerabilities in a more efficient and reliable way. We conduct a comprehensive evaluation of 20 proprietary and open-source LLMs. Overall, while most LLMs identify vulnerable codes well, they still tend to generate insecure codes and struggle with recognizing specific vulnerability types and performing repairs. Extensive experiments and qualitative analyses reveal key challenges and optimization directions, offering insights for future research in LLM code security.
The detection of sexism in online content remains an open problem, as harmful language disproportionately affects women and marginalized groups. While automated systems for sexism detection have been developed, they still face two key challenges: data sparsity and the nuanced nature of sexist language. Even in large, well-curated datasets like the Explainable Detection of Online Sexism (EDOS), severe class imbalance hinders model generalization. Additionally, the overlapping and ambiguous boundaries of fine-grained categories introduce substantial annotator disagreement, reflecting the difficulty of interpreting nuanced expressions of sexism. To address these challenges, we propose two prompt-based data augmentation techniques: Definition-based Data Augmentation (DDA), which leverages category-specific definitions to generate semantically-aligned synthetic examples, and Contextual Semantic Expansion (CSE), which targets systematic model errors by enriching examples with task-specific semantic features. To further improve reliability in fine-grained classification, we introduce an ensemble strategy that resolves prediction ties by aggregating complementary perspectives from multiple language models. Our experimental evaluation on the EDOS dataset demonstrates state-of-the-art performance across all tasks, with notable improvements of macro F1 by 1.5 points for binary classification (Task A) and 4.1 points for fine-grained classification (Task C).
Enhancing Cross-Lingual Transfer through Reversible Transliteration: A Huffman-Based Approach for Low-Resource Languages
PDF ↗As large language models (LLMs) are trained on increasingly diverse and extensive multilingual corpora, they demonstrate cross-lingual transfer capabilities. However, these capabilities often fail to effectively extend to low-resource languages, particularly those utilizing non-Latin scripts. While transliterating low-resource languages into Latin script presents a natural solution, there currently lacks a comprehensive framework for integrating transliteration into LLMs training and deployment. Taking a pragmatic approach, this paper innovatively combines character transliteration with Huffman coding to design a complete transliteration framework. Our proposed framework offers the following advantages: 1) Compression: Reduces storage requirements for low-resource language content, achieving up to 50% reduction in file size and 50-80% reduction in token count. 2) Accuracy: Guarantees 100% lossless conversion from transliterated text back to the source language. 3) Efficiency: Eliminates the need for vocabulary expansion for low-resource languages, improving training and inference efficiency. 4) Scalability: The framework can be extended to other low-resource languages. We validate the effectiveness of our framework across multiple downstream tasks, including text classification, machine reading comprehension, and machine translation. Experimental results demonstrate that our method significantly enhances the model’s capability to process low-resource languages while maintaining performance on high-resource languages. Our data and code are publicly available at https://github.com/CMLI-NLP/HuffmanTranslit.
CrisisTS: Coupling Social Media Textual Data and Meteorological Time Series for Urgency Classification
PDF ↗This paper proposes CrisisTS, the first multimodal and multilingual dataset for urgency classification composed of benchmark crisis datasets from French and English social media about various expected (e.g., flood, storm) and sudden (e.g., earthquakes, explosions) crises that have been mapped with open source geocoded meteorological time series data. This mapping is based on a simple and effective strategy that allows for temporal and location alignment even in the absence of location mention in the text. A set of multimodal experiments have been conducted relying on transformers and LLMs to improve overall performances while ensuring model generalizability. Our results show that modality fusion outperforms text-only models.
Large Language Models (LLMs) are increasingly used to assess NLP tasks due to their ability to generate human-like judgments. Single LLMs were used initially, however, recent work suggests using multiple LLMs as judges yields improved performance. An important step in exploiting multiple judgements is the combination stage, aggregation. Existing methods in NLP either assign equal weight to all LLM judgments or are designed for specific tasks such as hallucination detection. This work focuses on aggregating predictions from multiple systems where no reference labels are available. A new method called SkillAggregation is proposed, which learns to combine estimates from LLM judges without needing additional data or ground truth. It extends the Crowdlayer aggregation method, developed for image classification, to exploit the judge estimates during inference. The approach is compared to a range of standard aggregation methods on HaluEval-Dialogue, TruthfulQA and Chatbot Arena tasks. SkillAggregation outperforms Crowdlayer on all tasks, and yields the best performance over all approaches on the majority of tasks.
Fine-grained classification via LLMs is susceptible to more complex label biases compared to traditional classification tasks. Existing bias mitigation strategies, such as retraining, post-hoc adjustment, and parameter-efficient fine-tuning (PEFT) are primarily effective for simple classification biases, such as stereotypes, but fail to adequately address prediction propensity and discriminative ability biases. In this paper, we analyze these two bias phenomena and observe their progressive accumulation from intermediate to deeper layers within LLMs. To mitigate this issue, we propose a bias-aware optimization framework that incorporates two distinct label balance constraints with a PEFT strategy targeting an intermediate layer. Our approach adjusts less than 1% of the model’s parameters while effectively curbing bias amplification in deeper layers. Extensive experiments conducted across 12 datasets and 5 LLMs demonstrate that our method consistently outperforms or matches the performance of full-parameter fine-tuning and LoRA, achieving superior results with lower perplexity.
Current multi-task adversarial text attacks rely on abundant access to shared internal features and numerous queries, often limited to a single task type. As a result, these attacks are less effective against practical scenarios involving black-box feedback APIs, limited queries, or multiple task types. To bridge this gap, we propose Cluster and Ensemble Mutil-task Text Adversarial Attack (CEMA), an effective black-box attack that exploits the transferability of adversarial texts across different tasks. CEMA simplifies complex multi-task scenarios by using a deep-level substitute model trained in a plug-and-play manner for text classification, enabling attacks without mimicking the victim model. This approach requires only a few queries for training, converting multi-task attacks into classification attacks and allowing attacks across various tasks. CEMA generates multiple adversarial candidates using different text classification methods and selects the one that most effectively attacks substitute models. In experiments involving multi-task models with two, three, or six tasks—spanning classification, translation, summarization, and text-to-image generation—CEMA demonstrates significant attack success with as few as 100 queries. Furthermore, CEMA can target commercial APIs (e.g., Baidu and Google Translate), large language models (e.g., ChatGPT 4o), and image-generation models (e.g., Stable Diffusion V2), showcasing its versatility and effectiveness in real-world applications.
Self-supervised Quantized Representation for Seamlessly Integrating Knowledge Graphs with Large Language Models
PDF ↗Due to the presence of the natural gap between Knowledge Graph (KG) structures and the natural language, the effective integration of holistic structural information of KGs with Large Language Models (LLMs) has emerged as a significant question. To this end, we propose a two-stage framework to learn and apply quantized codes for each entity, aiming for the seamless integration of KGs with LLMs. Firstly, a self-supervised quantized representation (SSQR) method is proposed to compress both KG structural and semantic knowledge into discrete codes (i.e., tokens) that align the format of language sentences. We further design KG instruction-following data by viewing these learned codes as features to directly input to LLMs, thereby achieving seamless integration. The experiment results demonstrate that SSQR outperforms existing unsupervised quantized methods, producing more distinguishable codes. Moreover, the fine-tuned LLaMA2 and LLaMA3.1 also have superior performance on KG link prediction and triple classification tasks, utilizing only 16 tokens per entity instead of thousands in conventional prompting methods.
Medical imaging provides essential visual insights for diagnosis, and multimodal large language models (MLLMs) are increasingly utilized for its analysis due to their strong generalization capabilities; however, the underlying factors driving this generalization remain unclear. Current research suggests that multi-task training outperforms single-task as different tasks can benefit each other, but they often overlook the internal relationships within these tasks. To analyze this phenomenon, we attempted to employ **compositional generalization** (CG), which refers to the models’ ability to understand novel combinations by recombining learned elements, as a guiding framework. Since medical images can be precisely defined by **M**odality, **A**natomical area, and **T**ask, naturally providing an environment for exploring CG, we assembled 106 medical datasets to create **Med-MAT** for comprehensive experiments. The experiments confirmed that MLLMs can use CG to understand unseen medical images and identified CG as one of the main drivers of the generalization observed in multi-task training. Additionally, further studies demonstrated that CG effectively supports datasets with limited data and confirmed that MLLMs can achieve CG across classification and detection tasks, underscoring its broader generalization potential. Med-MAT is available at https://github.com/FreedomIntelligence/Med-MAT.
In the era of rapidly evolving large language models (LLMs), state-of-the-art rumor detection systems, particularly those based on Message Propagation Trees (MPTs), which represent a conversation tree with the post as its root and the replies as its descendants, are facing increasing threats from adversarial attacks that leverage LLMs to generate and inject malicious messages. Existing methods are based on the assumption that different nodes exhibit varying degrees of influence on predictions. They define nodes with high predictive influence as important nodes and target them for attacks. If the model treats nodes’ predictive influence more uniformly, attackers will find it harder to target high predictive influence nodes. In this paper, we propose Similarizing the predictive Influence of Nodes with Contrastive Learning (SINCon), a defense mechanism that encourages the model to learn graph representations where nodes with varying importance have a more uniform influence on predictions. Extensive experiments on the Twitter and Weibo datasets demonstrate that SINCon not only preserves high classification accuracy on clean data but also significantly enhances resistance against LLM-driven message injection attacks.