Repository-level code generation requires generated code to be compatible not only with the target repository but also with its dependency environment. Existing retrieval-based methods mainly retrieve context from the local repository, leaving external API usage dependent on the model’s pretrained knowledge, which can be insufficient for unseen or version-specific APIs. Moreover, current retrieval strategies largely focus on one-hop evidence and overlook the structural relationships among code components. We propose CrossCoder, a cross-repository code generation framework that explicitly incorporates external libraries into the retrieval context through a unified knowledge graph over repository and library entities. CrossCoder identifies important nodes via planning and semantic retrieval, then selectively expands neighboring nodes to retrieve richer multi-hop contextual evidence for generation. To further evaluate dependency-version compatibility, we introduce VersionExec, an execution-based benchmark derived from BigCodeBench that evaluates generation under different dependency versions. Experimental results on RepoExec, DevEval, and VersionExec demonstrate that CrossCoder consistently improves both functional correctness (up to 6.3% on pass@1) and robustness to dependency-version changes.

Deploying high-dimensional Vision-Language Model (VLM) embeddings is computationally prohibitive. While Matryoshka Representation Learning (MRL) allows elastic truncation, standard MRL optimizes dimensional hierarchies semi-independently. This lack of cross-scale structural constraints creates a severe topological disconnect causing representational collapse at lower dimensions and introduces optimization trade-offs that compromise high-dimensional capacity. To solve this, we propose **M-VLM2Vec**, a framework that shifts MRL from a compromised hierarchy to a synergistic geometric structure. Without additional supervision, we integrate two novel constraints into the multimodal contrastive objective: (1) Laplacian spectral alignment to preserve the global manifold topology across scales, and (2) Stiefel-constrained orthogonal projection to bidirectionally align local contrastive landscapes. By allowing nested spaces to mutually regularize, M-VLM2Vec eliminates competing gradients, preventing ultra-low-dimensional collapse while yielding striking, consistent performance gains across all embedding scales on the Multimodal Embedding Benchmark (MMEB).

Neural topic models are widely used to uncover latent thematic structure in text corpora, but often produce topics with limited semantic coherence and weak alignment between topics and documents. Recent refinement approaches based on large language models (LLMs) have shown promising improvements, yet typically rely on repeated generative inference, resulting in high computational cost and practical issues such as sensitivity to prompting, stochastic variability, and the introduction of out-of-vocabulary content. We propose BERT-NTM, an encoder-based semantic refinement framework for neural topic modeling that avoids generative inference altogether. Instead of generating or rewriting topic words, BERT-NTM leverages a pretrained BERT encoder to provide discriminative semantic signals during a refinement phase: topic–word distributions are refined by scoring candidate words within a topic using masked language modeling, and document–topic representations are improved by aligning document embeddings with mixtures of topic embeddings via a Maximum Mean Discrepancy (MMD) objective. Both refinement components are fully parallelizable, stable, and operate strictly within the model vocabulary. Experiments on standard topic modeling benchmarks demonstrate that BERT-NTM consistently improves topic coherence and document–topic alignment while substantially reducing runtime and GPU memory usage compared to LLM-based refinement methods.

Vision-language models (VLMs) achieve strong multimodal performance but remain expensive to deploy. Knowledge distillation (KD) is an effective approach for compressing these models and transferring their capabilities to more efficient student models. While KD for large language models (LLMs) is already well surveyed, no existing work provides systematic coverage of KD for VLMs, where the addition of a visual modality introduces challenges that text-only distillation does not face. This paper presents the first dedicated survey of knowledge distillation for vision-language models. We propose a unified taxonomy that organizes methods along three orthogonal views: knowledge type, teacher accessibility, and structural compatibility, and use it to organize methods across major VLM distillation settings. We also survey the associated datasets and evaluation protocols, identify open challenges and outline directions for future research. We further maintain a GitHub repository at https://github.com/HoangTran223/Awesome-Knowledge-Distillation-of-VLMs to compile all surveyed papers, which we will actively update.

Weight-only post-training quantization (PTQ) for large language models (LLMs) typically ends with the projection of floating-point weights onto a uniform integer lattice, introducing structured output error due to non-zero activation means. Existing bias correction methods compensate for this shift by adding bias terms, which is incompatible with modern bias-less architectures such as LLaMA and Mistral and only addresses the error indirectly. We propose **Complementary Lattice Correction (CLC)**, a lightweight post-quantization method that corrects projection error directly on the integer lattice without modifying the model architecture. CLC performs a budget-constrained greedy descent that selectively flips quantized weights to adjacent lattice levels. We further show that CLC provably reduces the first-moment error and admits a worst-case bound on the layer-wise mean-squared error change, which certifies non-increase under a verifiable budget condition. Experiments across multiple models, quantizers, and bit widths show that CLC consistently improves perplexity and downstream accuracy with negligible overhead.

Event extraction aims to identify triggers in text, classify them into predefined event types, and recognize associated arguments such as participants and locations. Large language model (LLM)-based generative approaches provide a unified paradigm for joint trigger identification, classification, and argument extraction; however, their computational cost necessitates knowledge distillation (KD) into smaller student models. Existing KD methods operate at the token level, treating each output position independently and thus failing to concentrate the distillation signal on event-relevant spans or to capture the relational structure among event components. We propose EventKD, a two-level distillation framework tailored for event extraction. At the token level, a distributional divergence loss incorporates event-aware span importance weighting to prioritize tokens corresponding to triggers, arguments, and type labels. At the span level, a Relational-Aware (RA) loss preserves pairwise geometric relationships among event-aware spans across Transformer layers. Experiments on four benchmarks (ACE05, MAVEN, GENEVA, and RAMS) demonstrate that EventKD consistently outperforms existing distillation methods on both trigger identification and argument extraction.

Dynamic topic models aim to reveal how themes emerge, evolve, and dissolve in time-stamped corpora, but existing approaches still face three major challenges: (i) encoders capture bag-of-words statistics but fail to align with the rich semantic priors of large pre-trained language models, (ii) temporal linkages are often modeled as rigid one-to-one chains, limiting the ability to track non-linear evolution such as topic splits or merges, and (iii) interpretability remains shallow, relying on noisy top-word lists that obscure thematic clarity. We propose L-DNTM (LLM-Augmented for Dynamic Neural Topic Model), a variational framework designed to capture more faithful temporal trajectories. Our model integrates three key components: multi-objective distillation to inject PLM-derived semantic knowledge into the encoder, entropy-regularized optimal transport to align entire topic constellations across time for smooth yet flexible evolution, and LLM-guided refinement to sharpen topic–word distributions for improved interpretability. Extensive experiments on diverse corpora show that L-DNTM yields more coherent, temporally consistent, and interpretable topic dynamics, and further enhances downstream classification and clustering tasks.

Knowledge Distillation (KD) has emerged as a crucial technique for compressing Large Language Models (LLMs). Although existing cross-tokenizer KD methods have made notable progress, their effectiveness remains constrained by suboptimal alignment across sequence and vocabulary levels. To address these limitations, we introduce Dual-Space Weighting and Time-Warped Alignment (DWA-KD), a novel cross-tokenizer distillation framework that enhances token-wise distillation through dual-space entropy-based weighting and achieves precise sequence-level alignment by leveraging both lexical and semantic information. At the token level, DWA-KD maps teacher representations into the student space and vice versa, performing dual-space KD via Kullback–Leibler divergence (KL). The process is modulated by dual-space entropy-based weights that up-weight tokens where the student is uncertain and the teacher is confident, thereby focusing learning on informative tokens rather than treating all positions equally. At the sequence level, DWA-KD applies Soft Dynamic Time Warping (Soft-DTW) to both the embedding and final hidden-state layers, enabling robust alignment of lexical and contextual semantics between teacher and student sequences. Extensive experiments across diverse NLP benchmarks demonstrate that DWA-KD consistently outperforms state-of-the-art KD baselines, while ablation studies confirm the complementary contributions of entropy-based token weighting and embedding and final hidden state layer Soft-DTW alignment.

Representation learning is fundamental to NLP, but building embeddings that work well at different computational budgets is challenging. Matryoshka Representation Learning (MRL) offers a flexible inference paradigm through nested embeddings; however, learning such structures requires explicit coordination of how information is arranged across embedding dimensionality and model depth. In this work, we propose MIPIC (Matryoshka Representation Learning via Self-Distilled Intra-Relational Alignment and Progressive Information Chaining), a unified training framework designed to produce structurally coherent and semantically compact Matryoshka representations. MIPIC promotes cross-dimensional structural consistency through Self-Distilled Intra-Relational Alignment (SIA), which aligns token-level geometric and attention-driven relations between full and truncated representations using top-k CKA self-distillation. Complementarily, it enables depth-wise semantic consolidation via Progressive Information Chaining (PIC), a scaffolded alignment strategy that incrementally transfers mature task semantics from deeper layers into earlier layers. Extensive experiments on STS, NLI, and classification benchmarks (spanning models from TinyBERT to BGEM3, Qwen3) demonstrate that MIPIC yields Matryoshka representations that are highly competitive across all capacities, with significant performance advantages observed under extreme low-dimensional.

Large Language Models (LLMs) lack persistent memory for long-term personalized conversations. Existing graph-based memory systems suffer from information dilution, absent provenance tracking, and uniform retrieval that ignores query context. We introduce MemORAI (Memory Organization and Retrieval via Adaptive Graph Intelligence), a framework that integrates three innovations: selective memory filtering with dual-layer compression to retain user-persona-relevant content, a provenance-enriched multi-relational graph tracking factual origins at the turn level, and query-adaptive subgraph retrieval with Dynamic Weighted PageRank that applies query-conditioned edge weighting. Evaluated on LOCOMO and LongMemEval benchmarks, MemORAI achieves state-of-the-art performance in memory retrieval and personalized response generation, demonstrating that selective storage, enriched representation, and adaptive retrieval are essential for coherent, personalized LLM agents.

Large language models (LLMs) have shown promise for automated unit test generation, but existing evaluations largely rely on standalone settings and a narrow set of programming languages, overestimating real-world readiness. We introduce XRepoTest, a multilingual, repository-level benchmark for unit test generation spanning five underexplored languages: Rust, Go, Julia, PHP, and Ruby. XRepoTest evaluates tests under realistic repository constraints using a containerized execution framework and multiple context augmentation strategies, including file-level, LSP-based, and retrieval-based context. Beyond standard metrics such as test pass rate and coverage, we propose Invocation Rate (IR) to assess whether generated tests meaningfully exercise the intended functionality. Experiments with 14 state-of-the-art LLMs (e.g., Claude 4.5, GPT-5.2, DeepSeek V4-pro, and Qwen families) reveal a substantial gap between standalone and repository-level performance, as well as trade-offs between richer context and test reliability. Overall, XRepoTest provides a challenging and informative benchmark to advance scalable, robust, and multilingual unit test generation in realistic software environments.

The integration of large language models (LLMs) into topic modeling has shown promise for improving topic interpretability and coherence, particularly by exploiting the generative capabilities of LLMs through instruction-based prompting. However, existing methods often incur substantial computational cost and inference complexity. To address these limitations, we propose LLMFeed, a novel framework that leverages feedback from an LLM to improve topic model optimization. Specifically, the LLM is used to evaluate topic quality, identify semantic meaning, and recommend representative topic words. These signals are then incorporated to (i) refine topic representations via Direct Preference Optimization (DPO), and (ii) guide contrastive learning for improved document-topic assignments. Extensive experiments show that LLMFeed achieves a 7.6% gain in topic coherence and over 10% improvement in doc-topic distributions on 20 Newsgroups dataset, while requiring only minutes to run, compared with hours for recent LLM-integrated topic modeling methods.

Adapting large vision–language models for embedding tasks (VLM2Vec) is standard practice, but distilling them into compact architectures remains challenging due to capacity mismatches in tokenization, dimensionality, and multimodal representations. To address this, we propose a robust distillation framework based on two complementary mechanisms: Principal Geometry Alignment (PGA) to stabilize training by aligning principal geometric components, and a Mutual Information Maximization objective to enforce global semantic consistency across intra- and inter-modal pairs. By jointly constraining geometric structure and semantic dependency, our method effectively mitigates model capacity discrepancies. Experimental results on the Massive Multimodal Embedding Benchmark (MMEB) across all four meta-tasks—under both In-Distribution and Out-of-Distribution settings—demonstrate that our approach consistently outperforms state-of-the-art distillation methods. Our source code is available at https://github.com/theanh-ktmt/PGA-KD.

Matryoshka Representation Learning (MRL) produces nested embeddings whose prefixes remain semantically meaningful, allowing representations to be truncated at inference time and enabling a single model to operate across multiple dimensional budgets. This paradigm has been adopted in several modern embedding models such as Qwen3-Embedding and OpenAI text-embedding-3. However, distilling MRL teachers into compact students remains largely unexplored, as conventional knowledge distillation methods focus on aligning single-scale embeddings and do not account for the multi-scale structure of Matryoshka representations. We introduce Matryoshka Representation Distillation and propose SUGAR, a new framework that preserves the structural properties of MRL embeddings during knowledge distillation. SUGAR aligns point-wise semantic fidelity, spectral capacity via effective-rank matching, and global subspace geometry through Grassmannian alignment. Our extensive experiments show that SUGAR produces compact students that retain strong performance across multiple embedding dimensions while preserving deployment flexibility.

Multi-hop question answering often fails when retrieval treats evidence as isolated matches to the original question, since the facts needed to answer a complex question are usually connected through intermediate entities, relations, and constraints. We propose HyperProve, a retrieval-augmented QA framework that addresses this challenge by coupling question decomposition with answer-conditioned expansion over a hypergraph of atomic facts. HyperProve does not use atomic facts, hypergraphs, or iterative retrieval in isolation; instead, it carries intermediate answers and supporting hyperedges as retrieval state, then uses that state to bias the next local hypergraph expansion. This design enables HyperProve to construct coherent evidence chains for final answer generation while making the retrieval process stateful and fact-centered. Across multi-hop QA benchmarks, HyperProve achieves the best overall performance in our evaluation, outperforming the strongest baselines by an average relative improvement of 6.2% in answer accuracy and 4.9% in F1.

The attention prefilling phase of long-context LLM inference scales quadratically, making self-attention a severe computational bottleneck. Traditional sparse attention methods mitigate this through fixed patterns or offline profiling, but lack the flexibility to adapt to input-dependent attention structure. Recent dynamic methods address this by routing heads to sparse patterns in real-time, but rely on indirect routing proxies with overhead and budget allocation mechanisms that overlook the post-softmax mass hierarchy. We present CRISP (Cliff-awaRe Input-adaptive Sparse Prefilling), which identifies and addresses two structural challenges in this dynamic routing paradigm. First, we show that the routing decision can be read directly off the structure of the proxy attention map. We replace the Jensen-Shannon Divergence (JSD) routing with C_struct, a structural proxy that measures mass at Vertical-Slash compatible positions and reproduces JSD’s routing decisions while eliminating both the pooled matmul and subsequent KL divergence overhead. Second, we formalize the post-softmax mass cliff and demonstrate theoretically that strictly cumulative coverage thresholds accumulate O(n) background noise at long contexts. CRISP navigates this via a sink-aware threshold grounded in the noise floor. Empirically, across InfiniteBench, RULER and LongBench on two model families, CRISP is the strongest sparse method overall and matches or exceeds exact dense attention on retrieval-heavy benchmarks, recovering up to +28.0 pp on retrieval tasks over baselines and achieving up to a 5.30 × attention speedup at 512k tokens, driven primarily by our O(n) noise elimination during selection while preserving structural integrity.

Continual Event Detection (CED) aims to identify and classify event triggers in evolving ontologies, where novel event types emerge incrementally. Unlike standard continual learning, CED faces highly imbalanced, long-tail distributions, semantic overlaps between event types, and a prevalence of negative spans, all of which exacerbate catastrophic forgetting. To address these challenges, we propose Mozila, a novel framework that integrates Sharpness-Aware Multi-Objective Optimization (SaMOO) and Language Model Head Preservation (LMHP). SaMOO adapts gradient-based multi-objective optimization with sharpness-aware minimization to balance competing objectives, navigate the Pareto front, and locate flatter minima, enhancing robustness across sequential tasks. Meanwhile, LMHP strengthens representation learning by preserving essential linguistic knowledge from the pretrained language model head through distribution alignment, explored with three alternative strategies. To further address data imbalance, we leverage a generative model to synthesize event trigger representations, mitigating replay buffer limitations. Comprehensive experiments on standard CED benchmarks demonstrate that Mozila consistently improves performance, achieving significant improvements over baseline methods. Ablation studies and error analyses highlight the individual contributions of each component and reveal key challenges in CED, including class imbalance and distinguishing similar event types across tasks. Our framework provides a robust and generalizable solution for continual event detection in dynamic, real-world environments. All resources are publicly available at: Mozila.

Cross-lingual topic modeling aims to discover shared semantic structures across languages, yet existing models depend on sparse bilingual resources and often yield incoherent or weakly aligned topics. Recent LLM-based refinements improve interpretability but are costly, document-level, and prone to hallucination, with prior white-box approaches requiring inaccessible token probabilities. We propose LLM-XTM, a framework that integrates LLM-guided topic refinement with self-consistency uncertainty quantification, enabling black-box, stable, and scalable enhancement of cross-lingual topic models. Experiments on multilingual corpora show that LLM-XTM achieves superior topic coherence and alignment while reducing reliance on bilingual dictionaries and expensive LLM calls.

Transformer inference becomes increasingly memory-bound as the Key–Value (KV) cache grows linearly with sequence length. While subquadratic architectures offer constant-memory inference, they rely on aggressive state compression that degrades performance on complex reasoning tasks. We propose Octopus, a framework that confers fixed-memory inference onto pretrained Transformers without the information loss of linearization. Octopus retrofits attention layers with Gated Selective Attention, a learnable module that enforces an adaptive sparsity policy over the context history. By dynamically scoring and retaining only high-utility KV states, this mechanism transforms the unbounded cache into a compact, evolving memory budget that filters out uninformative noise. Empirically, on the GSM8K benchmark, it outperforms state-of-the-art linearized baselines by over 36 points under identical memory constraints. Remarkably, Octopus also surpasses its own full-cache teacher, demonstrating that learned sparse retention serves as an effective regularizer for long-horizon reasoning.

Cross-Tokenizer Knowledge Distillation (CTKD) enables knowledge transfer between a large language model and a smaller student, even when they employ different tokenizers. While existing approaches mainly focus on token-level alignment strategies, which are often brittle and sensitive to discrepancies between tokenizers, we argue that the method of aggregating tokens into more robust representations before distillation is of equal importance. In this paper, we introduce SRA (Span Representation Alignment for Large Language Model Distillation), a novel framework that reframes CTKD through the physical lens of Multi-Particle Dynamical Systems. SRA shifts the fundamental unit of alignment from tokens to robust, tokenizer-agnostic spans. We model each span as a cluster of particles and represent its state by its Center of Mass (CoM) - an attention-weighted average that captures rich semantic information. We leverage the concept of span centers of mass with attention-derived weighting to prioritize the most salient spans. In addition, we employ a geometric regularizer to preserve the structural integrity of the representation space and introduce aligned span logit distillation to enhance knowledge transfer across models. In challenging cross-architecture distillation experiments, SRA consistently and significantly outperforms state-of-the-art CTKD baselines, validating our physically-grounded approach.

Knowledge Distillation (KD) has established itself as a pivotal technique for compressing large pre-trained language models. However, existing methods that force a student to strictly mimic the teacher’s sentence embeddings or internal features often incur prohibitive computational costs and yield suboptimal performance due to the inherent capacity gap. To address these challenges, we propose TALAS (Teacher-Anchored Layer Alignment with Sharpness-aware minimization), a unified framework that synergizes hierarchical (multi-layer) alignment with robust optimization. First, we introduce a Teacher-Anchored mechanism that selectively distills final sentence embeddings only into the student’s upper layers, thereby reducing overhead while respecting capacity constraints. Second, we bridge the semantic gap in lower layers via Layer-Aligned Self-Distillation, which propagates knowledge top-down using internal geometric relational constraints in the embedding space. Finally, to prevent the student from memorizing point-wise teacher noise, we integrate Adaptive Sharpness-Aware Minimization (ASAM) into the training objective, guiding the model towards flat minima for enhanced generalization. Empirical results on standard sentence embedding benchmarks demonstrate that TALAS consistently outperforms strong distillation baselines while achieving superior training efficiency in terms of computational cost and memory footprint.

Knowledge distillation is a key technique for compressing large language models (LLMs), but most existing methods align representations at fixed layers or token-level outputs, ignoring how representations evolve across depth. As a result, the student is only weakly guided to capture the teacher’s internal relational structure during distillation, which limits knowledge transfer. To address this limitation, we propose Multi-Granular Trajectory Alignment (MTA), a framework that aligns teacher and student representations along their layer-wise transformation trajectory. MTA adopts a layer-adaptive strategy: lower layers are aligned at the word level to preserve lexical information, while higher layers operate on phrase-level spans (e.g., noun and verb phrases) to capture compositional semantics. We instantiate this idea through a Dynamic Structural Alignment loss that matches the relative geometry among semantic units within each layer. This design is motivated by empirical findings that Transformer representations become increasingly abstract with depth, and is also consistent with linguistic views in which higher-level meaning emerges through the composition of lower-level lexical units. We further incorporate a Hidden Representation Alignment loss to directly align selected teacher–student layers. Experiments show that MTA consistently outperforms state-of-the-art baselines on standard benchmarks, with ablations confirming the contribution of each component.

Document retrieval plays a crucial role in numerous question-answering systems, yet research has concentrated on the general knowledge domain and resource-rich languages like English. In contrast, it remains largely underexplored in low-resource languages and cross-lingual scenarios within specialized domain knowledge such as legal. We present a novel dataset designed for cross-lingual retrieval between Vietnamese and English, which not only covers the general domain but also extends to the legal field. Additionally, we propose auxiliary loss function and symmetrical training strategy that significantly enhance the performance of state-of-the-art models on these retrieval tasks. Our contributions offer a significant resource and methodology aimed at improving cross-lingual retrieval in both legal and general QA settings, facilitating further advancements in document retrieval research across multiple languages and a broader spectrum of specialized domains. All the resources related to our work can be accessed at huggingface.co/datasets/bkai-foundation-models/crosslingual.

Uncovering hidden topics from short texts is challenging for traditional and neural models due to data sparsity, which limits word co-occurrence patterns, and label sparsity, stemming from incomplete reconstruction targets. Although data aggregation offers a potential solution, existing neural topic models often overlook it due to time complexity, poor aggregation quality, and difficulty in inferring topic proportions for individual documents. In this paper, we propose a novel model, GloCOM (Global Clustering COntexts for Topic Models), which addresses these challenges by constructing aggregated global clustering contexts for short documents, leveraging text embeddings from pre-trained language models. GloCOM can infer both global topic distributions for clustering contexts and local distributions for individual short texts. Additionally, the model incorporates these global contexts to augment the reconstruction loss, effectively handling the label sparsity issue. Extensive experiments on short text datasets show that our approach outperforms other state-of-the-art models in both topic quality and document representations.

Recent advanced frameworks in topic models have significantly enhanced the performance compared to conventional probabilistic approaches. Such models, mostly constructed from neural network architecture together with other advanced techniques such as contextual embedding, optimal transport distance and pre-trained language model, etc. have effectively improved the topic quality and document topic distribution. Despite the improvements, these methods lack considerations of effective optimization for complex objective functions that contain log-likelihood and additional regularization terms. In this study, we propose to apply an efficient optimization method to improve the generalization and performance of topic models. Our approach explicitly considers the sharpness of the loss landscape during optimization, which forces the optimizer to choose directions in the parameter space that lead to flatter minima, in which the models are typically more stable and robust to small perturbations in the data. Additionally, we propose an effective strategy to select the flatness region for parameter optimization by leveraging the optimal transport distance between doc-topic distributions and doc-cluster proportions, which can effectively enhance document representation. Experimental results on popular benchmark datasets demonstrate that our method effectively improves the performance of baseline topic models.

Data scarcity is a major challenge in Few-shot Continual Relation Extraction (FCRE), where models must learn new relations from limited data while retaining past knowledge. Current methods, restricted by minimal data streams, struggle with catastrophic forgetting and overfitting. To overcome this, we introduce a novel data augmentation strategy that transforms single input sentences into complex texts by integrating both old and new data. Our approach sharpens model focus, enabling precise identification of word relationships based on specified relation types. By embedding adversarial training effects and leveraging new training perspectives through special objective functions, our method enhances model performance significantly. Additionally, we explore Sharpness-Aware Minimization (SAM) in Few-shot Continual Learning. Our extensive experiments uncover fascinating behaviors of SAM across tasks and offer valuable insights for future research in this dynamic field.

Few-shot Continual Relation Extraction (FCRE) has emerged as a significant challenge in information extraction, necessitating that relation extraction (RE) systems can sequentially identify new relations with limited labeled samples. While existing studies have demonstrated promising results in FCRE, they often overlook the issue of similar relations, which is a critical factor contributing to catastrophic forgetting. In this work, we propose Sirus–a novel method that utilizes relation descriptions and dynamic clustering on these descriptions to identify similar relations. Leveraging this information, we introduce innovative loss functions specifically designed to enhance the distinction between relations, with a focus on learning to differentiate similar ones. Experimental results show that our approach can effectively mitigate the problem of catastrophic forgetting and outperforms state-of-the-art methods by a large margin. Additionally, we explore the potential of Large Language Model Embeddings (LLMEs) with representation learning and embedding capabilities, demonstrating their promise for advancing FCRE systems.

Existing toxic detection models face significant limitations, such as lack of transparency, customization, and reproducibility. These challenges stem from the closed-source nature of their training data and the paucity of explanations for their evaluation mechanism. To address these issues, we propose a dataset creation mechanism that integrates voting and chain-of-thought processes, producing a high-quality open-source dataset for toxic content detection. Our methodology ensures diverse classification metrics for each sample and includes both classification scores and explanatory reasoning for the classifications.We utilize the dataset created through our proposed mechanism to train our model, which is then compared against existing widely-used detectors. Our approach not only enhances transparency and customizability but also facilitates better fine-tuning for specific use cases. This work contributes a robust framework for developing toxic content detection models, emphasizing openness and adaptability, thus paving the way for more effective and user-specific content moderation solutions.

Neural topic modeling has substantially improved topic quality and document topic distribution compared to traditional probabilistic methods. These models often incorporate multiple loss functions. However, the disparate magnitudes of these losses can make hyperparameter tuning for these loss functions challenging, potentially creating obstacles for simultaneous optimization. While gradient-based Multi-objective Optimization (MOO) algorithms offer a potential solution, they are typically applied to shared parameters in multi-task learning, hindering their broader adoption, particularly in Neural Topic Models (NTMs). Furthermore, our experiments reveal that naïve MOO applications on NTMs can yield suboptimal results, even underperforming compared to implementations without the MOO mechanism. This paper proposes a novel approach to integrate MOO algorithms, independent of hard-parameter sharing architectures, and effectively optimizes multiple NTMs loss functions. Comprehensive evaluations on widely used benchmark datasets demonstrate that our approach significantly enhances baseline topic model performance and outperforms direct MOO applications on NTMs.

Cross-lingual topic modeling aims to uncover shared semantic themes across languages. Several methods have been proposed to address this problem, leveraging both traditional and neural approaches. While previous methods have achieved some improvements in topic diversity, they often struggle to ensure high topic coherence and consistent alignment across languages. We propose XTRA (Cross-Lingual Topic Modeling with Topic and Representation Alignments), a novel framework that unifies Bag-of-Words modeling with multilingual embeddings. XTRA introduces two core components: (1) representation alignment, aligning document-topic distributions via contrastive learning in a shared semantic space; and (2) topic alignment, projecting topic-word distributions into the same space to enforce cross-lingual consistency. This dual mechanism enables XTRA to learn topics that are interpretable (coherent and diverse) and well-aligned across languages. Experiments on multilingual corpora confirm that XTRA significantly outperforms strong baselines in topic coherence, diversity, and alignment quality.

Retrieval-Augmented Generation (RAG) enhances large language models by grounding their outputs in external knowledge. Recent advances in Graph-based RAG (GRAG) frameworks, such as GraphRAG, LightRAG, and HippoRAG2, integrate knowledge graphs into the retrieval process to improve multi-hop reasoning and semantic coherence. While effective in monolingual settings, these methods remain underexplored in cross-lingual scenarios and face limitations in semantic granularity and entity alignment. In this work, we propose MaGiX, the first GRAG framework tailored for English–Vietnamese cross-lingual question answering. MaGiX constructs a multi-granular cross-lingual knowledge graph using fine-grained attribute descriptions and cross-synonym edges, and incorporates a custom multilingual embedding model trained with contrastive learning for semantic alignment. During retrieval, MaGiX leverages graph-based reasoning and a semantic-aware reranking strategy to enhance cross-lingual relevance. Experiments across five benchmarks show that MaGiX substantially outperforms prior GRAG systems in both retrieval accuracy and generation quality, advancing structured retrieval for multilingual QA.

Recent advances in neural topic models (NTMs) have improved topic quality but still face challenges: weak document-topic alignment, high inference costs due to large pretrained language models (PLMs), and limited modeling of hierarchical topic structures. To address these issues, we introduce HiCOT (Hierarchical Clustering and Contrastive Learning with Optimal Transport for Neural Topic Modeling), a novel framework that enhances topic coherence and efficiency. HiCOT integrates Optimal Transport to refine document-topic relationships using compact PLM-based embeddings, captures semantic structure of the documents. Additionally, it employs hierarchical clustering combine with contrastive learning to disentangle topic-word and topic-topic relationships, ensuring clearer structure and better coherence. Experimental results on multiple benchmark datasets demonstrate HiCOT’s superior effectiveness over existing NTMs in topic coherence, topic performance, representation quality, and computational efficiency.

Discovering topics and learning document representations in topic space are two crucial aspects of topic modeling, particularly in the short-text setting, where inferring topic proportions for individual documents is highly challenging. Despite significant progress in neural topic modeling, effectively distinguishing document representations as well as topic embeddings remains an open problem. In this paper, we propose a novel method called Enhancing Global Clustering with Optimal Transport in Topic Modeling (EnCOT). Our approach utilizes an abstract global clusters concept to capture global information and then employs the Optimal Transport framework to align document representations in the topic space with global clusters, while also aligning global clusters with topics. This dual alignment not only enhances the separation of documents in the topic space but also facilitates learning of latent topics. Through extensive experiments, we demonstrate that our method outperforms state-of-the-art techniques in short-text topic modeling across commonly used metrics.

Knowledge distillation (KD) is crucial for compressing large text embedding models, but faces challenges when teacher and student models use different tokenizers (Cross-Tokenizer KD - CTKD). Vocabulary mismatches impede the transfer of relational knowledge encoded in deep representations, such as hidden states and attention matrices, which are vital for producing high-quality embeddings. Existing CTKD methods often focus on direct output alignment, neglecting this crucial structural information. We propose a novel framework tailored for CTKD embedding model distillation. We first map tokens one-to-one via Minimum Edit Distance (MinED). Then, we distill intra-model relational knowledge by aligning attention matrix patterns using Centered Kernel Alignment, focusing on the top-m most important tokens of the directly mapped tokens. Simultaneously, we align final hidden states via Optimal Transport with Importance-Scored Mass Assignment, which emphasizes semantically important token representations, based on importance scores derived from attention weights. We evaluate distillation from state-of-the-art embedding models (e.g., LLM2Vec, BGE) to a Bert-base-uncased model on embedding-reliant tasks such as text classification, sentence pair classification, and semantic textual similarity. Our proposed framework significantly outperforms existing CTKD baselines. By preserving attention structure and prioritizing key representations, our approach yields smaller, high-fidelity embedding models despite tokenizer differences.

To address the phenomenon of similar classes, existing methods in few-shot continual relation extraction (FCRE) face two main challenges: non-representative prototypes and representation bias, especially when the number of available samples is limited. In our work, we propose Minion to address these challenges. Firstly, we leverage the General Orthogonal Frame (GOF) structure, based on the concept of Neural Collapse, to create robust class prototypes with clear separation, even between analogous classes. Secondly, we utilize label description representations as global class representatives within the fast-slow contrastive learning paradigm. These representations consistently encapsulate the essential attributes of each relation, acting as global information that helps mitigate overfitting and reduces representation bias caused by the limited local few-shot examples within a class. Extensive experiments on well-known FCRE benchmarks show that our method outperforms state-of-the-art approaches, demonstrating its effectiveness for advancing RE system.

Continual event detection is a cornerstone in uncovering valuable patterns in many dynamic practical applications, where novel events emerge daily. Existing state-of-the-art approaches with replay buffers still suffer from catastrophic forgetting, partially due to overly simplistic objective aggregation. This oversight disregards complex trade-offs and leads to sub-optimal gradient updates, resulting in performance deterioration across objectives. While there are successful, widely cited multi-objective optimization frameworks for multi-task learning, they lack mechanisms to address data imbalance and evaluate whether a Pareto-optimal solution can effectively mitigate catastrophic forgetting, rendering them unsuitable for direct application to continual learning. To address these challenges, we propose SharpSeq, a novel continual learning paradigm leveraging sharpness-aware minimization combined with a generative model to balance training data distribution. Through extensive experiments on multiple real-world datasets, we demonstrate the superior performance of SharpSeq in continual event detection, proving the importance of our approach in mitigating catastrophic forgetting in continual event detection.

We study the problem of Event Causality Identification (ECI) that seeks to predict causal relation between event mentions in the text. In contrast to previous classification-based models, a few recent ECI methods have explored generative models to deliver state-of-the-art performance. However, such generative models cannot handle document-level ECI where long context between event mentions must be encoded to secure correct predictions. In addition, previous generative ECI methods tend to rely on external toolkits or human annotation to obtain necessary training signals. To address these limitations, we propose a novel generative framework that leverages Optimal Transport (OT) to automatically select the most important sentences and words from full documents. Specifically, we introduce hierarchical OT alignments between event pairs and the document to extract pertinent contexts. The selected sentences and words are provided as input and output to a T5 encoder-decoder model which is trained to generate both the causal relation label and salient contexts. This allows richer supervision without external tools. We conduct extensive evaluations on different datasets with multiple languages to demonstrate the benefits and state-of-the-art performance of ECI.

Large language models (LLMs) have become integral to our professional workflows and daily lives. Nevertheless, these machine companions of ours have a critical flaw: the huge amount of data which endows them with vast and diverse knowledge, also exposes them to the inevitable toxicity and bias. While most LLMs incorporate defense mechanisms to prevent the generation of harmful content, these safeguards can be easily bypassed with minimal prompt engineering. In this paper, we introduce the new Thoroughly Engineered Toxicity (TET) dataset, comprising manually crafted prompts designed to nullify the protective layers of such models. Through extensive evaluations, we demonstrate the pivotal role of TET in providing a rigorous benchmark for evaluation of toxicity awareness in several popular LLMs: it highlights the toxicity in the LLMs that might remain hidden when using normal prompts, thus revealing subtler issues in their behavior.

Continual Relation Extraction (CRE) aims to continuously train a model to learn new relations while preserving its ability on previously learned relations. Similar to other continual learning problems, in CRE, models experience representation shift, where learned deep space changes in the continual learning process, which leads to the downgrade in the performance of the old tasks. In this work, we will provide an insight into this phenomenon under the spectral viewpoint. Our key argument is that, for each class shape, if its eigenvectors (or spectral components) do not change much, the shape is well-preserved. We then conduct a spectral experiment and show that, for the shape of each class, the eigenvectors with larger eigenvalue are more preserved after learning new tasks which means these vectors are good at keeping class shapes. Based on this analysis, we propose a simple yet effective class-wise regularization that improve the eigenvalues in the representation learning. We observe that our proposed regularization leads to an increase in the eigenvalues. Extensive experiments on two benchmark datasets, FewRel and TACRED, show the effectiveness of our proposed method with significant improvement in performance compared to the state-of-the-art models. Further analyses also verify our hypothesis that larger eigenvalues lead to better performance and vice versa.

We study the problem of cross-lingual transfer learning for event detection (ED) where models trained on a source language are expected to perform well on data for a new target language. Among a few recent works for this problem, the main approaches involve representation matching (e.g., adversarial training) that aims to eliminate language-specific features from the representations to achieve the language-invariant representations. However, due to the mix of language-specific features with event-discriminative context, representation matching methods might also remove important features for event prediction, thus hindering the performance for ED. To address this issue, we introduce a novel approach for cross-lingual ED where representations are augmented with additional context (i.e., not eliminating) to bridge the gap between languages while enriching the contextual information to facilitate ED. At the core of our method involves a retrieval model that retrieves relevant sentences in the target language for an input sentence to compute augmentation representations. Experiments on three languages demonstrate the state-of-the-art performance of our model for cross-lingual ED.

Subevent Relation Extraction (SRE) is a task in Information Extraction that aims to recognize spatial and temporal containment relations between event mentions in text. Recent methods have utilized pre-trained language models to represent input texts for SRE. However, a key issue in existing SRE methods is the employment of sequential order of words in texts to feed into representation learning methods, thus unable to explicitly focus on important context words and their interactions to enhance representations. In this work, we introduce a new method for SRE that learns to induce effective graph structures for input texts to boost representation learning. Our method features a word alignment framework with dependency paths and optimal transport to identify important context words to form effective graph structures for SRE. In addition, to enable SRE research on non-English languages, we present a new multilingual SRE dataset for five typologically different languages. Extensive experiments reveal the state-of-the-art performance for our method on different datasets and languages.

A shift in data distribution can have a significant impact on performance of a text classification model. Recent methods addressing unsupervised domain adaptation for textual tasks typically extracted domain-invariant representations through balancing between multiple objectives to align feature spaces between source and target domains. While effective, these methods induce various new domain-sensitive hyperparameters, thus are impractical as large-scale language models are drastically growing bigger to achieve optimal performance. To this end, we propose to leverage meta-learning framework to train a neural network-based self-paced learning procedure in an end-to-end manner. Our method, called Meta Self-Paced Domain Adaption (MSP-DA), follows a novel but intuitive domain-shift variation of cluster assumption to derive the meta train-test dataset split based on the self-pacing difficulties of source domain’s examples. As a result, MSP-DA effectively leverages self-training and self-tuning domain-specific hyperparameters simultaneously throughout the learning process. Extensive experiments demonstrate our framework substantially improves performance on target domains, surpassing state-of-the-art approaches. Detailed analyses validate our method and provide insight into how each domain affects the learned hyperparameters.